Merge lp:~ltsp-docwriters/ltsp/ltsp-docs-trunk into lp:~ltsp-upstream/ltsp/ltsp-trunk
- ltsp-docs-trunk
- Merge into ltsp-trunk
Status: | Rejected |
---|---|
Rejected by: | Alkis Georgopoulos |
Proposed branch: | lp:~ltsp-docwriters/ltsp/ltsp-docs-trunk |
Merge into: | lp:~ltsp-upstream/ltsp/ltsp-trunk |
Diff against target: |
13293 lines (+13179/-0) (has conflicts) 22 files modified
.bzrignore (+8/-0) AUTHORS (+17/-0) COPYING (+339/-0) LTSPManual-C.omf (+18/-0) LTSPManual.xml (+3113/-0) Makefile.in (+111/-0) config.guess (+1465/-0) config.sub (+1569/-0) configure (+3233/-0) configure.in (+49/-0) debian/changelog (+141/-0) debian/compat (+1/-0) debian/control (+20/-0) debian/copyright (+66/-0) debian/ltsp-docs.doc-base (+11/-0) debian/rules (+4/-0) debian/source/format (+1/-0) debian/watch (+10/-0) install-sh (+323/-0) lts.conf.xml (+1878/-0) release.conf (+3/-0) savedinfo.xml (+799/-0) Conflict adding file .bzrignore. Moved existing file to .bzrignore.moved. Conflict adding file COPYING. Moved existing file to COPYING.moved. Conflict adding file release.conf. Moved existing file to release.conf.moved. |
To merge this branch: | bzr merge lp:~ltsp-docwriters/ltsp/ltsp-docs-trunk |
Related bugs: | |
Related blueprints: |
Reviewer | Review Type | Date Requested | Status |
---|---|---|---|
Alkis Georgopoulos | Disapprove | ||
Review via email: mp+318211@code.launchpad.net |
Commit message
Description of the change
Installing LTSP using the LTSP-PNP method
At the time of writing the version of LTSP in Debian Stretch is 5.5.9-2, Debian Jessie is 5.5.4-4, while in Debian Wheezy 5.4.2-6+deb7u1. ltsp-pnp uses the server / (its entire root file system) as a template for the default "chroot". Using this default there is less flexibilty since the clients must run the same version of distribution and platform as the server. The upside is that the model is easier to maintain. Thus a 32bit version (Stretch i386 or Jessie i386 or Wheezy i386) is suggested. While creating additional separate chroots is possible, the usual application where there is no separate chroot (sometimes referred to as ltsp-pnp) is the norm and nbd (rather than nfs) is used to provide a squashfs image.
The use of dnsmasq provides ease of configurability and maintenance and unlike isc dhcpd dnsmasq can be used to provide dhcp-proxy. Often the router is already providing dhcp so this additional option with dnsmasq is the easiest way to avoid conflict arising from there being two dhcp servers on the same subnet.
=======
Note that at the time of writing the Jessie kernel was linux-image-
dpkg-reconfigure linux-image-
This reports update-initramfs: Generating /boot/initrd.
Richard Kweskin (rkwesk-ltsp) wrote : | # |
Alkis Georgopoulos (alkisg) wrote : | # |
I'm setting the state of old merge requests that have been stale for years, to rejected; if someone wants, he can do another merge request in git format, since now ltsp uses git.
Unmerged revisions
- 163. By Vagrant Cascadian
-
Upload ltsp-docs 1.2-1 to unstable.
- 162. By Vagrant Cascadian
-
Update version to 1.2.
- 161. By Vagrant Cascadian
-
debian/copyright: Updated and switched to copyright-format 1.0.
- 160. By Vagrant Cascadian
-
debian/control: Updated Standards-Version to 3.9.6, some changes
needed. - 159. By Vagrant Cascadian
-
debian/control: Remove obsoleted DM-Upload-Allowed field.
- 158. By Vagrant Cascadian
-
Update to debhelper 9.
- 157. By Vagrant Cascadian
-
debian/control: Document Vcs-Browser and Vcs-Bzr fields.
- 156. By Vagrant Cascadian
-
Fix spelling error ressources -> resources.
- 155. By Vagrant Cascadian
-
LDM_PASSWORD_HASH: document versions of ldm/ltsp which contain this feature.
- 154. By Vagrant Cascadian
-
update LDM_PASSWORD_HASH description to mention sudo rather than software installation.
Preview Diff
1 | === added file '.bzrignore' |
2 | --- .bzrignore 1970-01-01 00:00:00 +0000 |
3 | +++ .bzrignore 2017-02-24 11:47:08 +0000 |
4 | @@ -0,0 +1,8 @@ |
5 | +Makefile |
6 | +autom4te.cache |
7 | +config.log |
8 | +config.status |
9 | +LTSPManual.html |
10 | +LTSPManual.pdf |
11 | +LTSPManual.txt |
12 | +lts.conf.5 |
13 | |
14 | === renamed file '.bzrignore' => '.bzrignore.moved' |
15 | === added file 'AUTHORS' |
16 | --- AUTHORS 1970-01-01 00:00:00 +0000 |
17 | +++ AUTHORS 2017-02-24 11:47:08 +0000 |
18 | @@ -0,0 +1,17 @@ |
19 | +Scott Balneaves |
20 | +Jordan Erickson |
21 | +Francis Giraldeau |
22 | +Richard Johnson |
23 | +David Johnston |
24 | +Chuck Liebow |
25 | +James McQuillan |
26 | +Jonathan Mueller |
27 | +Gideon Romm |
28 | +Joel Sass |
29 | +Robin Shepheard |
30 | +Susan Stewart |
31 | +Brian Tilma |
32 | +David Van Assche |
33 | +Carol Wiebe |
34 | +Vagrant Cascadian |
35 | +Alkis Georgopoulos |
36 | |
37 | === added file 'COPYING' |
38 | --- COPYING 1970-01-01 00:00:00 +0000 |
39 | +++ COPYING 2017-02-24 11:47:08 +0000 |
40 | @@ -0,0 +1,339 @@ |
41 | + GNU GENERAL PUBLIC LICENSE |
42 | + Version 2, June 1991 |
43 | + |
44 | + Copyright (C) 1989, 1991 Free Software Foundation, Inc., |
45 | + 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA |
46 | + Everyone is permitted to copy and distribute verbatim copies |
47 | + of this license document, but changing it is not allowed. |
48 | + |
49 | + Preamble |
50 | + |
51 | + The licenses for most software are designed to take away your |
52 | +freedom to share and change it. By contrast, the GNU General Public |
53 | +License is intended to guarantee your freedom to share and change free |
54 | +software--to make sure the software is free for all its users. This |
55 | +General Public License applies to most of the Free Software |
56 | +Foundation's software and to any other program whose authors commit to |
57 | +using it. (Some other Free Software Foundation software is covered by |
58 | +the GNU Lesser General Public License instead.) You can apply it to |
59 | +your programs, too. |
60 | + |
61 | + When we speak of free software, we are referring to freedom, not |
62 | +price. Our General Public Licenses are designed to make sure that you |
63 | +have the freedom to distribute copies of free software (and charge for |
64 | +this service if you wish), that you receive source code or can get it |
65 | +if you want it, that you can change the software or use pieces of it |
66 | +in new free programs; and that you know you can do these things. |
67 | + |
68 | + To protect your rights, we need to make restrictions that forbid |
69 | +anyone to deny you these rights or to ask you to surrender the rights. |
70 | +These restrictions translate to certain responsibilities for you if you |
71 | +distribute copies of the software, or if you modify it. |
72 | + |
73 | + For example, if you distribute copies of such a program, whether |
74 | +gratis or for a fee, you must give the recipients all the rights that |
75 | +you have. You must make sure that they, too, receive or can get the |
76 | +source code. And you must show them these terms so they know their |
77 | +rights. |
78 | + |
79 | + We protect your rights with two steps: (1) copyright the software, and |
80 | +(2) offer you this license which gives you legal permission to copy, |
81 | +distribute and/or modify the software. |
82 | + |
83 | + Also, for each author's protection and ours, we want to make certain |
84 | +that everyone understands that there is no warranty for this free |
85 | +software. If the software is modified by someone else and passed on, we |
86 | +want its recipients to know that what they have is not the original, so |
87 | +that any problems introduced by others will not reflect on the original |
88 | +authors' reputations. |
89 | + |
90 | + Finally, any free program is threatened constantly by software |
91 | +patents. We wish to avoid the danger that redistributors of a free |
92 | +program will individually obtain patent licenses, in effect making the |
93 | +program proprietary. To prevent this, we have made it clear that any |
94 | +patent must be licensed for everyone's free use or not licensed at all. |
95 | + |
96 | + The precise terms and conditions for copying, distribution and |
97 | +modification follow. |
98 | + |
99 | + GNU GENERAL PUBLIC LICENSE |
100 | + TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION |
101 | + |
102 | + 0. This License applies to any program or other work which contains |
103 | +a notice placed by the copyright holder saying it may be distributed |
104 | +under the terms of this General Public License. The "Program", below, |
105 | +refers to any such program or work, and a "work based on the Program" |
106 | +means either the Program or any derivative work under copyright law: |
107 | +that is to say, a work containing the Program or a portion of it, |
108 | +either verbatim or with modifications and/or translated into another |
109 | +language. (Hereinafter, translation is included without limitation in |
110 | +the term "modification".) Each licensee is addressed as "you". |
111 | + |
112 | +Activities other than copying, distribution and modification are not |
113 | +covered by this License; they are outside its scope. The act of |
114 | +running the Program is not restricted, and the output from the Program |
115 | +is covered only if its contents constitute a work based on the |
116 | +Program (independent of having been made by running the Program). |
117 | +Whether that is true depends on what the Program does. |
118 | + |
119 | + 1. You may copy and distribute verbatim copies of the Program's |
120 | +source code as you receive it, in any medium, provided that you |
121 | +conspicuously and appropriately publish on each copy an appropriate |
122 | +copyright notice and disclaimer of warranty; keep intact all the |
123 | +notices that refer to this License and to the absence of any warranty; |
124 | +and give any other recipients of the Program a copy of this License |
125 | +along with the Program. |
126 | + |
127 | +You may charge a fee for the physical act of transferring a copy, and |
128 | +you may at your option offer warranty protection in exchange for a fee. |
129 | + |
130 | + 2. You may modify your copy or copies of the Program or any portion |
131 | +of it, thus forming a work based on the Program, and copy and |
132 | +distribute such modifications or work under the terms of Section 1 |
133 | +above, provided that you also meet all of these conditions: |
134 | + |
135 | + a) You must cause the modified files to carry prominent notices |
136 | + stating that you changed the files and the date of any change. |
137 | + |
138 | + b) You must cause any work that you distribute or publish, that in |
139 | + whole or in part contains or is derived from the Program or any |
140 | + part thereof, to be licensed as a whole at no charge to all third |
141 | + parties under the terms of this License. |
142 | + |
143 | + c) If the modified program normally reads commands interactively |
144 | + when run, you must cause it, when started running for such |
145 | + interactive use in the most ordinary way, to print or display an |
146 | + announcement including an appropriate copyright notice and a |
147 | + notice that there is no warranty (or else, saying that you provide |
148 | + a warranty) and that users may redistribute the program under |
149 | + these conditions, and telling the user how to view a copy of this |
150 | + License. (Exception: if the Program itself is interactive but |
151 | + does not normally print such an announcement, your work based on |
152 | + the Program is not required to print an announcement.) |
153 | + |
154 | +These requirements apply to the modified work as a whole. If |
155 | +identifiable sections of that work are not derived from the Program, |
156 | +and can be reasonably considered independent and separate works in |
157 | +themselves, then this License, and its terms, do not apply to those |
158 | +sections when you distribute them as separate works. But when you |
159 | +distribute the same sections as part of a whole which is a work based |
160 | +on the Program, the distribution of the whole must be on the terms of |
161 | +this License, whose permissions for other licensees extend to the |
162 | +entire whole, and thus to each and every part regardless of who wrote it. |
163 | + |
164 | +Thus, it is not the intent of this section to claim rights or contest |
165 | +your rights to work written entirely by you; rather, the intent is to |
166 | +exercise the right to control the distribution of derivative or |
167 | +collective works based on the Program. |
168 | + |
169 | +In addition, mere aggregation of another work not based on the Program |
170 | +with the Program (or with a work based on the Program) on a volume of |
171 | +a storage or distribution medium does not bring the other work under |
172 | +the scope of this License. |
173 | + |
174 | + 3. You may copy and distribute the Program (or a work based on it, |
175 | +under Section 2) in object code or executable form under the terms of |
176 | +Sections 1 and 2 above provided that you also do one of the following: |
177 | + |
178 | + a) Accompany it with the complete corresponding machine-readable |
179 | + source code, which must be distributed under the terms of Sections |
180 | + 1 and 2 above on a medium customarily used for software interchange; or, |
181 | + |
182 | + b) Accompany it with a written offer, valid for at least three |
183 | + years, to give any third party, for a charge no more than your |
184 | + cost of physically performing source distribution, a complete |
185 | + machine-readable copy of the corresponding source code, to be |
186 | + distributed under the terms of Sections 1 and 2 above on a medium |
187 | + customarily used for software interchange; or, |
188 | + |
189 | + c) Accompany it with the information you received as to the offer |
190 | + to distribute corresponding source code. (This alternative is |
191 | + allowed only for noncommercial distribution and only if you |
192 | + received the program in object code or executable form with such |
193 | + an offer, in accord with Subsection b above.) |
194 | + |
195 | +The source code for a work means the preferred form of the work for |
196 | +making modifications to it. For an executable work, complete source |
197 | +code means all the source code for all modules it contains, plus any |
198 | +associated interface definition files, plus the scripts used to |
199 | +control compilation and installation of the executable. However, as a |
200 | +special exception, the source code distributed need not include |
201 | +anything that is normally distributed (in either source or binary |
202 | +form) with the major components (compiler, kernel, and so on) of the |
203 | +operating system on which the executable runs, unless that component |
204 | +itself accompanies the executable. |
205 | + |
206 | +If distribution of executable or object code is made by offering |
207 | +access to copy from a designated place, then offering equivalent |
208 | +access to copy the source code from the same place counts as |
209 | +distribution of the source code, even though third parties are not |
210 | +compelled to copy the source along with the object code. |
211 | + |
212 | + 4. You may not copy, modify, sublicense, or distribute the Program |
213 | +except as expressly provided under this License. Any attempt |
214 | +otherwise to copy, modify, sublicense or distribute the Program is |
215 | +void, and will automatically terminate your rights under this License. |
216 | +However, parties who have received copies, or rights, from you under |
217 | +this License will not have their licenses terminated so long as such |
218 | +parties remain in full compliance. |
219 | + |
220 | + 5. You are not required to accept this License, since you have not |
221 | +signed it. However, nothing else grants you permission to modify or |
222 | +distribute the Program or its derivative works. These actions are |
223 | +prohibited by law if you do not accept this License. Therefore, by |
224 | +modifying or distributing the Program (or any work based on the |
225 | +Program), you indicate your acceptance of this License to do so, and |
226 | +all its terms and conditions for copying, distributing or modifying |
227 | +the Program or works based on it. |
228 | + |
229 | + 6. Each time you redistribute the Program (or any work based on the |
230 | +Program), the recipient automatically receives a license from the |
231 | +original licensor to copy, distribute or modify the Program subject to |
232 | +these terms and conditions. You may not impose any further |
233 | +restrictions on the recipients' exercise of the rights granted herein. |
234 | +You are not responsible for enforcing compliance by third parties to |
235 | +this License. |
236 | + |
237 | + 7. If, as a consequence of a court judgment or allegation of patent |
238 | +infringement or for any other reason (not limited to patent issues), |
239 | +conditions are imposed on you (whether by court order, agreement or |
240 | +otherwise) that contradict the conditions of this License, they do not |
241 | +excuse you from the conditions of this License. If you cannot |
242 | +distribute so as to satisfy simultaneously your obligations under this |
243 | +License and any other pertinent obligations, then as a consequence you |
244 | +may not distribute the Program at all. For example, if a patent |
245 | +license would not permit royalty-free redistribution of the Program by |
246 | +all those who receive copies directly or indirectly through you, then |
247 | +the only way you could satisfy both it and this License would be to |
248 | +refrain entirely from distribution of the Program. |
249 | + |
250 | +If any portion of this section is held invalid or unenforceable under |
251 | +any particular circumstance, the balance of the section is intended to |
252 | +apply and the section as a whole is intended to apply in other |
253 | +circumstances. |
254 | + |
255 | +It is not the purpose of this section to induce you to infringe any |
256 | +patents or other property right claims or to contest validity of any |
257 | +such claims; this section has the sole purpose of protecting the |
258 | +integrity of the free software distribution system, which is |
259 | +implemented by public license practices. Many people have made |
260 | +generous contributions to the wide range of software distributed |
261 | +through that system in reliance on consistent application of that |
262 | +system; it is up to the author/donor to decide if he or she is willing |
263 | +to distribute software through any other system and a licensee cannot |
264 | +impose that choice. |
265 | + |
266 | +This section is intended to make thoroughly clear what is believed to |
267 | +be a consequence of the rest of this License. |
268 | + |
269 | + 8. If the distribution and/or use of the Program is restricted in |
270 | +certain countries either by patents or by copyrighted interfaces, the |
271 | +original copyright holder who places the Program under this License |
272 | +may add an explicit geographical distribution limitation excluding |
273 | +those countries, so that distribution is permitted only in or among |
274 | +countries not thus excluded. In such case, this License incorporates |
275 | +the limitation as if written in the body of this License. |
276 | + |
277 | + 9. The Free Software Foundation may publish revised and/or new versions |
278 | +of the General Public License from time to time. Such new versions will |
279 | +be similar in spirit to the present version, but may differ in detail to |
280 | +address new problems or concerns. |
281 | + |
282 | +Each version is given a distinguishing version number. If the Program |
283 | +specifies a version number of this License which applies to it and "any |
284 | +later version", you have the option of following the terms and conditions |
285 | +either of that version or of any later version published by the Free |
286 | +Software Foundation. If the Program does not specify a version number of |
287 | +this License, you may choose any version ever published by the Free Software |
288 | +Foundation. |
289 | + |
290 | + 10. If you wish to incorporate parts of the Program into other free |
291 | +programs whose distribution conditions are different, write to the author |
292 | +to ask for permission. For software which is copyrighted by the Free |
293 | +Software Foundation, write to the Free Software Foundation; we sometimes |
294 | +make exceptions for this. Our decision will be guided by the two goals |
295 | +of preserving the free status of all derivatives of our free software and |
296 | +of promoting the sharing and reuse of software generally. |
297 | + |
298 | + NO WARRANTY |
299 | + |
300 | + 11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY |
301 | +FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN |
302 | +OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES |
303 | +PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED |
304 | +OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF |
305 | +MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS |
306 | +TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE |
307 | +PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, |
308 | +REPAIR OR CORRECTION. |
309 | + |
310 | + 12. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING |
311 | +WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR |
312 | +REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, |
313 | +INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING |
314 | +OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED |
315 | +TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY |
316 | +YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER |
317 | +PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE |
318 | +POSSIBILITY OF SUCH DAMAGES. |
319 | + |
320 | + END OF TERMS AND CONDITIONS |
321 | + |
322 | + How to Apply These Terms to Your New Programs |
323 | + |
324 | + If you develop a new program, and you want it to be of the greatest |
325 | +possible use to the public, the best way to achieve this is to make it |
326 | +free software which everyone can redistribute and change under these terms. |
327 | + |
328 | + To do so, attach the following notices to the program. It is safest |
329 | +to attach them to the start of each source file to most effectively |
330 | +convey the exclusion of warranty; and each file should have at least |
331 | +the "copyright" line and a pointer to where the full notice is found. |
332 | + |
333 | + <one line to give the program's name and a brief idea of what it does.> |
334 | + Copyright (C) <year> <name of author> |
335 | + |
336 | + This program is free software; you can redistribute it and/or modify |
337 | + it under the terms of the GNU General Public License as published by |
338 | + the Free Software Foundation; either version 2 of the License, or |
339 | + (at your option) any later version. |
340 | + |
341 | + This program is distributed in the hope that it will be useful, |
342 | + but WITHOUT ANY WARRANTY; without even the implied warranty of |
343 | + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
344 | + GNU General Public License for more details. |
345 | + |
346 | + You should have received a copy of the GNU General Public License along |
347 | + with this program; if not, write to the Free Software Foundation, Inc., |
348 | + 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA. |
349 | + |
350 | +Also add information on how to contact you by electronic and paper mail. |
351 | + |
352 | +If the program is interactive, make it output a short notice like this |
353 | +when it starts in an interactive mode: |
354 | + |
355 | + Gnomovision version 69, Copyright (C) year name of author |
356 | + Gnomovision comes with ABSOLUTELY NO WARRANTY; for details type `show w'. |
357 | + This is free software, and you are welcome to redistribute it |
358 | + under certain conditions; type `show c' for details. |
359 | + |
360 | +The hypothetical commands `show w' and `show c' should show the appropriate |
361 | +parts of the General Public License. Of course, the commands you use may |
362 | +be called something other than `show w' and `show c'; they could even be |
363 | +mouse-clicks or menu items--whatever suits your program. |
364 | + |
365 | +You should also get your employer (if you work as a programmer) or your |
366 | +school, if any, to sign a "copyright disclaimer" for the program, if |
367 | +necessary. Here is a sample; alter the names: |
368 | + |
369 | + Yoyodyne, Inc., hereby disclaims all copyright interest in the program |
370 | + `Gnomovision' (which makes passes at compilers) written by James Hacker. |
371 | + |
372 | + <signature of Ty Coon>, 1 April 1989 |
373 | + Ty Coon, President of Vice |
374 | + |
375 | +This General Public License does not permit incorporating your program into |
376 | +proprietary programs. If your program is a subroutine library, you may |
377 | +consider it more useful to permit linking proprietary applications with the |
378 | +library. If this is what you want to do, use the GNU Lesser General |
379 | +Public License instead of this License. |
380 | |
381 | === renamed file 'COPYING' => 'COPYING.moved' |
382 | === added file 'LTSPManual-C.omf' |
383 | --- LTSPManual-C.omf 1970-01-01 00:00:00 +0000 |
384 | +++ LTSPManual-C.omf 2017-02-24 11:47:08 +0000 |
385 | @@ -0,0 +1,18 @@ |
386 | +<?xml version="1.0" encoding="UTF-8"?> |
387 | +<!DOCTYPE omf PUBLIC "-//OMF//DTD Scrollkeeper OMF Variant V1.0" |
388 | + "http://scrollkeeper.sourceforge.net/dtds/scrollkeeper-omf-1.0/scrollkeeper-omf.dtd"> |
389 | +<omf> |
390 | + <resource> |
391 | + <creator>ltsp-developer@lists.sourceforge.net (LTSP Doc Writers)</creator> |
392 | + <maintainer>ltsp-developer@lists.sourceforge.net (LTSP Doc Writers)</maintainer> |
393 | + <title>Linux Terminal Server Project Administrator's Reference</title> |
394 | + <date>2008-11-18</date> |
395 | + <subject category="System|Administration"/> |
396 | + <description>Linux Terminal Server Project Administrator's Reference.</description> |
397 | + <format mime="text/xml" dtd="-//OASIS//DTD DocBook XML V4.1.2//EN"/> |
398 | + <identifier url="file:///usr/share/gnome/help/ltsp/C/LTSPManual.xml"/> |
399 | + <language code="C"/> |
400 | + <relation seriesid="03c4cb81-ee92-8741-c9ba-b2317b03f6d220"/> |
401 | + <rights type="GPL" license.version="2.0" holder="Scott Balneaves and other authors"/> |
402 | + </resource> |
403 | +</omf> |
404 | |
405 | === added file 'LTSPManual.xml' |
406 | --- LTSPManual.xml 1970-01-01 00:00:00 +0000 |
407 | +++ LTSPManual.xml 2017-02-24 11:47:08 +0000 |
408 | @@ -0,0 +1,3113 @@ |
409 | +<?xml version="1.0" encoding="UTF-8"?> |
410 | +<!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.1.2//EN" |
411 | +"http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd" |
412 | +[ |
413 | +<!ENTITY ldm "<command>ldm</command>ldm;"> |
414 | +<!ENTITY ltspfs "<command>ltspfs</command>"> |
415 | +<!ENTITY rdesktop "<command>rdesktop</command>"> |
416 | +<!ENTITY lts.conf "<filename>lts.conf</filename>"> |
417 | +]> |
418 | +<book> |
419 | + <title>Linux Terminal Server Project Administrator's Reference</title> |
420 | + <subtitle> |
421 | + A Guide to LTSP Networks |
422 | + </subtitle> |
423 | + <bookinfo> |
424 | + <title>Linux Terminal Server Project Administrator's Reference</title> |
425 | + <authorgroup> |
426 | + <author><firstname>Scott</firstname><surname>Balneaves</surname></author> |
427 | + <author><firstname>Jordan</firstname><surname>Erickson</surname></author> |
428 | + <author><firstname>Francis</firstname><surname>Giraldeau</surname></author> |
429 | + <author><firstname>Richard</firstname><surname>Johnson</surname></author> |
430 | + <author><firstname>David</firstname><surname>Johnston</surname></author> |
431 | + <author><firstname>Chuck</firstname><surname>Liebow</surname></author> |
432 | + <author><firstname>James</firstname><surname>McQuillan</surname></author> |
433 | + <author><firstname>Jonathan</firstname><surname>Mueller</surname></author> |
434 | + <author><firstname>Gideon</firstname><surname>Romm</surname></author> |
435 | + <author><firstname>Joel</firstname><surname>Sass</surname></author> |
436 | + <author><firstname>Robin</firstname><surname>Shepheard</surname></author> |
437 | + <author><firstname>Susan</firstname><surname>Stewart</surname></author> |
438 | + <author><firstname>Brian</firstname><surname>Tilma</surname></author> |
439 | + <author><firstname>David</firstname><surname>Van Assche</surname></author> |
440 | + <author><firstname>Carol</firstname><surname>Wiebe</surname></author> |
441 | + <author><firstname>Vagrant</firstname><surname>Cascadian</surname></author> |
442 | + </authorgroup> |
443 | + <date>Aug, 2015</date> |
444 | + <edition>1.2</edition> |
445 | + <pubdate>Aug 2015</pubdate> |
446 | + <copyright> |
447 | + <year>2008</year> |
448 | + <year>2015</year> |
449 | + <holder>Scott Balneaves and other authors</holder> |
450 | + </copyright> |
451 | + <legalnotice> |
452 | + <para> |
453 | + Permission to use, copy, modify and distribute |
454 | + the software and its accompanying documentation for any |
455 | + purpose and without fee is hereby granted in perpetuity under |
456 | + the terms of the GNU GPL2, |
457 | + provided that a copy of the GNU GPL2 appears |
458 | + with it. |
459 | + </para> |
460 | + </legalnotice> |
461 | + <legalnotice> |
462 | + <para> |
463 | + The copyright holder makes no representation about the |
464 | + suitability of this software for any purpose. It is provided |
465 | + <quote>as is</quote> without expressed or implied |
466 | + warranty. If you modify the software in any way, identify your |
467 | + software as a variant of LTSP. |
468 | + </para> |
469 | + </legalnotice> |
470 | + </bookinfo> |
471 | + <chapter id='intro'> |
472 | + <title>Linux Terminal Server Project - LTSP</title> |
473 | + <sect1> |
474 | + <title>Introduction to LTSP and Thin Client Computing</title> |
475 | + <para> |
476 | + One of the key technologies included in modern GNU/Linux |
477 | + operating systems is the Linux Terminal Server Project (LTSP) |
478 | + which allows you to boot thin clients from an LTSP server. For |
479 | + educational environments, LTSP lowers hardware costs by |
480 | + enabling the use of older or less powerful machines as thin |
481 | + clients, as well as reduced administration overhead by having |
482 | + only to install and maintain the software on the server. When |
483 | + a workstation fails, it can simply be replaced without data |
484 | + loss or re-installation of the operating system. |
485 | + </para> |
486 | + <para> |
487 | + Thin client computing has been around for a long time in the |
488 | + UNIX world. Although the implementation has evolved quite a |
489 | + bit, the concept has remained the same: |
490 | + </para> |
491 | + <orderedlist> |
492 | + <listitem> |
493 | + <para> |
494 | + The thin client only takes care of the basic |
495 | + functions like display, keyboard, mouse and sound. |
496 | + </para> |
497 | + </listitem> |
498 | + <listitem> |
499 | + <para> |
500 | + The server does the heavy weightlifting. All the |
501 | + applications run on the server, and they simply |
502 | + display on the thin client. |
503 | + </para> |
504 | + </listitem> |
505 | + </orderedlist> |
506 | + <para> |
507 | + Because the thin clients have a limited number of tasks to |
508 | + manage, the hardware for the thin client can be small and |
509 | + cheap. The thin clients themselves are basically maintenance |
510 | + free. They last longer because they have no storage with |
511 | + moving parts like hard disks. If they break no data is lost |
512 | + since nothing is stored on the client itself. Simply swap the |
513 | + client with another one and go back to work. If your thin |
514 | + client is stolen or put in the trash, no data ends up in the |
515 | + hands of unauthorized people. |
516 | + </para> |
517 | + <para> |
518 | + The terminal server runs all applications and contains all |
519 | + the data. All the regular maintenance (software updates, |
520 | + administration) takes place on the terminal server. The number |
521 | + of thin clients that a terminal server can support is |
522 | + proportional to the power of the server. Because GNU/Linux |
523 | + makes efficient use of resources, you can support a surprising |
524 | + number of thin clients from a machine which might only be |
525 | + considered a powerful single user system running other |
526 | + operating systems. Please see for more details. |
527 | + </para> |
528 | + <para> |
529 | + In a thin client computing environment, the stability of the |
530 | + server is important. It's important to make sure that your |
531 | + server has good power fallback facilities, like installing a |
532 | + UPS, and depending on how much availability is required, |
533 | + redundant power supplies may be called for. As well, users who |
534 | + have the resources may decide to invest in multiple disks for |
535 | + RAID support, and other options which may be needed in a High |
536 | + Availability environment. However, you certainly don't need |
537 | + them in all environments, and GNU/Linux's high quality means |
538 | + that in all but the most demanding environments, this won't be |
539 | + needed. |
540 | + </para> |
541 | + </sect1> |
542 | + <sect1> |
543 | + <title>LTSP Security</title> |
544 | + <para> |
545 | + Security has become a key challenge for administrators and |
546 | + LTSP both recognizes and handles this quite well. Often |
547 | + schools lack the specialized IT staff or time to lock and |
548 | + clean up computers. |
549 | + </para> |
550 | + <para> |
551 | + Operating systems with LTSP included, being Linux-based |
552 | + operating systems, enjoy the security advantages of its |
553 | + Unix-like and open source heritages. This translates into |
554 | + higher quality code and no spyware and viruses, like they |
555 | + plague other operating systems. |
556 | + </para> |
557 | + <para> |
558 | + In addition, it has a strict, proactive security policy |
559 | + which means that many common problems, such as open ports or |
560 | + misconfigured software, never make it into the released |
561 | + product. Finally, LTSP based systems are true multi-user |
562 | + operating systems, making it easy to allow users to complete |
563 | + their tasks without having a level of access that could |
564 | + compromise the system. |
565 | + </para> |
566 | + </sect1> |
567 | + <sect1> |
568 | + <title>LTSP Manageability</title> |
569 | + <para> |
570 | + With administrators and especially school IT departments |
571 | + deploying and administering an increasing number of computers, |
572 | + it is difficult to find time to manage individual machines. |
573 | + LTSP thin client technology, makes deployment and management |
574 | + simple and easy. A single server is all that is needed to set |
575 | + up, manage and administrate an entire network It is recognized |
576 | + that not every school's setup is the same, so LTSP (and the |
577 | + underlying operating system) has been made to easily customize |
578 | + your unique needs. |
579 | + </para> |
580 | + </sect1> |
581 | + <sect1> |
582 | + <title>It's Green!</title> |
583 | + <para> |
584 | + With the ongoing debate about climate change, questions are |
585 | + finally being asked and answered in the fields of IT, |
586 | + education and thin client technology in general. A recent |
587 | + study compared the energy and resource consumption of a |
588 | + regular PC and Thin Client setups. You can find that study |
589 | + here: <ulink url="http://it.umsicht.fraunhofer.de/TCecology/index_en.html"> |
590 | + http://it.umsicht.fraunhofer.de/TCecology/index_en.html |
591 | + </ulink> |
592 | + </para> |
593 | + <para> |
594 | + They found that thin clients use half the energy of |
595 | + traditional workstations, which not only helps on the cost |
596 | + savings (calculate that a 40 terminal thin client lab will |
597 | + save approximately $500-$800 per year), but is ecologically |
598 | + effective in avoiding electronic waste and high carbon |
599 | + emissions. Thin client production, assembly and logistics |
600 | + costs far less and requires less energy than traditional PC |
601 | + manufacturing. The recycling of old machinery also helps the |
602 | + environment, making LTSP a green solution to the environmental |
603 | + and power saving issues many IT managers face today. |
604 | + </para> |
605 | + </sect1> |
606 | + <sect1> |
607 | + <title>Cost Effective </title> |
608 | + <para> |
609 | + With ever-increasing demands on school budgets, expensive |
610 | + technology is often last on the list. LTSP can help you offer |
611 | + what your students increasingly require from computer |
612 | + technology, without breaking the bank. GNU/Linux is and always |
613 | + will be free to acquire, use and modify, including the |
614 | + underlying LTSP structure that holds it all together. |
615 | + </para> |
616 | + <para> |
617 | + Need to set up another machine? Or another 100? Just install |
618 | + them! With GNU/Linux you'll have no more expensive OS upgrades |
619 | + and licenses, and having specialized programs on only some |
620 | + computers will become a thing of the past. When you build your |
621 | + network on Open Source software, you are freed to seek support |
622 | + for your computers from whomever you wish. |
623 | + </para> |
624 | + <para> |
625 | + GNU/Linux with LTSP can also help you save hardware costs, by |
626 | + allowing you to redeploy older machines as thin clients using |
627 | + LTSP technology. Whether you choose to set up many smaller |
628 | + labs with various LTSP servers or one giant setup with a load |
629 | + balanced LTSP setup (various servers working together to |
630 | + manage the users and applications logging on) the cost savings |
631 | + are always enormous. |
632 | + </para> |
633 | + </sect1> |
634 | + <sect1> |
635 | + <title>Well Supported</title> |
636 | + <para> |
637 | + LTSP support is available from the LTSP community. Many |
638 | + of the authors of the software included in LTSP, including the |
639 | + respective developers of the various LTSP GNU/Linux |
640 | + distribution implementations themselves, can be contacted |
641 | + directly via mailing lists and IRC channels. |
642 | + </para> |
643 | + <para> |
644 | + There are many forms of support available, including mailing |
645 | + lists, Wiki websites, IRC channels, and bug trackers. There |
646 | + are also special support groups for using LTSP and GNU/Linux. |
647 | + </para> |
648 | + <para> |
649 | + The official IRC support channel is found on freenode.org at |
650 | + #ltsp |
651 | + </para> |
652 | + <para> |
653 | + The official LTSP mailing list is found here:</para> |
654 | + <para> |
655 | + <ulink url="https://lists.sourceforge.net/lists/listinfo/ltsp-discuss"> |
656 | + https://lists.sourceforge.net/lists/listinfo/ltsp-discuss |
657 | + </ulink> |
658 | + </para> |
659 | + <para> |
660 | + In fact, some of the money that would have gone to |
661 | + purchasing software can instead be spent to hire local experts |
662 | + to help train you, and to help you support your network. LTSP |
663 | + can help you take more control over your network while also |
664 | + benefiting your local economy. With LTSP based systems, these |
665 | + choices are yours. |
666 | + </para> |
667 | + </sect1> |
668 | + <sect1> |
669 | + <title>Built for Education, Government and business</title> |
670 | + <para> |
671 | + LTSP based distros come with translations for many languages |
672 | + and localization features that allow people from all over the |
673 | + world to enjoy their computing experience. Accessibility |
674 | + features strive to provide a pleasant, high-quality computing |
675 | + experience to disabled users. |
676 | + </para> |
677 | + <para> |
678 | + The Free nature of GNU/Linux means that the applications a user |
679 | + is used to at their school or workplace is also available to |
680 | + them at home. Users can install their favorite |
681 | + GNU/Linux distribution at home, and have all the same |
682 | + functionality they are used to. In fact, many GNU/Linux |
683 | + distributions have Live CD's, which allow users to try, or even |
684 | + fully use the distribution at home, without even installing it |
685 | + on their home machine. |
686 | + </para> |
687 | + <para> |
688 | + The LTSP server software allows administrators, IT managers |
689 | + and teachers to create a low cost computer lab so that users |
690 | + can have access to the opportunities that GNU/Linux and the |
691 | + Internet can provide. |
692 | + </para> |
693 | + <para> |
694 | + Since setups can be adjusted to many situations, each thin |
695 | + client lab can be uniquely tailored to fit the business, |
696 | + agency or educational facility in question. |
697 | + </para> |
698 | + </sect1> |
699 | + </chapter> |
700 | + <chapter id='basic-concepts'> |
701 | + <title>Basic Concepts: Networks and Networking </title> |
702 | + <para> |
703 | + There are two components of a network: hardware and software. |
704 | + This section will give an introduction to both. |
705 | + </para> |
706 | + <sect1> |
707 | + <title>Hardware </title> |
708 | + <para> |
709 | + Networking works by breaking files and other data into |
710 | + little packets of information. These packets are transferred |
711 | + over a network. The difference between various types of |
712 | + networks is how they transfer packets. |
713 | + </para> |
714 | + <para> |
715 | + There are two types of networking hardware: wired and |
716 | + wireless. |
717 | + </para> |
718 | + <para> |
719 | + An important fact to remember is that a network will be only |
720 | + as fast as the slowest part. Making sure that your network |
721 | + setup matches your intended use case is an important |
722 | + consideration in an LTSP network. |
723 | + </para> |
724 | + <sect2> |
725 | + <title>Wired</title> |
726 | + <para> |
727 | + Wired networking transfers packets over a cable that |
728 | + resembles a telephone cord, but with more wires. Wired |
729 | + networks can transfer packets at one of three possible |
730 | + speeds: 10 Mbit/sec, 100 Mbit/sec, or (Gigabit) 1000 |
731 | + Mbit/sec. |
732 | + </para> |
733 | + <para> |
734 | + A network is only useful if it can connect multiple |
735 | + computers. There are some pieces of hardware that allow |
736 | + multiple computers to be connected in a network. They look |
737 | + alike, but they function differently and, likewise, |
738 | + operate at different speeds. |
739 | + </para> |
740 | + <para><emphasis>Hub</emphasis></para> |
741 | + <para> |
742 | + A hub is the simplest way to connect multiple computers. |
743 | + A hub has a lot of ports in the front and usually has |
744 | + several small lights corresponding to each port. The hub |
745 | + takes a message it receives on one port and re-sends it to |
746 | + all the ports. As a result, only one port can talk at a |
747 | + time. |
748 | + </para> |
749 | + <para><emphasis>Switch</emphasis></para> |
750 | + <para> |
751 | + A switch looks a lot like a hub; it has a lot of ports |
752 | + in the front and usually has several small lights |
753 | + corresponding to each port. However, a switch is unlike a |
754 | + hub because it only makes a connection between the ports |
755 | + it needs to. A switch can have multiple connections at the |
756 | + same time. This allows a switch to be faster than a hub. |
757 | + </para> |
758 | + <para><emphasis>Router</emphasis></para> |
759 | + <para> |
760 | + A router is used to make a connection between two |
761 | + networks. Routers are also commonly used to connect a LAN |
762 | + (local area network) to the Internet. |
763 | + </para> |
764 | + </sect2> |
765 | + <sect2> |
766 | + <title>Wireless</title> |
767 | + <para> |
768 | + Some people may wish to try using LTSP in a wireless |
769 | + environment, for various reasons. This represents some |
770 | + challenges. |
771 | + </para> |
772 | + <para> |
773 | + Wireless networks typically have more latency than wired |
774 | + networks, which generally makes interactive programs feel |
775 | + slow and unresponsive. As well, wireless adaptors cannot |
776 | + directly PXE boot, as you need to set things such as ESSID, |
777 | + keys, etc., which wouldn't be there in a PXE capable card. |
778 | + </para> |
779 | + <para> |
780 | + However, for those wishing to use LTSP wirelessly, it is |
781 | + still possible, but requires more hardware. Wireless |
782 | + bridge boxes are available, which contain both an ethernet |
783 | + and a wireless network connection. One can typically |
784 | + connect to them like a small Internet router box, and |
785 | + program them with the information pertinent to your |
786 | + network. You can then use a standard wired network card |
787 | + connected directly to the bridge, and the bridge itself |
788 | + will handle the wireless part. |
789 | + </para> |
790 | + <para> |
791 | + This method has been used with success by users of LTSP in |
792 | + the past. The latency of wireless makes the experience |
793 | + slower, however, depending on the application you wish to |
794 | + use, you may find it acceptible. |
795 | + </para> |
796 | + </sect2> |
797 | + </sect1> |
798 | + <sect1> |
799 | + <title>Software</title> |
800 | + <para> |
801 | + The most common network infrastructure services include: |
802 | + </para> |
803 | + <itemizedlist> |
804 | + <listitem> |
805 | + <para> |
806 | + DHCP (Dynamic Host Configuration Protocol)</para> |
807 | + <para> |
808 | + Each computer on a network needs a unique identifier |
809 | + called an IP address. The IP address allows packets to be |
810 | + directed to the computer, much like a street address |
811 | + allows mail to be delivered to the correct house. An IP |
812 | + address follows a specific form: four groups of digits |
813 | + forming a number from 0 to 255. For example, a local IP |
814 | + address might be 192.168.2.50. |
815 | + </para> |
816 | + <para> |
817 | + For convenience, a computer's IP address can be given |
818 | + by a server running the Dynamic Host Configuration |
819 | + Protocol (DHCP) service. DHCP automatically provides |
820 | + network settings to the computers on the network. With |
821 | + DHCP, there is no need to keep track of each computer's |
822 | + IP address. |
823 | + </para> |
824 | + </listitem> |
825 | + <listitem> |
826 | + <para> |
827 | + DNS (Domain Name System)</para> |
828 | + <para> |
829 | + DNS is a service that runs on a server, and it is like a |
830 | + phone book for computers, except that it stores IP |
831 | + addresses instead of phone numbers. Your computer talks |
832 | + to a DNS server every time you refer to another computer |
833 | + system with a name instead of an IP address. For example: |
834 | + www.ltsp.org, wikipedia.org, and google.com are all DNS |
835 | + hostnames. |
836 | + </para> |
837 | + </listitem> |
838 | + <listitem> |
839 | + <para> |
840 | + NTP (Network Time Protocol) |
841 | + </para> |
842 | + <para> |
843 | + NTP is a service that runs on a server and allows other |
844 | + computers to synchronize their clocks. The server |
845 | + synchronizes with an extremely accurate atomic clock, and |
846 | + then the clients synchronize with the server. |
847 | + </para> |
848 | + </listitem> |
849 | + <listitem> |
850 | + <para> |
851 | + Web Server |
852 | + </para> |
853 | + <para> |
854 | + A Web server answers queries using protocols such as |
855 | + HTTP, and sends content such as web pages back to clients. |
856 | + Your Web browser almost exclusively talks to Web servers. |
857 | + </para> |
858 | + </listitem> |
859 | + <listitem> |
860 | + <para> |
861 | + Web Proxy |
862 | + </para> |
863 | + <para> |
864 | + A Web proxy is a service that runs on a server and |
865 | + accesses Web sites on behalf of the clients. A proxy can |
866 | + cache some data to allow faster repeated access to |
867 | + commonly accessed pages. This is not really needed in |
868 | + essence for ltsp thin clients, since nothing runs on them, |
869 | + it all runs on the server. But in order to allow for |
870 | + content filtering, a proxy is required. In the case of a |
871 | + mixed network, where some clients are independent from the |
872 | + the thin client network, a proxy server is useful. The |
873 | + most common and recommended proxy solution is called |
874 | + Squid, which can be easily installed through your distro's |
875 | + package manager. |
876 | + </para> |
877 | + </listitem> |
878 | + <listitem> |
879 | + <para> |
880 | + Content Filter or Net Guardian |
881 | + </para> |
882 | + <para> |
883 | + A typical network requires a filtering policy to be |
884 | + implemented, which can easily be done by software like |
885 | + dansguardian, squidguard or squid-filter. This allows an |
886 | + administrator to block and control unwanted traffic like: |
887 | + </para> |
888 | + <orderedlist> |
889 | + <listitem> |
890 | + <para> |
891 | + banner ads, |
892 | + </para> |
893 | + </listitem> |
894 | + <listitem> |
895 | + <para> |
896 | + user behaviour tracking via cookies, |
897 | + </para> |
898 | + </listitem> |
899 | + <listitem> |
900 | + <para> |
901 | + animated pictures, |
902 | + </para> |
903 | + </listitem> |
904 | + <listitem> |
905 | + <para> |
906 | + JavaScript, VBScript, ActiveX (dangerous as well |
907 | + as annoying). |
908 | + </para> |
909 | + </listitem> |
910 | + </orderedlist> |
911 | + </listitem> |
912 | + <listitem> |
913 | + <para> |
914 | + Firewall & Port Blocker |
915 | + </para> |
916 | + <para> |
917 | + A firewall is usually a service on the server, but often |
918 | + DSL routers have the basic functionality of a firewall |
919 | + too. A firewall can protect your server (and clients) by |
920 | + restricting or allowing computers on the Internet from |
921 | + initiating connections into your server or network. There |
922 | + are many programs available for different distros. On |
923 | + Ubuntu and Debian we recommend using gufw (uncomplicated |
924 | + firewall), while Fedora has Fedora Firewall GUI, and SuSE |
925 | + has Yast2 Firewall. If they are not already installed, you |
926 | + can simply install them with your distro's package manager |
927 | + </para> |
928 | + </listitem> |
929 | + </itemizedlist> |
930 | + </sect1> |
931 | + <sect1> |
932 | + <title>How LTSP Works</title> |
933 | + <orderedlist> |
934 | + <listitem> |
935 | + <para> |
936 | + LTSP is a collection of software that turns a normal |
937 | + GNU/Linux installation into a terminal server. This allows |
938 | + low-powered, low-cost thin-clients (or legacy hardware |
939 | + you already own) to be used as terminals to the |
940 | + thin-client server. LTSP is unique from other |
941 | + thin-client systems in that it is considered by many |
942 | + as the easiest to maintain. Other thin-client systems |
943 | + require each client to have software that boots the |
944 | + system to a point to be able to connect to the |
945 | + terminal server. This could be a full-blown operating |
946 | + system, or a minimal OS that simply provides an |
947 | + interface to connect to the server. Systems such as |
948 | + this generally require more maintenance and |
949 | + administration, as the local software that boots the |
950 | + thin-clients may become corrupt or contain bugs that |
951 | + require attention. LTSP, on the other hand, requires |
952 | + no client-side software. It requires only a PXE |
953 | + capable network interface, which many thin-clients and |
954 | + PCs have built-in already. This means that you need |
955 | + absolutely no physical storage media (hard disk, |
956 | + compact-flash, etc.) for your thin-client to boot to |
957 | + LTSP. This significantly reduces the amount of |
958 | + administration required to keep your network running. |
959 | + The process of booting a thin-client to an LTSP server |
960 | + is as follows: |
961 | + </para> |
962 | + </listitem> |
963 | + <listitem> |
964 | + <para> |
965 | + Thin-clients boot via a protocol called PXE |
966 | + (Pre-eXecution Environment) |
967 | + </para> |
968 | + </listitem> |
969 | + <listitem> |
970 | + <para> |
971 | + PXE requests an IP address from a local DHCP |
972 | + server |
973 | + </para> |
974 | + </listitem> |
975 | + <listitem> |
976 | + <para> |
977 | + The DHCP server passes additional parameters to |
978 | + the thin-client and downloads a Linux initramfs filesystem image |
979 | + via TFTP into a RAM disk on the client itself. |
980 | + </para> |
981 | + </listitem> |
982 | + <listitem> |
983 | + <para> |
984 | + The thin-client then boots the downloaded Linux |
985 | + initramfs image, detects hardware, and connects to the LTSP |
986 | + server's X session (normally handled by <link linkend="ldm">LDM</link>). |
987 | + </para> |
988 | + </listitem> |
989 | + </orderedlist> |
990 | + <para> |
991 | + From here, all operations such as authenticating your |
992 | + username and password, launching applications, and viewing |
993 | + websites are actually handled on the LTSP server rather than |
994 | + the thin-client. The LTSP server transfers all graphical |
995 | + information to the thin-client over the network. This allows |
996 | + very low powered thin-clients to utilize the power of the |
997 | + server for all operations. It also allows for large client |
998 | + deployments with reduced overall resource utilization, as 50 |
999 | + thin-clients all running the popular OpenOffice suite under |
1000 | + different sessions generally only require enough RAM for a |
1001 | + single instance of OpenOffice (excluding per-user |
1002 | + configuration which is minimal). The server shares memory |
1003 | + between user sessions, so libraries for applications are only |
1004 | + loaded once and referenced for each user session. |
1005 | + </para> |
1006 | + </sect1> |
1007 | + </chapter> |
1008 | + <chapter id='tc-hardware'> |
1009 | + <title>LTSP Thin Client hardware requirements</title> |
1010 | + <para> |
1011 | + A lot of LTSP deployments are in classroom environments, and |
1012 | + usually, in these situations, the primary goal is to re-use |
1013 | + existing hardware that the school already owns. However, |
1014 | + specifically designed thin clients can be used also. |
1015 | + </para> |
1016 | + <sect1> |
1017 | + <title>Hardware reuse and sizing</title> |
1018 | + <para> |
1019 | + A person setting up a LTSP thin client environment for the |
1020 | + first time, typically asks two questions: |
1021 | + </para> |
1022 | + <orderedlist> |
1023 | + <listitem> |
1024 | + <para> |
1025 | + Will my existing machines work as terminals, or, |
1026 | + what should I buy to use as a terminal? |
1027 | + </para> |
1028 | + </listitem> |
1029 | + <listitem> |
1030 | + <para> |
1031 | + How big a server do I need? |
1032 | + </para> |
1033 | + </listitem> |
1034 | + </orderedlist> |
1035 | + <para> |
1036 | + Chances are, hardware that you already have is more than |
1037 | + sufficient for terminals. One of the great advantages of an |
1038 | + LTSP Server is that you can set up a high quality lab of |
1039 | + terminals for your students to use, by leveraging the machines |
1040 | + you already have. As for servers, usually, it's very easy to |
1041 | + turn any high-end single user desktop machine into a terminal |
1042 | + server capable of handling many thin clients. We'll present |
1043 | + some guidelines that should help in making the most of your |
1044 | + resources. |
1045 | + </para> |
1046 | + </sect1> |
1047 | + <sect1> |
1048 | + <title>Clients</title> |
1049 | + <sect2> |
1050 | + <title>Older hardware</title> |
1051 | + <para> |
1052 | + There are three things to consider when trying to re-use |
1053 | + existing hardware: |
1054 | + </para> |
1055 | + <orderedlist> |
1056 | + <listitem> |
1057 | + <para> |
1058 | + CPU |
1059 | + </para> |
1060 | + </listitem> |
1061 | + <listitem> |
1062 | + <para> |
1063 | + Network |
1064 | + </para> |
1065 | + </listitem> |
1066 | + <listitem> |
1067 | + <para> |
1068 | + Thin client memory |
1069 | + </para> |
1070 | + </listitem> |
1071 | + <listitem> |
1072 | + <para> |
1073 | + Video card |
1074 | + </para> |
1075 | + </listitem> |
1076 | + </orderedlist> |
1077 | + </sect2> |
1078 | + <sect2> |
1079 | + <title>CPU</title> |
1080 | + <para> |
1081 | + For using the default, secure mode of LTSP, you'll need |
1082 | + to have a slightly faster CPU. Any 533 MHz or better CPU |
1083 | + should provide acceptable performance. |
1084 | + </para> |
1085 | + <para> |
1086 | + If you have slower clients, in the range of 233 MHz to |
1087 | + 533 MHz, you may be able to use them, if you're willing to |
1088 | + reduce the security of your thin client network. More |
1089 | + information on this is available in the chapter on |
1090 | + <link linkend="ldm">LDM</link>. |
1091 | + </para> |
1092 | + </sect2> |
1093 | + <sect2> |
1094 | + <title>Network</title> |
1095 | + <para> |
1096 | + A thin client boots over the network, using a small |
1097 | + program called a network boot loader. This network boot |
1098 | + loader is sometimes located on the card itself, or, for |
1099 | + older cards without one, the user can provide one on a |
1100 | + floppy or CDRom which can be used to boot the thin client. |
1101 | + </para> |
1102 | + <para> |
1103 | + Three common network boot loaders which can be used are: |
1104 | + </para> |
1105 | + <orderedlist> |
1106 | + <listitem> |
1107 | + <para> |
1108 | + <emphasis>PXE:</emphasis> This one is the most |
1109 | + common, and many network cards and motherboards |
1110 | + with built-in network cards support this. If you |
1111 | + have one of these, you'll be able to boot without |
1112 | + any problems. |
1113 | + </para> |
1114 | + </listitem> |
1115 | + <listitem> |
1116 | + <para> |
1117 | + <emphasis>Etherboot/gPXE:</emphasis> For older cards |
1118 | + that don't have PXE included on them, you can use |
1119 | + the Free Software equivalent, Etherboot, or it's |
1120 | + newer replacement, gPXE. This |
1121 | + excellent alternative to PXE can either be booted |
1122 | + from a floppy, memory stick, or CDRom, or, if |
1123 | + you're handy with electronics, be burned onto a |
1124 | + EPROM if your card has a socket for one. More |
1125 | + information on the project can be found at |
1126 | + http://www.etherboot.org, and you can download |
1127 | + ready-to-use Etherboot images at |
1128 | + http://www.rom-o-matic.org. |
1129 | + </para> |
1130 | + </listitem> |
1131 | + <listitem> |
1132 | + <para> |
1133 | + <emphasis>Yaboot:</emphasis> For Macintosh PowerPC |
1134 | + machines (iMac's and later), you can use the built |
1135 | + in Yaboot network boot. |
1136 | + </para> |
1137 | + </listitem> |
1138 | + </orderedlist> |
1139 | + </sect2> |
1140 | + <sect2> |
1141 | + <title>Thin client memory</title> |
1142 | + <para> |
1143 | + The bare minimum for a thin client to work is about |
1144 | + 48MB, but it will be unusably slow, so it is recommended |
1145 | + to install at least 128MB Ram, with 256MB Ram if you can |
1146 | + spare it. This will really help speed up thin clients. |
1147 | + </para> |
1148 | + </sect2> |
1149 | + <sect2> |
1150 | + <title>Video Card</title> |
1151 | + <para> |
1152 | + Typically, any video card that uses the PCI bus and has |
1153 | + 16 MB or more of memory, should make a reasonable client. |
1154 | + </para> |
1155 | + </sect2> |
1156 | + </sect1> |
1157 | + </chapter> |
1158 | + <chapter id='server-hardware'> |
1159 | + <title>LTSP Server requirements</title> |
1160 | + <para> |
1161 | + An LTSP thin client network is quite scalable; a moderately |
1162 | + powerful machine can serve several thin clients, and if you need |
1163 | + to add more thin clients, you can either expand the capabilities |
1164 | + of the existing server, or, simply add more servers. |
1165 | + </para> |
1166 | + <sect1> |
1167 | + <title>Recommended specs</title> |
1168 | + <para> |
1169 | + Server sizing in an LTSP network is more art than science. |
1170 | + Ask any LTSP administrator how big a server you need to use, |
1171 | + and you'll likely be told "It depends". How big a server you |
1172 | + need does depend largely on what it is you're planning on |
1173 | + doing with your thin client network. The server requirements |
1174 | + needed for a network where the only use will be a little light |
1175 | + web-browsing, with no Java or Flash, will be greatly different |
1176 | + from a network where you want to do heavy graphics, |
1177 | + interactive games, and Flash animation. Here are some common |
1178 | + guidelines that should fit most "average" cases. |
1179 | + </para> |
1180 | + <sect2> |
1181 | + <title>Memory</title> |
1182 | + <para> |
1183 | + A GNU/Linux based operating system makes efficient use |
1184 | + of memory. The usual formula that's used for adding memory |
1185 | + to a thin client server is: |
1186 | + </para> |
1187 | + <para> |
1188 | + 256 + (192 * users) MB |
1189 | + </para> |
1190 | + <para> |
1191 | + So, if your target is to have a server with 20 |
1192 | + terminals, you'll need: |
1193 | + </para> |
1194 | + <para> |
1195 | + 256 + (192 * 20) = 256 + 3840 = 4096 MB |
1196 | + </para> |
1197 | + <para> |
1198 | + So, you'll need 4 1 Gig memory sticks. Making sure |
1199 | + you've got enough memory is the single most important |
1200 | + thing you can do to help the performance of an LTSP thin |
1201 | + client server. If you do not have enough memory in your |
1202 | + server, you'll find your server will have to use the hard |
1203 | + drive as an overflow "virtual" memory. Hard drives are |
1204 | + much slower than memory, so you'll find things getting |
1205 | + very slow if this happens. |
1206 | + </para> |
1207 | + <para> |
1208 | + If you intend to make heavy use of graphics work in your |
1209 | + curriculum, you may want to add even more, perhaps |
1210 | + doubling the previous estimate. |
1211 | + </para> |
1212 | + </sect2> |
1213 | + <sect2> |
1214 | + <title>Processors</title> |
1215 | + <para> |
1216 | + How fast a processor you need is entirely dependant on |
1217 | + what programs you plan to use. Interactive games require a |
1218 | + bit more than say, a word processor. If you plan to use |
1219 | + Java and Flash plug-ins in your web browser, these can |
1220 | + consume a lot of processing power. For a "mixed" model, |
1221 | + i.e. some people playing TuxMath, a few people browsing |
1222 | + the web, and a few people typing in OpenOffice.org, a 2GHz |
1223 | + or better processor should be able to adequately handle 20 |
1224 | + people with some minor delays. A 3GHz processor would be |
1225 | + better. |
1226 | + </para> |
1227 | + <para> |
1228 | + For larger networks, moving to an SMP (Symmetric Multi |
1229 | + Processing), or multiple CPU server may be advantageous. |
1230 | + If you plan to handle 30 or more clients, a newer |
1231 | + dual-core Xeon server or dual-core Opteron will provide |
1232 | + good results. |
1233 | + </para> |
1234 | + <para> |
1235 | + Remember, if you need to serve a large number of |
1236 | + clients, it will be worth your while to configure multiple |
1237 | + LTSP servers, each handling some of the terminals. |
1238 | + </para> |
1239 | + </sect2> |
1240 | + <sect2> |
1241 | + <title>Disks</title> |
1242 | + <para> |
1243 | + It's advisable to use some form of RAID in the terminal |
1244 | + servers. Besides saving your data when a single disks |
1245 | + fails, it improves the performance (especially read |
1246 | + performance, which is the most common type of file |
1247 | + access). For people on a budget, setting up software RAID |
1248 | + 1, with 2 SATA disks with NCQ (Native Command Queueing) |
1249 | + will provide good results. If you have a bigger budget, |
1250 | + and a bigger network, setting up your server with RAID 10 |
1251 | + along with 10,000 RPM western digital VelociRaptors will |
1252 | + give you the best speeds possible. This will provide you |
1253 | + with top notch performance and reliability. |
1254 | + </para> |
1255 | + </sect2> |
1256 | + </sect1> |
1257 | + </chapter> |
1258 | + <chapter id='network'> |
1259 | + <title>Network</title> |
1260 | + <para> |
1261 | + If you have more than 20 users, it is recommended to use Gigabit |
1262 | + Ethernet connected to a gigabit port on a switch for your LTSP |
1263 | + servers. Although normal usage ranges from 0.5 to 2mbit, clients |
1264 | + can peak quite high (70mbit), especially when watching multimedia |
1265 | + content. |
1266 | + </para> |
1267 | + <para> |
1268 | + Booting a thin client involves several steps. Understanding what |
1269 | + is happening along the way will make it much easier to solve |
1270 | + problems, should they arise. |
1271 | + </para> |
1272 | + <para> |
1273 | + There are four basic services required to boot an LTSP thin |
1274 | + client. They are: |
1275 | + </para> |
1276 | + <orderedlist> |
1277 | + <listitem> |
1278 | + <para> |
1279 | + DHCP |
1280 | + </para> |
1281 | + </listitem> |
1282 | + <listitem> |
1283 | + <para> |
1284 | + TFTP |
1285 | + </para> |
1286 | + </listitem> |
1287 | + <listitem> |
1288 | + <para> |
1289 | + NFS or NBD |
1290 | + </para> |
1291 | + </listitem> |
1292 | + <listitem> |
1293 | + <para> |
1294 | + SSH |
1295 | + </para> |
1296 | + </listitem> |
1297 | + </orderedlist> |
1298 | + </chapter> |
1299 | + <chapter id='chroot'> |
1300 | + <title>The LTSP chroot environment</title> |
1301 | + <para> |
1302 | + In order to turn a computer into a thin client, we need to run a |
1303 | + mini version of GNU/Linux on the workstation. It needs to boot |
1304 | + this mini version of GNU/Linux over the network, since it probably |
1305 | + won't have a hard drive on it's own. This mini GNU/Linux |
1306 | + installation needs to live somewhere, and the best place for it is |
1307 | + on the server. |
1308 | + </para> |
1309 | + <para> |
1310 | + This scaled-down GNU/Linux installation, customized so that |
1311 | + it's efficient to boot over the network, is called a chroot |
1312 | + environment. You can have several of them, based upon several |
1313 | + different CPU architectures. |
1314 | + </para> |
1315 | + <para> |
1316 | + They'll normally live under <filename>/opt/ltsp</filename> on the server, with |
1317 | + sub directories for each of the architectures. For instance, if you |
1318 | + have a lab full of old Power PC Macs, and older PC's, you'll |
1319 | + have an <filename>/opt/ltsp/ppc</filename> and an |
1320 | + <filename>/opt/ltsp/i386</filename> directory on the |
1321 | + server. |
1322 | + </para> |
1323 | + <para> |
1324 | + This is the LTSP project's preferred area to store the chroot, |
1325 | + however, different distros that support LTSP are free to change |
1326 | + this. Check with your distro's specific LTSP documentation to see |
1327 | + where the LTSP chroot is stored. |
1328 | + </para> |
1329 | + <para> |
1330 | + The reason why it is called a chroot environment is that to |
1331 | + install it, the GNU/Linux command chroot is called to actually set |
1332 | + the installation root to |
1333 | + <filename>/opt/ltsp/<arch></filename>. From there, a |
1334 | + scaled-down version of the distribution is installed. What this |
1335 | + means is that for you to manage the chroot, performing such things |
1336 | + as updates, all you need to do is use the chroot command to change |
1337 | + the root of your installation. Then you can use all your tools |
1338 | + like you normally would. |
1339 | + </para> |
1340 | + <sect1> |
1341 | + <title>The boot process of a thin client</title> |
1342 | + <orderedlist> |
1343 | + <listitem> |
1344 | + <para> |
1345 | + Load the Linux kernel into the memory of the thin |
1346 | + client. This can be done several different ways, |
1347 | + including: |
1348 | + </para> |
1349 | + </listitem> |
1350 | + <listitem> |
1351 | + <para> |
1352 | + Each of the above booting methods will be explained |
1353 | + later in this chapter. But for now, it should be noted |
1354 | + that it makes sense in almost all cases to use a PXE |
1355 | + based network card during booting for the fastest, |
1356 | + and easiest to setup method. |
1357 | + </para> |
1358 | + </listitem> |
1359 | + <listitem> |
1360 | + <para> |
1361 | + Once the kernel has been loaded into memory, it will |
1362 | + begin executing. |
1363 | + </para> |
1364 | + </listitem> |
1365 | + <listitem> |
1366 | + <para> |
1367 | + The kernel will initialize the entire system and all |
1368 | + of the peripherals that it recognizes. |
1369 | + </para> |
1370 | + </listitem> |
1371 | + <listitem> |
1372 | + <para> |
1373 | + This is where the fun really begins. During the |
1374 | + kernel loading process, an initramfs image will also |
1375 | + be loaded into memory. |
1376 | + </para> |
1377 | + </listitem> |
1378 | + <listitem> |
1379 | + <para> |
1380 | + Normally, when the kernel is finished booting, it will |
1381 | + launch the new task launcher <command>upstart</command>, |
1382 | + which will handle starting up a server or workstation. |
1383 | + But, in this case, we've instructed the kernel to load |
1384 | + a small shell script instead. This shell script is |
1385 | + called <command>/init</command>,and lives in the root |
1386 | + of the initramfs. |
1387 | + </para> |
1388 | + </listitem> |
1389 | + <listitem> |
1390 | + <para> |
1391 | + The <command>/init</command> script begins by mounting |
1392 | + <filename>/proc</filename> and |
1393 | + <filename>/sys</filename>, starts <command>udev</command> to |
1394 | + discover and initialize hardware, especially the |
1395 | + network card, which is needed for every aspect of the |
1396 | + boot from here on. As well, it creates a small ram |
1397 | + disk, where any local storage that is needed (to |
1398 | + configure the <filename>xorg.conf</filename> file, for |
1399 | + instance) can be written to. |
1400 | + </para> |
1401 | + </listitem> |
1402 | + <listitem> |
1403 | + <para> |
1404 | + The <emphasis>loopback</emphasis> network interface is |
1405 | + configured. This is the networking interface that has |
1406 | + <emphasis>127.0.0.1</emphasis> as its IP address. |
1407 | + </para> |
1408 | + </listitem> |
1409 | + <listitem> |
1410 | + <para> |
1411 | + A small DHCP client called <command>ipconfig</command> |
1412 | + will then be run, to make another query from the DHCP |
1413 | + server. This separate user-space query gets |
1414 | + information supplied in the dhcpd.conf file, like the |
1415 | + nfs root server, default gateway, and other important |
1416 | + parameters. |
1417 | + </para> |
1418 | + </listitem> |
1419 | + <listitem> |
1420 | + <para> |
1421 | + When <command>ipconfig</command> gets a reply from the |
1422 | + server, the information it receives is used to |
1423 | + configure the Ethernet interface, and determine the |
1424 | + server to mount the root from. |
1425 | + </para> |
1426 | + </listitem> |
1427 | + <listitem> |
1428 | + <para> |
1429 | + Up to this point, the root filesystem has been a ram |
1430 | + disk. Now, the <command>/init</command> script will |
1431 | + mount a new root filesystem via either NBD or NFS. In |
1432 | + the case of NBD, the image that is normally loaded is |
1433 | + <filename>/opt/ltsp/images/<arch>.img</filename>. |
1434 | + If the root is mounted via NFS, then the directory |
1435 | + that is exported from the server is typically |
1436 | + <filename>/opt/ltsp/<arch></filename>. |
1437 | + It can't just mount the new filesystem as |
1438 | + <filename>/</filename>. It must |
1439 | + first mount it to a separate directory. Then, it will |
1440 | + do a <command>run-init</command>, which will swap the |
1441 | + current root filesystem for a new filesystem. When it |
1442 | + completes, the filesystem will be mounted on |
1443 | + <filename>/</filename>. At |
1444 | + this point, any directories that need to be writable |
1445 | + for regular start up to occur, like |
1446 | + <filename>/tmp</filename>, or <filename>/var</filename>, are |
1447 | + mounted at this time. |
1448 | + </para> |
1449 | + </listitem> |
1450 | + <listitem> |
1451 | + <para> |
1452 | + Once the mounting of the new root filesystem is |
1453 | + complete, we are done with the <command>/init</command> shell script and |
1454 | + we need to invoke the real <command>/sbin/init</command> program. |
1455 | + </para> |
1456 | + </listitem> |
1457 | + <listitem> |
1458 | + <para> |
1459 | + The <command>init</command> program will read the |
1460 | + <filename>/etc/event.d</filename> |
1461 | + directory and begin setting up the thin client |
1462 | + environment. From there, upstart will begin reading |
1463 | + the start up commands in <filename>/etc/rcS.d</filename>. |
1464 | + </para> |
1465 | + </listitem> |
1466 | + <listitem> |
1467 | + <para> |
1468 | + It will execute the <command>ltsp-client-setup</command> command |
1469 | + which will configure many aspects of the thin client environment, |
1470 | + such as checking if local devices need starting, loading any |
1471 | + specified modules, etc. |
1472 | + </para> |
1473 | + </listitem> |
1474 | + <listitem> |
1475 | + <para> |
1476 | + Next, the <command>init</command> program will begin |
1477 | + to execute commands in the <filename>/etc/rc2.d</filename> directory |
1478 | + </para> |
1479 | + </listitem> |
1480 | + <listitem> |
1481 | + <para> |
1482 | + One of the items in the <filename>/etc/rc2.d</filename> |
1483 | + directory is the <command>ltsp-client-core</command> |
1484 | + command that will be run while the thin client is |
1485 | + booting. |
1486 | + </para> |
1487 | + </listitem> |
1488 | + <listitem> |
1489 | + <para> |
1490 | + The <s.conf; file will be parsed, |
1491 | + and all of the parameters in that file that pertain to |
1492 | + this thin client will be set as environment variables |
1493 | + for the <command>ltsp-client-core</command> script |
1494 | + to use. |
1495 | + </para> |
1496 | + </listitem> |
1497 | + <listitem> |
1498 | + <para> |
1499 | + If Sound is configured at this point, the |
1500 | + <command>pulseaudio</command> |
1501 | + daemon is started, to allow remote audio connections |
1502 | + from the server to connect and play on the thin |
1503 | + client. |
1504 | + </para> |
1505 | + </listitem> |
1506 | + <listitem> |
1507 | + <para> |
1508 | + If the thin client has local device support enabled, |
1509 | + the <command>ltspfsd</command> program is started to |
1510 | + allow the server to read from devices such as memory |
1511 | + sticks or CD-Roms attached to the thin client. |
1512 | + </para> |
1513 | + </listitem> |
1514 | + <listitem> |
1515 | + <para> |
1516 | + At this point, any of the screen sessions you've |
1517 | + defined in your <s.conf; will be |
1518 | + executed. |
1519 | + </para> |
1520 | + <para> |
1521 | + Screen sessions are what you want to launch on all |
1522 | + of the virtual screens on your terminal. These are the |
1523 | + standard virtual screens that all GNU/Linux |
1524 | + distributions usually have, i.e. |
1525 | + <keycombo><keycap>Alt</keycap><keycap>F1</keycap></keycombo>, through |
1526 | + <keycombo><keycap>Alt</keycap><keycap>F10</keycap></keycombo>. |
1527 | + </para> |
1528 | + <para> |
1529 | + By default, a standard character based getty will be |
1530 | + run on screen 1 (<varname>SCREEN_01</varname> in the <s.conf; file). |
1531 | + </para> |
1532 | + <para> |
1533 | + As well, if nothing else is specified in the |
1534 | + <s.conf; |
1535 | + file, an &ldm; screen script is run |
1536 | + on <varname>SCREEN_07</varname>. The LTSP Display |
1537 | + Manager (&ldm;) |
1538 | + is the default login manager for LTSP. |
1539 | + </para> |
1540 | + </listitem> |
1541 | + <listitem> |
1542 | + <para> |
1543 | + If <varname>SCREEN_07</varname> is set to a value of |
1544 | + &ldm;, or <command>startx</command>, then the X |
1545 | + Windows System will be launched, giving you a graphical user interface. |
1546 | + </para> |
1547 | + <para> |
1548 | + By default, the Xorg server will auto-probe the card, |
1549 | + create a default <filename>/etc/X11/xorg.conf</filename> file on the |
1550 | + ram-disk in the terminal, and start up xorg with that custom config. |
1551 | + </para> |
1552 | + </listitem> |
1553 | + <listitem> |
1554 | + <para> |
1555 | + The X server will either start an encrypted <command>ssh</command> |
1556 | + tunnel to the server, in the case of &ldm;, or an an |
1557 | + XDMCP query to the LTSP server, in the case of |
1558 | + <command>startx</command>. Either way, |
1559 | + a login box will appear on the terminal. |
1560 | + </para> |
1561 | + </listitem> |
1562 | + <listitem> |
1563 | + <para> |
1564 | + At this point, the user can log in. They'll get a |
1565 | + session on the server. |
1566 | + </para> |
1567 | + <para> |
1568 | + This confuses a lot of people at first. They are |
1569 | + sitting at a thin client, but they are running a |
1570 | + session on the server. All commands they run will be |
1571 | + run on the server, but the output will be displayed on |
1572 | + the thin client. |
1573 | + </para> |
1574 | + </listitem> |
1575 | + </orderedlist> |
1576 | + </sect1> |
1577 | + </chapter> |
1578 | + <chapter id='network-boot'> |
1579 | + <title>Network booting the thin client</title> |
1580 | + <para> |
1581 | + Getting the thin client to boot over the network can be |
1582 | + accomplished in a variety of ways: |
1583 | + </para> |
1584 | + <orderedlist> |
1585 | + <listitem> |
1586 | + <para> |
1587 | + Boot ROM |
1588 | + </para> |
1589 | + </listitem> |
1590 | + <listitem> |
1591 | + <para> |
1592 | + Local media |
1593 | + </para> |
1594 | + </listitem> |
1595 | + </orderedlist> |
1596 | + <sect1> |
1597 | + <title>Boot ROM</title> |
1598 | + <para> |
1599 | + Depending on your network card, it may already contain a |
1600 | + boot ROM, or you may be able to use an EPROM programmer to |
1601 | + create your own. Check the hardware documentation for the |
1602 | + network card in your thin client for details. |
1603 | + </para> |
1604 | + <sect2> |
1605 | + <title>Etherboot</title> |
1606 | + <para> |
1607 | + Etherboot is a very popular open-source bootrom project. |
1608 | + It contains drivers for many common network cards, and |
1609 | + works very well with LTSP. |
1610 | + </para> |
1611 | + <para> |
1612 | + ROM images suitable for booting from floppy, CD-ROM, |
1613 | + etc., can be obtained from http://www.rom-o-matic.org |
1614 | + </para> |
1615 | + <para> |
1616 | + Linux kernels must be tagged with the <command>mknbi-linux</command>, |
1617 | + which will prepare the kernel for network booting, by |
1618 | + prefixing the kernel with some additional code, and |
1619 | + appending the initrd to the end of the kernel. |
1620 | + </para> |
1621 | + <para> |
1622 | + The kernels that are supplied with LTSP are already |
1623 | + tagged, and ready to boot with Etherboot. |
1624 | + </para> |
1625 | + </sect2> |
1626 | + <sect2> |
1627 | + <title>PXE</title> |
1628 | + <para> |
1629 | + Part of the 'Wired for Management' specification from the |
1630 | + late 1990's included a specification for a bootrom |
1631 | + technology known as the |
1632 | + <emphasis>Pre-boot Execution Environment</emphasis> |
1633 | + commonly abbreviated as <emphasis>PXE</emphasis>. |
1634 | + </para> |
1635 | + <para> |
1636 | + A PXE bootrom can load at most a 32 kilo-byte file. A |
1637 | + Linux kernel is quite a bit larger than that. Therefore, |
1638 | + we setup PXE to load a 2nd stage boot loader called |
1639 | + <command>pxelinux</command>, which is small enough |
1640 | + to be loaded. It knows how to load |
1641 | + much larger files, such as a Linux kernel. |
1642 | + </para> |
1643 | + </sect2> |
1644 | + </sect1> |
1645 | + <sect1> |
1646 | + <title>Local media</title> |
1647 | + <para> |
1648 | + If your network card in the thin client doesn't have a boot |
1649 | + ROM built in, and you don't have access to an EPROM burner, |
1650 | + have no fear! Chances are, that old machine has a floppy |
1651 | + drive, or CD-ROM in it. If so, then you can use local media to |
1652 | + boot the thin client. |
1653 | + </para> |
1654 | + <sect2> |
1655 | + <title>Floppy disk</title> |
1656 | + <para> |
1657 | + Booting Etherboot from a floppy is an excellent way of |
1658 | + booting an LTSP thin client that doesn't have a boot ROM. |
1659 | + Etherboot is loaded in the boot sector of the floppy. |
1660 | + Then, it will act just like a bootrom. The boot code will |
1661 | + be executed, the network card will be initialized, and the |
1662 | + kernel will be loaded from the network server. |
1663 | + </para> |
1664 | + </sect2> |
1665 | + <sect2> |
1666 | + <title>Hard disk</title> |
1667 | + <para> |
1668 | + The hard disk can be used with LILO or GRUB, to load the |
1669 | + Linux kernel and initrd. You can also load the Etherboot |
1670 | + bootrom image from the hard disk, and it will act like a |
1671 | + bootrom. |
1672 | + </para> |
1673 | + </sect2> |
1674 | + <sect2> |
1675 | + <title>CD-ROM</title> |
1676 | + <para> |
1677 | + A bootable CD-ROM can be loaded either with a Linux |
1678 | + kernel, or an Etherboot image. |
1679 | + </para> |
1680 | + </sect2> |
1681 | + <sect2> |
1682 | + <title>USB Memory device</title> |
1683 | + <para> |
1684 | + Just like a CD-ROM, Floppy disk and Hard disk, you can |
1685 | + use a USB Memory device to boot an Etherboot module. |
1686 | + </para> |
1687 | + </sect2> |
1688 | + </sect1> |
1689 | + <sect1> |
1690 | + <title>Installation</title> |
1691 | + <para> |
1692 | + With the integration of LTSP into distributions, installation |
1693 | + of LTSP is now usually as easy as adding the LTSP packages in |
1694 | + your distro's package manager. Consult your distribution's |
1695 | + documentation for details on how to install LTSP on your |
1696 | + particular system. |
1697 | + </para> |
1698 | + <para> |
1699 | + However, as a general guideline, usually after you've installed |
1700 | + your distributions' LTSP packages, and configured your network |
1701 | + interfaces, and some kind of DHCP3 server, you'd run (as root): |
1702 | + </para> |
1703 | + <screen> |
1704 | +sudo ltsp-build-client |
1705 | + </screen> |
1706 | + <para> |
1707 | + If you are on a 64-bit system but your clients |
1708 | + have another architecture use the --arch option |
1709 | + e.g. <screen>ltsp-build-client --arch i386</screen> |
1710 | + </para> |
1711 | + <para> |
1712 | + After that, you should be able to boot your first thin |
1713 | + client. |
1714 | + </para> |
1715 | + </sect1> |
1716 | + </chapter> |
1717 | + <chapter id='customizing-ltsconf'> |
1718 | + <title>Customizing thin client behaviour</title> |
1719 | + <para> |
1720 | + By default, most thin clients will automatically configure |
1721 | + themselves correctly, and just work when they're plugged in. |
1722 | + However, sometimes you may wish to customize their behaviour. You |
1723 | + would do this by editing the <s.conf; file. |
1724 | + </para> |
1725 | + <sect1> |
1726 | + <title>Location of the lts.conf file</title> |
1727 | + <para> |
1728 | + In order to speed up LTSP, by default, we're using NBD |
1729 | + (Network Block Devices) rather than NFS. To do this, we'd had |
1730 | + to move the the lts.conf file out of the chroot and into the |
1731 | + TFTP directory, in /var/lib/tftpboot/ltsp/<arch>, where |
1732 | + <arch> is the architecture you are working on (usually i386, |
1733 | + but could be something else, like amd64 for example). This |
1734 | + means you can make changes to the file immediately, and simply |
1735 | + reboot the terminal, without recompiling the image. |
1736 | + </para> |
1737 | + </sect1> |
1738 | + <sect1> |
1739 | + <title>Sample lts.conf file</title> |
1740 | + <para> |
1741 | + Here is an example of the lts.conf file: </para> |
1742 | + <screen> |
1743 | +################ |
1744 | +# Global defaults for all clients |
1745 | +# if you refer to the local server, just use the |
1746 | +# "server" keyword as value |
1747 | +# see lts_parameters.txt for valid values |
1748 | +################ |
1749 | + |
1750 | +[default] |
1751 | + X_COLOR_DEPTH=16 |
1752 | + LOCALDEV=True |
1753 | + SOUND=True |
1754 | + NBD_SWAP=True |
1755 | + SYSLOG_HOST=server |
1756 | + XKBLAYOUT=de |
1757 | + |
1758 | +################ |
1759 | +#[MAC ADDRESS]: Per thin client settings |
1760 | +################ |
1761 | + |
1762 | +[00:11:25:84:CE:BA] |
1763 | + XSERVER = vesa |
1764 | + X_MOUSE_DEVICE=/dev/ttyS0 |
1765 | + X_MOUSE_PROTOCOL=intellimouse |
1766 | + |
1767 | +############### |
1768 | +# A Thin Client Print server |
1769 | +# (switch off X by pointing tty7 to shell, |
1770 | +# to save resources) |
1771 | +############### |
1772 | + |
1773 | +[00:11:25:93:CF:00] |
1774 | + PRINTER_0_DEVICE=/dev/usblp0 |
1775 | + SCREEN_07=shell |
1776 | + |
1777 | +############### |
1778 | +# A workstation that executes a specific |
1779 | +# command after login |
1780 | +############### |
1781 | + |
1782 | +[00:11:25:93:CF:02] |
1783 | + LDM_SESSION=/usr/bin/myloginscript |
1784 | + </screen> |
1785 | + </sect1> |
1786 | + </chapter> |
1787 | + <chapter id='lts-conf'> |
1788 | + <title>Format of the lts.conf file</title> |
1789 | + <para> |
1790 | + When LTSP was designed, one of the issues that needed to be |
1791 | + dealt with was varying hardware configurations for the thin |
1792 | + client. Certainly, whatever combination of processor, network card |
1793 | + and video card available today would not be available in 3 months, |
1794 | + when you want to add more thin clients to the network. So, LTSP |
1795 | + devised a way of specifying the configuration of each thin client. |
1796 | + The configuration file is called lts.conf and it lives in the |
1797 | + <filename>/var/lib/tftpboot/ltsp/<arch>/</filename> directory. |
1798 | + </para> |
1799 | + <para> |
1800 | + The format of the lts.conf allows for 'default' settings and |
1801 | + individual thin client settings. If all of your thin clients are |
1802 | + identical, you could specify all of the configuration settings in |
1803 | + the <varname>'[Default]'</varname> section. The file must have a first line |
1804 | + containing <varname>'[Default]'</varname> in any case. |
1805 | + </para> |
1806 | + <sect1> |
1807 | + <title>Section headings </title> |
1808 | + <para> |
1809 | + Section headings begin with an identifier in the form |
1810 | + [Default] which is used for all computers as mentioned above, |
1811 | + and [MAC address] for individual workstations, in the form of |
1812 | + [XX:XX:XX:XX:XX:XX], where X is the digits 0-9, or A-F. |
1813 | + </para> |
1814 | + <para> |
1815 | + You can usually read the MAC address for a network card from |
1816 | + a sticker on the card itself, or use some kind of network tool |
1817 | + to discover it. The best way to check for the MAC address of |
1818 | + the machine is by starting it up, checking its IP number and |
1819 | + doing a <command>arp -an</command> on the server, this should then tell you |
1820 | + which IP number has which MAC address. |
1821 | + </para> |
1822 | + </sect1> |
1823 | + <sect1> |
1824 | + <title>Variable Assignments</title> |
1825 | + <para> |
1826 | + After the section heading, you can then define variables. |
1827 | + Variables are ether boolean values, requiring a True/False or |
1828 | + Y/N answer. Note that you can either use True or False, Yes or |
1829 | + No, or Y or N. Whichever you prefer. Other variables may |
1830 | + simply be strings, supplied after the = sign. The general |
1831 | + format of an assignment looks like: |
1832 | + </para> |
1833 | + <screen> |
1834 | +VARIABLE = value |
1835 | + </screen> |
1836 | + <para> |
1837 | + Comments can be inserted into the file for your |
1838 | + documentation purposes. Comments start with a # character, and |
1839 | + everything after the # for the rest of the line is considered |
1840 | + a comment. |
1841 | + </para> |
1842 | + </sect1> |
1843 | + <sect1> |
1844 | + <title>The LIKE keyword</title> |
1845 | + <para> |
1846 | + The <varname>LIKE</varname> keyword allows you to define a |
1847 | + general set of parameters under a unique identifier, and then |
1848 | + assign individual workstations that set of parameters using the |
1849 | + <varname>LIKE</varname> keyword. An example will illustrate |
1850 | + it's use. |
1851 | + </para> |
1852 | + <para>Let's assume you have 3 kinds of thin clients on your |
1853 | + network. One set, which are used in the lab, have older video |
1854 | + cards, and must use 16 bit colour at 1024x768 resolution. |
1855 | + Another set need to have their video ram set to 8 megs, and a |
1856 | + third set which auto-detect everything correctly. We don't |
1857 | + need to specify anything for the third set, but having some |
1858 | + symbolic names for the first two would help us to maintain the |
1859 | + <s.conf; file. |
1860 | + </para> |
1861 | + <para> |
1862 | + Here's an example <s.conf; that |
1863 | + illustrates how this would be done: |
1864 | + </para> |
1865 | + <screen> |
1866 | +[Lab] |
1867 | + X_COLOR_DEPTH = 16 |
1868 | + X_MODE_0 = 1024x768 |
1869 | + |
1870 | +[Lowram] |
1871 | + X_VIDEO_RAM = 8096 |
1872 | + |
1873 | +[00:40:32:71:77:A1] |
1874 | + LIKE = Lab |
1875 | + |
1876 | +[00:70:84:BB:27:52] |
1877 | + LIKE = Lowram |
1878 | + </screen> |
1879 | + <para> |
1880 | + As you can see, using the <varname>LIKE</varname> keyword can |
1881 | + make your <s.conf; more readable, by |
1882 | + allowing you to group related parameters together into a single |
1883 | + symbolic name. |
1884 | + </para> |
1885 | + </sect1> |
1886 | + </chapter> |
1887 | + <chapter id='general-parms'> |
1888 | + <title>General thin client parameters</title> |
1889 | + <para> |
1890 | + There are several variables that one can define in the lts.conf |
1891 | + file which control how the thin client interacts with the server. |
1892 | + These are: |
1893 | + </para> |
1894 | + <xi:include |
1895 | + href='lts.conf.xml' |
1896 | + xpointer='lts-conf-general-list' |
1897 | + xmlns:xi='http://www.w3.org/2001/XInclude' /> |
1898 | + </chapter> |
1899 | + <chapter id='localdev-parms'> |
1900 | + <title>Local Device thin client parameters</title> |
1901 | + <para> |
1902 | + Local devices such as USB sticks, CD-ROM drives, or even floppy |
1903 | + disks need special configuration in order to be accessed from the |
1904 | + thin client. The following values allow to enable or disable the |
1905 | + use of various local devices: |
1906 | + </para> |
1907 | + <xi:include |
1908 | + href='lts.conf.xml' |
1909 | + xpointer='lts-conf-localdev-list' |
1910 | + xmlns:xi='http://www.w3.org/2001/XInclude' /> |
1911 | + </chapter> |
1912 | + <chapter id='screen-scripts'> |
1913 | + <title>Screen Scripts</title> |
1914 | + <para> |
1915 | + Screen scripts are how LTSP determines what type of login will |
1916 | + run on what virtual screen. Most GNU/Linux machines have 12 |
1917 | + virtual consoles, which you can access by pressing |
1918 | + <keycombo><keycap>Control</keycap><keycap>Alt</keycap><keycap>F1</keycap></keycombo>, through |
1919 | + <keycombo><keycap>Control</keycap><keycap>Alt</keycap><keycap>F12</keycap></keycombo>. |
1920 | + On some distributions there is a text based getty that is |
1921 | + started on screen 1, but you normally can't log into it, as there |
1922 | + are no local users on the thin client. |
1923 | + </para> |
1924 | + <para> |
1925 | + However, for debugging purposes, you may want to set up root to |
1926 | + log in on the thin client. You may need to do this if you're |
1927 | + debugging problems with local devices, for example. Fortunately, |
1928 | + it's easy to do: on the server, as root, just chroot into the LTSP chroot, |
1929 | + and set the password with passwd. |
1930 | + </para> |
1931 | + <screen> |
1932 | +chroot /opt/ltsp/<arch> |
1933 | +passwd |
1934 | + </screen> |
1935 | + <para> |
1936 | + By default, if there's nothing else mentioned in |
1937 | + <s.conf;, |
1938 | + an LDM session will be started on screen 7. |
1939 | + </para> |
1940 | + <xi:include |
1941 | + href='lts.conf.xml' |
1942 | + xpointer='lts-conf-screen-list' |
1943 | + xmlns:xi='http://www.w3.org/2001/XInclude' /> |
1944 | + </chapter> |
1945 | + <chapter id='rdesktop'> |
1946 | + <title>The "rdesktop" screen script</title> |
1947 | + <para> |
1948 | + LTSP includes an &rdesktop; screen script that |
1949 | + can be used to bring up a full screen RDP connection using |
1950 | + &rdesktop; on the thin client. This |
1951 | + screen script can be invoked in one of two ways. |
1952 | + </para> |
1953 | + <sect1> |
1954 | + <title>Single Line</title> |
1955 | + <para> |
1956 | + For example, by adding the following |
1957 | + <s.conf; parameter: |
1958 | + </para> |
1959 | + <screen> |
1960 | +SCREEN_07 = "rdesktop -a 16 192.168.0.253" |
1961 | + </screen> |
1962 | + <para> |
1963 | + That is, calling it with the arguments you normally would on the |
1964 | + command line. This method of invocation is useful in that you can have |
1965 | + multiple screens pointing to different RDP servers with different |
1966 | + arguments and switch between them. |
1967 | + </para> |
1968 | + </sect1> |
1969 | + <sect1> |
1970 | + <title>Multiple Line</title> |
1971 | + <para> |
1972 | + For example, by adding the following parameters to |
1973 | + <s.conf;: |
1974 | + </para> |
1975 | + <screen> |
1976 | +SCREEN_07 = rdesktop |
1977 | +RDP_SERVER = 192.168.0.253 |
1978 | +RDP_OPTIONS = "-a 16" |
1979 | + </screen> |
1980 | + <para> |
1981 | + This method will apply the same arguments and server to all screens. |
1982 | + </para> |
1983 | + </sect1> |
1984 | + <sect1> |
1985 | + <title>RDP + Local Devices</title> |
1986 | + <para> |
1987 | + When you run an &rdesktop; screen script, |
1988 | + &rdesktop; runs on the thin |
1989 | + client. The thin client, by default, has no way of automounting |
1990 | + removable devices, and the normal localdev approach used in an |
1991 | + &ldm; session in which local devices invoke a call over ssh to have the |
1992 | + Linux server mount the device obviously won't work. |
1993 | + </para> |
1994 | + <para> |
1995 | + To address this issue in a controlled way, we have chosen to use |
1996 | + <spfs; as a local automounter for RDP |
1997 | + sessions. To add local device support, you must: |
1998 | + </para> |
1999 | + <orderedlist> |
2000 | + <listitem> |
2001 | + <para> |
2002 | + Install ltspfs in the chroot. |
2003 | + </para> |
2004 | + </listitem> |
2005 | + <listitem> |
2006 | + <para> |
2007 | + Use folder redirection in &rdesktop; to map the local /media/root |
2008 | + folder created by ltspfs to the server as a shared drive. For |
2009 | + example, you could add the following &rdesktop; argument: |
2010 | + </para> |
2011 | + <screen> |
2012 | +-r disk:Drives=/media/root |
2013 | + </screen> |
2014 | + <para> |
2015 | + With this redirection, you should get a "Drives" share in Windows |
2016 | + under My Computer. Inside the "Drives" share, a folder will appear for |
2017 | + each local device. The local device will be mounted with ltspfs, so |
2018 | + they can just be removed when the device is not being written to |
2019 | + without "unmounting". |
2020 | + </para> |
2021 | + </listitem> |
2022 | + </orderedlist> |
2023 | + </sect1> |
2024 | + <sect1> |
2025 | + <title>RDP + Local Sound</title> |
2026 | + <para> |
2027 | + You should be able to add sound to your RDP session with the following |
2028 | + &rdesktop; argument: |
2029 | + </para> |
2030 | + <screen> |
2031 | +-r sound:local |
2032 | + </screen> |
2033 | + </sect1> |
2034 | + </chapter> |
2035 | + <chapter id='modules-scripts'> |
2036 | + <title>Modules and startup scripts</title> |
2037 | + <para> |
2038 | + For the most part, LTSP does a very good job of detecting what |
2039 | + hardware's on your thin client. However, it's possible that you |
2040 | + may want to manually specify a kernel module to load after boot. |
2041 | + Alternatively, you may have a script you've written that you've |
2042 | + put in the chroot, and want to make sure gets run at startup. LTSP |
2043 | + provides some hooks to allow you to do this. |
2044 | + </para> |
2045 | + <xi:include |
2046 | + href='lts.conf.xml' |
2047 | + xpointer='lts-conf-modules-list' |
2048 | + xmlns:xi='http://www.w3.org/2001/XInclude' /> |
2049 | + </chapter> |
2050 | + <chapter id='sound'> |
2051 | + <title>Sound in LTSP</title> |
2052 | + <para> |
2053 | + Sound in LTSP is handled by running the |
2054 | + <command>pulseaudio</command> daemon on the |
2055 | + thin client, which sits on top of the ALSA kernel drivers. The |
2056 | + thin client's kernel should detect the thin client sound hardware |
2057 | + via the usual udev mechanisms, and enable the sound card. At boot |
2058 | + time, the <command>pulseaudio</command> daemon is run, which allows the thin client to |
2059 | + receive audio streams via network connections. |
2060 | + </para> |
2061 | + <para> |
2062 | + On login, the LDM sets both the <varname>PULSE_SERVER</varname> and |
2063 | + <varname>ESPEAKER</varname> |
2064 | + environment variables for the X windows session, to allow the |
2065 | + server to re-route the sound over a TCP/IP socket to the thin |
2066 | + client. |
2067 | + </para> |
2068 | + <xi:include |
2069 | + href='lts.conf.xml' |
2070 | + xpointer='lts-conf-sound-list' |
2071 | + xmlns:xi='http://www.w3.org/2001/XInclude' /> |
2072 | + </chapter> |
2073 | + <chapter id='xwin-parms'> |
2074 | + <title>X-Windows parameters</title> |
2075 | + <para> |
2076 | + Setting up X windows on the thin client's normally a pretty easy |
2077 | + operation. The thin client uses X.org's own auto configuration |
2078 | + mode to let X determine what it thinks is installed in the box. |
2079 | + </para> |
2080 | + <para> |
2081 | + However, sometimes, this doesn't always work. Either due to |
2082 | + strange/buggy hardware, or buggy drivers in X.org, or because X |
2083 | + detects default settings that you don't want. For instance, it may |
2084 | + detect that your monitor is capable of doing 1280x1024, but you'd |
2085 | + prefer it to come up in 1024x768 resolution. Fortunately, you can |
2086 | + tweak individual X settings, or, alternatively, simply provide |
2087 | + your own <filename>xorg.conf</filename> to use. |
2088 | + </para> |
2089 | + <sect1> |
2090 | + <title>X.org Configuration</title> |
2091 | + <xi:include |
2092 | + href='lts.conf.xml' |
2093 | + xpointer='lts-conf-xorg-list' |
2094 | + xmlns:xi='http://www.w3.org/2001/XInclude' /> |
2095 | + </sect1> |
2096 | + </chapter> |
2097 | + <chapter id='xrandr'> |
2098 | + <title>XRANDR setting for managing displays</title> |
2099 | + <para> |
2100 | + The new Xorg Xserver has the ability to figure out (for the most part, |
2101 | + to the extent that the driver helps in the process) the best mode for |
2102 | + the videocard. Moreover, with the new dependency upon hal and Xrandr, |
2103 | + it is recommended to add input devices with hal and modify video modes |
2104 | + with Xrandr 1.2 calls. In essence, the xorg.conf becomes a place really |
2105 | + to fix deficiencies in poorly written drivers or to force certain |
2106 | + abnormal driver behavior in a particular environment in a way that can |
2107 | + not be otherwise done through hal or Xrandr. |
2108 | + </para> |
2109 | + <sect1> |
2110 | + <title>New Xorg structure within LTSP</title> |
2111 | + <para> |
2112 | + To accommodate this, Xorg now understands partial xorg.conf files. |
2113 | + Meaning you only add the sections that you need to force. Otherwise, it |
2114 | + discovers everything. That's why you might see minimalist xorg.conf |
2115 | + files in your LTSP chroot. |
2116 | + </para> |
2117 | + <para> |
2118 | + The <filename>screen-session.d/</filename> directory (located in the |
2119 | + chroot's <filename>/usr/share/ltsp directory</filename>) is a structure of shell scripts all |
2120 | + of which are sourced in order (similar to |
2121 | + <filename>Xsession.d/</filename> or <filename>rc.d/</filename> that you |
2122 | + may be familiar with). These scripts are executed upon the beginning of |
2123 | + each session but before the Xserver (if the session runs an Xserver) is |
2124 | + launched. You can make whatever script you want that may need to run at |
2125 | + that point. For LTSP, one thing we use it for is to set up |
2126 | + <emphasis>how</emphasis> the |
2127 | + Xserver will be launched. This entails not just generating a |
2128 | + <filename>xorg.conf</filename> |
2129 | + file as needed, but also configuring the parameters that the Xserver |
2130 | + should be launched with. The nice thing about a collection of sourced |
2131 | + scripts is that it gives flexibility to the distribution or to the |
2132 | + administrator to add additional scripts that may be required for that |
2133 | + distribution or for a particular network environment that will not |
2134 | + modify existing files (and therefore require more maintenance to care |
2135 | + for updates in the upstream code). |
2136 | + </para> |
2137 | + </sect1> |
2138 | + <sect1> |
2139 | + <title>Script structure</title> |
2140 | + <para> |
2141 | + Each script is named with a prefix letter, then an order number, then a |
2142 | + name. The prefix letter determines when the scripts of that prefix are |
2143 | + executed and the order number determines in what order. |
2144 | + </para> |
2145 | + <para> |
2146 | + PREFIXES: |
2147 | + Prefixes that may be used include: |
2148 | + </para> |
2149 | + <para> |
2150 | + S - Is a script that runs at the beginning of a session (screen script) |
2151 | + K - Is a script that runs at the end of a session (screen script) |
2152 | + XS - Is a script that is only run at the beginning of screen scripts |
2153 | + that run an Xserver |
2154 | + </para> |
2155 | + <para> |
2156 | + All of the scripts that generate a xorg.conf or modify the Xserver |
2157 | + arguments are XS* scripts. |
2158 | + </para> |
2159 | + <para> |
2160 | + These scripts are mostly organized by the particular lts.conf parameter |
2161 | + or function that they affect. For example, XS85-xvideoram adds the |
2162 | + ability to specify the X_VIDEO_RAM parameter in lts.conf and force the |
2163 | + amount of video ram used by the driver. |
2164 | + </para> |
2165 | + <para> |
2166 | + If you are going to create your own script, I recommend looking at other |
2167 | + scripts to understand the structure. Since many hacks may impact the |
2168 | + same xorg.conf sections, each section has a function of hacks assigned |
2169 | + to it, and in your script, you would create a function and add it to the |
2170 | + list of functions for that section. For example, if you add something |
2171 | + to the Monitor section (that cannot already be added through existing |
2172 | + functions) you would create a function in your script and add it to the |
2173 | + monitor_hacks function list. Again, easier to read the code and look at |
2174 | + examples to understand how to write a new script. |
2175 | + </para> |
2176 | + <para> |
2177 | + Also, please note that one of the lts.conf parameters you can specify |
2178 | + is: <varname>CONFIGURE_X_COMMAND</varname> |
2179 | + This should be set to a path to a script. So, if you have the old |
2180 | + configure-x.sh and like it better, simply copy it into the chroot, to |
2181 | + say, |
2182 | + <command>/opt/ltsp/<arch>/usr/share/ltsp/configure-x.sh</command> |
2183 | + and then in <s.conf;, |
2184 | + specify: <screen>CONFIGURE_X_COMMAND = |
2185 | + "/usr/share/ltsp/configure-x.sh"</screen> and |
2186 | + you will be back to where you were. |
2187 | + </para> |
2188 | + </sect1> |
2189 | + <sect1> |
2190 | + <title>XRandR parameters</title> |
2191 | + <xi:include |
2192 | + href='lts.conf.xml' |
2193 | + xpointer='lts-conf-xrandr-list' |
2194 | + xmlns:xi='http://www.w3.org/2001/XInclude' /> |
2195 | + </sect1> |
2196 | + </chapter> |
2197 | + <chapter id='printer'> |
2198 | + <title>Printer configuration parameters</title> |
2199 | + <para> |
2200 | + Sometimes, it's convenient to hang a printer off of a thin |
2201 | + client in a lab, so that the computer lab has access to local |
2202 | + printing resources. Fortunately, LTSP can accommodate printing on |
2203 | + the workstation. |
2204 | + </para> |
2205 | + <para> |
2206 | + LTSP can connect up to 3 printers per workstation to the network |
2207 | + via a small daemon called JetPipe. Both parallel and USB printers |
2208 | + are supported. JetPipe makes the printer look like a standard HP |
2209 | + Jet Direct printer interface. You can then create any cups printer |
2210 | + on your server, and point it at the printer via a Jet Direct |
2211 | + connection. |
2212 | + </para> |
2213 | + <para> |
2214 | + In your <filename>dhcpd.conf</filename> file that controls your |
2215 | + thin client IP assignments, you'll want to assign a static IP for |
2216 | + the terminal with the printers, to guarantee that it gets the same |
2217 | + IP address every time it boots. Otherwise, your printing won't |
2218 | + work if the terminal leases a different IP address. |
2219 | + </para> |
2220 | + <xi:include |
2221 | + href='lts.conf.xml' |
2222 | + xpointer='lts-conf-printer-list' |
2223 | + xmlns:xi='http://www.w3.org/2001/XInclude' /> |
2224 | + </chapter> |
2225 | + <chapter id='keyboard'> |
2226 | + <title>Keyboard parameters</title> |
2227 | + <para> |
2228 | + All of the keyboard support files are copied into the |
2229 | + <filename>/opt/ltsp/<arch></filename> hierarchy, so configuring international keyboard |
2230 | + support is simply a matter of configuring X.org. There are several |
2231 | + configuration parameters for this. |
2232 | + </para> |
2233 | + <para> |
2234 | + The values for the above parameters are from the X.org |
2235 | + documentation. Whatever is valid for X.org is valid for these |
2236 | + parameters. |
2237 | + </para> |
2238 | + <para> |
2239 | + We would like to add documentation to show what values are |
2240 | + needed for each type of international keyboard. If you work with |
2241 | + this and can configure your international keyboards, feedback to |
2242 | + LTSP would be greatly appreciated. |
2243 | + </para> |
2244 | + <xi:include |
2245 | + href='lts.conf.xml' |
2246 | + xpointer='lts-conf-keyboard-list' |
2247 | + xmlns:xi='http://www.w3.org/2001/XInclude' /> |
2248 | + </chapter> |
2249 | + <chapter id='touchscreen'> |
2250 | + <title>Touchscreen configuration</title> |
2251 | + <para> |
2252 | + Description to be added later. |
2253 | + </para> |
2254 | + <xi:include |
2255 | + href='lts.conf.xml' |
2256 | + xpointer='lts-conf-touchscreen-list' |
2257 | + xmlns:xi='http://www.w3.org/2001/XInclude' /> |
2258 | + </chapter> |
2259 | + <chapter id='localapps'> |
2260 | + <title>Local Applications</title> |
2261 | + <para> |
2262 | + Description to be added later. |
2263 | + </para> |
2264 | + <xi:include |
2265 | + href='lts.conf.xml' |
2266 | + xpointer='lts-conf-localapps-list' |
2267 | + xmlns:xi='http://www.w3.org/2001/XInclude' /> |
2268 | + </chapter> |
2269 | + <chapter id='ldm'> |
2270 | + <title>The LDM display manager</title> |
2271 | + <sect1> |
2272 | + <title>Introduction</title> |
2273 | + <para> |
2274 | + The LTSP Display Manager, or &ldm; is the |
2275 | + display manager specifically written by the LTSP project to |
2276 | + handle logins to a GNU/Linux server. It is the default display |
2277 | + manager for LTSP thin clients running under LTSP, and has a |
2278 | + lot of useful features: |
2279 | + </para> |
2280 | + <orderedlist> |
2281 | + <listitem> |
2282 | + <para> |
2283 | + It is written in C, for speed and efficiency on low |
2284 | + end clients. |
2285 | + </para> |
2286 | + </listitem> |
2287 | + <listitem> |
2288 | + <para> |
2289 | + It supports logging in via either a greeter (a |
2290 | + graphical login application) or autologin. |
2291 | + </para> |
2292 | + </listitem> |
2293 | + <listitem> |
2294 | + <para> |
2295 | + It can be configured to encrypt X Windows traffic, |
2296 | + for increased security, or leave it unencrypted, for |
2297 | + better performance on slower clients. |
2298 | + </para> |
2299 | + </listitem> |
2300 | + <listitem> |
2301 | + <para> |
2302 | + It contains a simple load-balancing system, to allow |
2303 | + the system administrator to allow load balancing |
2304 | + across several servers. |
2305 | + </para> |
2306 | + </listitem> |
2307 | + </orderedlist> |
2308 | + <para> |
2309 | + We'll go over the <s.conf; entries you'll |
2310 | + need to control these features below. |
2311 | + </para> |
2312 | + </sect1> |
2313 | + <sect1> |
2314 | + <title>Theory of operation</title> |
2315 | + <para> |
2316 | + To help understand the following sections, a bit of an |
2317 | + explanation of how &ldm; does it's work is |
2318 | + needed. Most thin client display managers tend to run up on |
2319 | + the server. The &ldm; display manager is |
2320 | + unique in that it runs on the thin client itself. This allows |
2321 | + the thin client to have a lot of choice as to how it will set |
2322 | + up the connection. A typical login session goes as follows: |
2323 | + </para> |
2324 | + <orderedlist> |
2325 | + <listitem> |
2326 | + <para> |
2327 | + &ldm; launches and starts up the X |
2328 | + Windows display on the thin client. |
2329 | + </para> |
2330 | + </listitem> |
2331 | + <listitem> |
2332 | + <para> |
2333 | + &ldm; starts up the greeter, which is |
2334 | + a graphical program which presents the user with a |
2335 | + nice login display and allows them to select their |
2336 | + session, language, and host they'd like to log into. |
2337 | + </para> |
2338 | + </listitem> |
2339 | + <listitem> |
2340 | + <para> |
2341 | + &ldm; collects the information from |
2342 | + the greeter, and starts an ssh session with the |
2343 | + server. This ssh connection is used to create an ssh |
2344 | + master socket, which is used by all subsequent |
2345 | + operations. |
2346 | + </para> |
2347 | + </listitem> |
2348 | + <listitem> |
2349 | + <para> |
2350 | + Now, the users selected session is started via the |
2351 | + master socket. Depending on whether or not an |
2352 | + encrypted connection has been requested, via the |
2353 | + <varname>LDM_DIRECTX</varname> parameter, the session is either connected |
2354 | + back to the local display via the ssh tunnel, or via a |
2355 | + regular TCP/IP connection. |
2356 | + </para> |
2357 | + </listitem> |
2358 | + <listitem> |
2359 | + <para> |
2360 | + During the session, any memory sticks, or other |
2361 | + local devices that are plugged in, communicate their |
2362 | + status to the server via the ssh control socket. |
2363 | + </para> |
2364 | + </listitem> |
2365 | + <listitem> |
2366 | + <para> |
2367 | + When the user exits the session, the ssh connection is |
2368 | + closed down, the X server is stopped, and &ldm; |
2369 | + restarts itself, so everything starts with a clean slate. |
2370 | + </para> |
2371 | + </listitem> |
2372 | + </orderedlist> |
2373 | + </sect1> |
2374 | + <sect1> |
2375 | + <title>Encrypted versus unencrypted sessions</title> |
2376 | + <para> |
2377 | + By default, LTSP5 encrypts the X session between the server. |
2378 | + This makes your session more secure, but at the cost of |
2379 | + increased processing power required on the thin client and on |
2380 | + the server. If processing power is a concern to you, it's very |
2381 | + easy to specify that the connection for either an individual |
2382 | + workstation, or the default setting should use an unencrypted |
2383 | + connection. To do so, simply specify: |
2384 | + </para> |
2385 | + <screen> |
2386 | +LDM_DIRECTX = True |
2387 | + </screen> |
2388 | + <para> |
2389 | + in your <s.conf; file in the appropriate |
2390 | + section. |
2391 | + </para> |
2392 | + </sect1> |
2393 | + <sect1> |
2394 | + <title>Auto login features</title> |
2395 | + <para> |
2396 | + This new version of LDM supports auto login of accounts, if |
2397 | + specified in the <s.conf; file. Simply |
2398 | + create a config section for each of the terminals you want to |
2399 | + log in automatically (you can use either MAC address, IP |
2400 | + address, or hostname) and specify the variable |
2401 | + <varname>LDM_USERNAME</varname> |
2402 | + and <varname>LDM_PASSWORD</varname>. Note that you must have |
2403 | + created these accounts on the server, with the passwords |
2404 | + specified. An example will serve to illustrate how to use |
2405 | + this: |
2406 | + </para> |
2407 | + </sect1> |
2408 | + <sect1> |
2409 | + <title>Load balancing features</title> |
2410 | + <para> |
2411 | + In this version of LTSP, there's a simple load-balancing |
2412 | + solution implemented that allows administrators to have |
2413 | + multiple LTSP servers on the network, and allow the thin |
2414 | + client to pick which one of the servers it would like to log |
2415 | + into. |
2416 | + </para> |
2417 | + <para> |
2418 | + The host selection system is simple and flexible enough to |
2419 | + allow administrators to implement their own policy on how they |
2420 | + want the load balancing to happen: either on a random, |
2421 | + load-based, or round robin system. See for details. |
2422 | + </para> |
2423 | + </sect1> |
2424 | + <sect1> |
2425 | + <title>RC script capabilities</title> |
2426 | + <para> |
2427 | + LDM has a very good system for handling user-supplied rc.d |
2428 | + scripts. This allows people looking to add site-specific |
2429 | + customizations to their LTSP setups an easy way to integrate |
2430 | + this functionality into LTSP. |
2431 | + </para> |
2432 | + <para> |
2433 | + These rc.d scripts can be placed in |
2434 | + <filename>$CHROOT/usr/share/ldm/rc.d/</filename>. They are |
2435 | + executed in the usual rc.d type method, so you must make sure |
2436 | + that any script you write will not make a call to |
2437 | + <command>exit</command>. |
2438 | + </para> |
2439 | + <para> |
2440 | + The files start with the letter I, S, K, or X, and have two |
2441 | + digits after them, allowing you to place them in order of |
2442 | + execution. The letters stand for: |
2443 | + </para> |
2444 | + <itemizedlist> |
2445 | + <listitem> |
2446 | + <para> |
2447 | + <emphasis>I</emphasis> scripts are executed at the |
2448 | + start of LDM, before the greeter has been presented. |
2449 | + </para> |
2450 | + </listitem> |
2451 | + <listitem> |
2452 | + <para> |
2453 | + <emphasis>S</emphasis> scripts are executed after the user |
2454 | + has logged in, but before the X session is run. |
2455 | + </para> |
2456 | + </listitem> |
2457 | + <listitem> |
2458 | + <para> |
2459 | + <emphasis>X</emphasis> scripts are executed while the X |
2460 | + session is being executed. |
2461 | + </para> |
2462 | + </listitem> |
2463 | + <listitem> |
2464 | + <para> |
2465 | + <emphasis>K</emphasis> scripts are executed after the X |
2466 | + session has ended, but before the user logs out entirely. |
2467 | + </para> |
2468 | + </listitem> |
2469 | + </itemizedlist> |
2470 | + <para> |
2471 | + Your scripts can make use of the following environment |
2472 | + variables in the S, X, and K scripts: |
2473 | + </para> |
2474 | + <variablelist> |
2475 | + <varlistentry> |
2476 | + <term><varname>LDM_USERNAME</varname></term> |
2477 | + <listitem> |
2478 | + <para> |
2479 | + This is the username the user supplied at login. |
2480 | + </para> |
2481 | + </listitem> |
2482 | + </varlistentry> |
2483 | + <varlistentry> |
2484 | + <term><varname>LDM_SOCKET</varname></term> |
2485 | + <listitem> |
2486 | + <para> |
2487 | + The path to the ssh control socket that LDM has open |
2488 | + for communication with the server. |
2489 | + </para> |
2490 | + </listitem> |
2491 | + </varlistentry> |
2492 | + <varlistentry> |
2493 | + <term><varname>LDM_SERVER</varname></term> |
2494 | + <listitem> |
2495 | + <para> |
2496 | + The current server that LDM is connected to. |
2497 | + </para> |
2498 | + </listitem> |
2499 | + </varlistentry> |
2500 | + <varlistentry> |
2501 | + <term><varname>LDMINFO_IPADDR</varname></term> |
2502 | + <listitem> |
2503 | + <para> |
2504 | + The IP address of the thin client. |
2505 | + </para> |
2506 | + </listitem> |
2507 | + </varlistentry> |
2508 | + </variablelist> |
2509 | + <para> |
2510 | + You can use these variables to create scripts that customize |
2511 | + behaviors at login time. For instance, lets say you were |
2512 | + running the GNOME desktop environment, and wanted to force your |
2513 | + users to have blank-only mode for their screen savers, to save |
2514 | + network bandwidth. |
2515 | + </para> |
2516 | + <para> |
2517 | + Since the script is actually running <emphasis>on the thin |
2518 | + client itself</emphasis>, you want this script to set this |
2519 | + up on the server, where the Gnome session is running. That's |
2520 | + where you can make use of the <varname>LDM_SOCKET</varname> and |
2521 | + <varname>LDM_SERVER</varname> environment variables to run an |
2522 | + <command>ssh</command> command on the server, using the control |
2523 | + socket that LDM has set up. Here's an example script. You |
2524 | + could install this into |
2525 | + <filename>$CHROOT/usr/share/ldm/rc.d/X01-set-blankonly</filename>: |
2526 | + </para> |
2527 | + <screen> |
2528 | +# |
2529 | +# sourced with . |
2530 | +# |
2531 | +# Script to automatically switch gnome screensaver to blank |
2532 | +# only mode |
2533 | +# |
2534 | +ssh -S ${LDM_SOCKET} ${LDM_SERVER} "/usr/bin/gconftool-2 --set --type string /apps/gnome-screensaver/mode blank-only" |
2535 | + </screen> |
2536 | + <para> |
2537 | + Using this mechanism, it's easy to customize your LTSP setup to |
2538 | + your needs. |
2539 | + </para> |
2540 | + </sect1> |
2541 | + <sect1> |
2542 | + <title>LDM lts.conf parameters</title> |
2543 | + <xi:include |
2544 | + href='lts.conf.xml' |
2545 | + xpointer='lts-conf-ldm-list' |
2546 | + xmlns:xi='http://www.w3.org/2001/XInclude' /> |
2547 | + </sect1> |
2548 | + <sect1> |
2549 | + <title>Multiple server setup</title> |
2550 | + <para> |
2551 | + A multiple server setup is useful for larger thin client |
2552 | + networks. Instead of using one big server, it makes it |
2553 | + possible to use smaller servers, and dispatch users on them. |
2554 | + You can adjust computing resources as the demand grows simply |
2555 | + by adding a new server. To make sure that every server behaves |
2556 | + the same from the users point of view, new services and |
2557 | + configurations that are required will be discussed. In |
2558 | + addition, some configurations specific to thin clients will be |
2559 | + presented. |
2560 | + </para> |
2561 | + </sect1> |
2562 | + </chapter> |
2563 | + <chapter id='infrastructure'> |
2564 | + <title>Infrastructure setup</title> |
2565 | + <sect1> |
2566 | + <title>Network topology</title> |
2567 | + <para> |
2568 | + The network topology is the same as a standalone server |
2569 | + setup, except that there are more than one server on the thin |
2570 | + client LAN. |
2571 | + </para> |
2572 | + <para> |
2573 | + You will need to select one server to behave as the primary |
2574 | + server. This server will be used to run additional services, |
2575 | + hold users files, and network boot thin clients. |
2576 | + </para> |
2577 | + <para> |
2578 | + Secondary servers will be used only to run desktop sessions. |
2579 | + They are simpler, and will be configured to use the central |
2580 | + services from the primary server. |
2581 | + </para> |
2582 | + </sect1> |
2583 | + <sect1> |
2584 | + <title>Common authentication</title> |
2585 | + <para> |
2586 | + A user should be able to start a session with the same login |
2587 | + and password, no matter which server it connects to. For this |
2588 | + purpose, a central authentication mechanism must be used. |
2589 | + There are many possibilities offered. Here are the major |
2590 | + technologies: |
2591 | + </para> |
2592 | + <orderedlist> |
2593 | + <listitem> |
2594 | + <para> |
2595 | + LDAP authentication: On the master server, setup an |
2596 | + OpenLDAP server. Configure each servers to use this |
2597 | + LDAP server as the authentication base. |
2598 | + </para> |
2599 | + </listitem> |
2600 | + <listitem> |
2601 | + <para> |
2602 | + NIS authentication: On the master server, setup a |
2603 | + NIS server. Configure each server to use this NIS |
2604 | + server for the authentication. |
2605 | + </para> |
2606 | + </listitem> |
2607 | + <listitem> |
2608 | + <para> |
2609 | + Winbind authentication: Useful if you already have |
2610 | + an Active Directory server. |
2611 | + </para> |
2612 | + </listitem> |
2613 | + </orderedlist> |
2614 | + <para> |
2615 | + For detailed instructions, see their respective manuals.</para> |
2616 | + </sect1> |
2617 | + <sect1> |
2618 | + <title>Shared home directories</title> |
2619 | + <para> |
2620 | + Shared home directories are easy to setup using an NFS server |
2621 | + on eithe the primary LTSP server, or even better, a standalone |
2622 | + NFS server. Other more modern, faster (and consequently more |
2623 | + expensive) options include a SAN, and maybe even moving to a |
2624 | + fibre-channel raid SAN. Consult your distribution's |
2625 | + documentation for details and suggestions for setting up an NFS |
2626 | + server. |
2627 | + </para> |
2628 | + </sect1> |
2629 | + <sect1> |
2630 | + <title>Shared printers</title> |
2631 | + <para> |
2632 | + For printers to be accessible on each server, the cups |
2633 | + server must be configured to share printers. Refer to the CUPS |
2634 | + manual for your distribution for detailed setup instructions. |
2635 | + </para> |
2636 | + <para> |
2637 | + Note that an LTSP thin client acting as a print server |
2638 | + resembles a JetDirect print server. |
2639 | + </para> |
2640 | + </sect1> |
2641 | + <sect1> |
2642 | + <title>Managing the SSH known hosts file</title> |
2643 | + <para> |
2644 | + For security reasons, a thin client won't connect to an |
2645 | + untrusted server. You must add the keys of secondary servers |
2646 | + inside the client root on the primary server. To do this, |
2647 | + first export the key file of the secondary server using LTSP's |
2648 | + tools. As root, run: |
2649 | + </para> |
2650 | + <screen> |
2651 | +ltsp-update-sshkeys --export ssh_known_hosts.myhostname |
2652 | + </screen> |
2653 | + <para> |
2654 | + Then, copy the file |
2655 | + <filename>ssh_known_hosts.myhostnam</filename> |
2656 | + to the primary server, in the directory |
2657 | + <filename>/etc/ltsp/</filename> |
2658 | + and run <command>ltsp-update-sshkeys</command> on the primary |
2659 | + server. Then, thin clients will trust the freshly added |
2660 | + server, and will be able to connect to it. |
2661 | + </para> |
2662 | + <para> |
2663 | + If a secondary server changes it's IP address, then this |
2664 | + procedure must be repeated. |
2665 | + </para> |
2666 | + </sect1> |
2667 | + <sect1> |
2668 | + <!-- FIXME: this should be generalized a bit, but since |
2669 | + iptables SHOULD be available on ALL distros, it can probably |
2670 | + stay. |
2671 | + --> |
2672 | + <title>Setting Network Forwarding or Masquerading</title> |
2673 | + <para> |
2674 | + The purpose of IP Masquerading is to allow machines with |
2675 | + private, non-routable IP addresses on your network to access |
2676 | + the Internet through the machine doing the masquerading. |
2677 | + Traffic from your private network destined for the Internet |
2678 | + must be manipulated for replies to be routable back to the |
2679 | + machine that made the request. To do this, the kernel must |
2680 | + modify the <emphasis>source</emphasis> IP address of each |
2681 | + packet so that replies will be routed back to it, rather than |
2682 | + to the private IP address that made the request, which is |
2683 | + impossible over the Internet. Linux uses |
2684 | + <emphasis>Connection Tracking</emphasis> |
2685 | + (conntrack) to keep track of which connections belong to which |
2686 | + machines and reroute each return packet accordingly. Traffic |
2687 | + leaving your private network is thus "masqueraded" as having |
2688 | + originated from your gateway machine. This process is referred |
2689 | + to in Microsoft documentation as Internet Connection Sharing. |
2690 | + </para> |
2691 | + <para> |
2692 | + IP Forwarding with IP Tables |
2693 | + </para> |
2694 | + <orderedlist> |
2695 | + <listitem> |
2696 | + <para> |
2697 | + to enable IPv4 packet forwarding by editing |
2698 | + /etc/sysctl.conf and uncomment the following |
2699 | + line:</para> |
2700 | + <screen>net.ipv4.ip_forward=1</screen> |
2701 | + </listitem> |
2702 | + <listitem> |
2703 | + <para> |
2704 | + If you wish to enable IPv6 forwarding also |
2705 | + uncomment: |
2706 | + </para> |
2707 | + <screen>net.ipv6.conf.default.forwarding=1</screen> |
2708 | + </listitem> |
2709 | + <listitem> |
2710 | + <para> |
2711 | + Next, execute the sysctl command to enable the new |
2712 | + settings in the configuration file: |
2713 | + </para> |
2714 | + <screen>sudo sysctl -p</screen> |
2715 | + </listitem> |
2716 | + <listitem> |
2717 | + <para> |
2718 | + IP Masquerading can now be accomplished with a |
2719 | + single iptables rule, which may differ slightly based |
2720 | + on your network configuration: |
2721 | + </para> |
2722 | + <screen>sudo iptables -t nat -A POSTROUTING -s 192.168.0.0/16 -o eth0 -j MASQUERADE</screen> |
2723 | + <para> |
2724 | + The above command assumes that your private address |
2725 | + space is 192.168.0.0/16 and that your Internet-facing |
2726 | + device is eth0. The syntax is broken down as follows: |
2727 | + </para> |
2728 | + </listitem> |
2729 | + </orderedlist> |
2730 | + </sect1> |
2731 | + </chapter> |
2732 | + <chapter id='session-dispatching'> |
2733 | + <title>Session dispatching</title> |
2734 | + <sect1> |
2735 | + <title>Define the server list</title> |
2736 | + <para> |
2737 | + LDM is a login manager for thin clients. Users can select a |
2738 | + server from the available ones in the host selection dialogue |
2739 | + box. |
2740 | + </para> |
2741 | + <para> |
2742 | + The displayed server list is defined by the |
2743 | + <varname>LDM_SERVER</varname> |
2744 | + parameter. This parameter accepts a list of server IP address |
2745 | + or host names, separated by space. If you use host names, then |
2746 | + your DNS resolution must work on the thin client. If defined |
2747 | + in the <s.conf; file, the list order will |
2748 | + be static, and the first server in the list will be selected |
2749 | + by default. |
2750 | + </para> |
2751 | + <para> |
2752 | + You can also compute a new order for the server list, by |
2753 | + creating the script |
2754 | + <command>/opt/ltsp/<arch>/usr/share/ltsp/get_hosts</command> |
2755 | + . The parameter |
2756 | + <varname>LDM_SERVER</varname> |
2757 | + overrides the |
2758 | + script. In consequence, this parameter must not be defined if |
2759 | + the <command>get_hosts</command> is going to be used. The |
2760 | + <command>get_hosts</command> |
2761 | + script writes on the standard output each server IP address or |
2762 | + host names, in the chosen order. |
2763 | + </para> |
2764 | + </sect1> |
2765 | + <sect1> |
2766 | + <title>Dispatching method</title> |
2767 | + <para> |
2768 | + You can change this behaviour by using a script to rearrange |
2769 | + the list. The simplest way to do it is by randomizing the |
2770 | + list. First, define a custom variable in the file |
2771 | + <s.conf; |
2772 | + , for example <varname>MY_SERVER_LIST</varname>, that will |
2773 | + contain the list of servers, the same way as |
2774 | + <varname>LDM_SERVER</varname> |
2775 | + Then, put the following script in |
2776 | + <command>/opt/ltsp/<arch>/usr/share/ltsp/get_hosts</command> |
2777 | + </para> |
2778 | + <screen> |
2779 | +#!/bin/bash |
2780 | +# Randomize the server list contained in MY_SERVER_LIST parameter |
2781 | +TMP_LIST="" |
2782 | +SHUFFLED_LIST="" |
2783 | + |
2784 | +for i in $MY_SERVER_LIST; do |
2785 | + rank=$RANDOM |
2786 | + let "rank %= 100" |
2787 | + TMP_LIST="$TMP_LIST\n${rank}_$i" |
2788 | +done |
2789 | + |
2790 | +TMP_LIST=$(echo -e $TMP_LIST | sort) |
2791 | +for i in $TMP_LIST; do |
2792 | + SHUFFLED_LIST="$SHUFFLED_LIST $(echo $i | cut -d_ -f2)" |
2793 | +done |
2794 | +echo $SHUFFLED_LIST |
2795 | + </screen> |
2796 | + <para> |
2797 | + More advanced load balancing algorithms can be written. For |
2798 | + example, load balancing can be done by querying ldminfod for |
2799 | + the server rating. By querying ldminfod, you can get the |
2800 | + current rating state of the server. This rating goes from 0 to |
2801 | + 100, higher is better. Here is an example of such a query: |
2802 | + <screen>nc localhost 9571 | grep rating | cut -d: -f2</screen> |
2803 | + </para> |
2804 | + </sect1> |
2805 | + <sect1> |
2806 | + <title>Network Swap</title> |
2807 | + <para> |
2808 | + Just like on a full fledged workstation, it helps to have |
2809 | + swap defined for your thin client. "Swap" is an area of disk |
2810 | + space set aside to allow you to transfer information out of |
2811 | + ram, and temporarily store it on a hard drive until it's |
2812 | + needed again. It makes the workstation look like it has more |
2813 | + memory than it actually does. For instance, if your |
2814 | + workstation has 64 Megabytes of ram and you configure 64 |
2815 | + Megabytes of swap, it's theoretically possible to load a 128 |
2816 | + Megabyte program. |
2817 | + </para> |
2818 | + <para> |
2819 | + We say, "theoretically", because in practice, you want to |
2820 | + avoid swapping as much as possible. A hard drive is several |
2821 | + orders of magnitude slower than ram, and, of course, on a thin |
2822 | + client, you don't even have a hard drive! You have to first |
2823 | + push the data through the network to the server's hard drive, |
2824 | + thus making your swapping even slower. In practice, it's best |
2825 | + to make sure you have enough ram in your thin client to handle |
2826 | + all your average memory needs. |
2827 | + </para> |
2828 | + <para> |
2829 | + However, sometimes that's not possible. Sometimes, you're |
2830 | + re-using old hardware, or you've simply got a program that |
2831 | + isn't normally used, but does consume a lot of ram on the thin |
2832 | + client when it does. Fortunately, LTSP supports swapping over |
2833 | + the network via NBD, or Network Block Devices. We include a |
2834 | + small shell script called nbdswapd, which is started via |
2835 | + inetd. It handles creating the swap file, and setting up the |
2836 | + swapping, and removing the swap file when it's no longer |
2837 | + needed, after the terminal shuts down. |
2838 | + </para> |
2839 | + <para> |
2840 | + By default, swap files are 64 Megabytes in size. This was |
2841 | + chosen to give your workstation a little extra ram, but not |
2842 | + use up too much disk space. If you get some random odd |
2843 | + behaviour, such as Firefox crashing when viewing web pages |
2844 | + with a lot of large pictures, you may want to try increasing |
2845 | + the size of the swap files. You can do so by creating a file |
2846 | + in the directory <filename>/etc/ltsp</filename> on the LTSP |
2847 | + server, called <filename>nbdswapd.conf</filename>. In it, you |
2848 | + can set the SIZE variable to the number of Megabytes you wish |
2849 | + the file to be sized to. For instance, to create 128 Megabyte |
2850 | + files, you'll want: SIZE=128 in the <filename>nbdswapd.conf</filename> file. |
2851 | + </para> |
2852 | + <para> |
2853 | + Please note that this is a global setting for all swap files. |
2854 | + If your server has 40 thin clients, each using 128 Megs of |
2855 | + memory, you'll need 128 * 40 = 5120, or a little over 5 |
2856 | + Gigabytes of space in your <filename>/tmp</filename> |
2857 | + directory, where the swap files are stored. |
2858 | + </para> |
2859 | + </sect1> |
2860 | + <sect1> |
2861 | + <!-- FIXME: Beginning to get too distro specific, and if people |
2862 | + want us to move to DNSMASQ, then this all changes anyway. Try to |
2863 | + generalize a bit more. |
2864 | + --> |
2865 | + <title>Managing DHCP</title> |
2866 | + <para> |
2867 | + DHCP stands for Dynamic Host Configuration Protocol and is the |
2868 | + very first thing your thin client uses to obtain an IP address |
2869 | + from the network, in order to allow it to start booting. In |
2870 | + LTSP, the dhcpd file is located in <filename>/etc/ltsp</filename>. |
2871 | + Any changes you want to make to booting behaviour should be made there. |
2872 | + </para> |
2873 | + <para> |
2874 | + By default, LTSP ships a <filename>dhcpd.conf</filename> that |
2875 | + serves thin clients in a dynamic range (i.e. it will hand out |
2876 | + ip addresses to anyone who asks for them) from 192.168.0.20 to |
2877 | + 192.168.0.250. The default dhcpd.conf file looks like: |
2878 | + </para> |
2879 | + <screen> |
2880 | +# |
2881 | +# Default LTSP dhcpd.conf config file. |
2882 | +# |
2883 | + |
2884 | +authoritative; |
2885 | + |
2886 | +subnet 192.168.0.0 netmask 255.255.255.0 { |
2887 | + range 192.168.0.20 192.168.0.250; |
2888 | + option domain-name "example.com"; |
2889 | + option domain-name-servers 192.168.0.1; |
2890 | + option broadcast-address 192.168.0.255; |
2891 | + option routers 192.168.0.1; |
2892 | +# next-server 192.168.0.1; |
2893 | +# get-lease-hostnames true; |
2894 | + option subnet-mask 255.255.255.0; |
2895 | + option root-path "/opt/ltsp/i386"; |
2896 | + if substring( option vendor-class-identifier, 0, 9 ) = "PXEClient" { |
2897 | + filename "/ltsp/i386/pxelinux.0"; |
2898 | + } else { |
2899 | + filename "/ltsp/i386/nbi.img"; |
2900 | + } |
2901 | +} |
2902 | + </screen> |
2903 | + <para> |
2904 | + This <filename>dhcpd.conf</filename> should handle most |
2905 | + situations. |
2906 | + </para> |
2907 | + <para> |
2908 | + By default, LTSP will detect an unused network interface and |
2909 | + configure it to be 192.168.0.254. LTSP's recommended single |
2910 | + server installation is to use a separate network interface for |
2911 | + the thin clients. If, however, you're not using two network |
2912 | + interfaces, or you already have an interface in the 192.168.0 |
2913 | + range, then you might have to configure the thin client |
2914 | + interface differently, which means you may have to adjust the |
2915 | + <filename>dhcpd.conf</filename> accordingly. |
2916 | + </para> |
2917 | + <para> |
2918 | + If the network interface that you're going to connect the thin |
2919 | + clients to has, say, a TCP/IP address of 10.0.20.254, you'll |
2920 | + want to replace every occurrence of 192.168.0 with 10.0.20 in |
2921 | + the <filename>dhcpd.conf</filename> file. |
2922 | + </para> |
2923 | + <para> |
2924 | + Always remember, you'll need to re-start the dhcp server if |
2925 | + you make any changes. You can do this by issuing the command: |
2926 | + </para> |
2927 | + <screen>sudo invoke-rc.d dhcp3-server restart</screen> |
2928 | + <para>(at the command prompt.)</para> |
2929 | + </sect1> |
2930 | + </chapter> |
2931 | + <chapter id='static-entries-dhcp'> |
2932 | + <title>Adding static entries to the dhcpd.conf</title> |
2933 | + <para> |
2934 | + Sometimes, you may need to have a certain terminal boot with a |
2935 | + guaranteed fixed TCP/IP address every time. Say, if you're |
2936 | + connecting a printer to the terminal, and need to make sure the |
2937 | + print server can find it at a fixed address. To create a fixed |
2938 | + address, use a low number in the range of 2-19, or otherwise, if |
2939 | + you change the range statement in the <filename>dhcpd.conf</filename>. |
2940 | + </para> |
2941 | + <para> |
2942 | + To create a static entry, simply add the following after the |
2943 | + "option root-path" line: |
2944 | + </para> |
2945 | + <screen> |
2946 | +host hostname { |
2947 | + hardware ethernet MA:CA:DD:RE:SS:00; |
2948 | + fixed-address 192.168.0.2; |
2949 | +} |
2950 | + </screen> |
2951 | + <para> |
2952 | + Substitude the MAC |
2953 | + address for the mac address of the thin client you wish to fix the |
2954 | + address of. The fixed-address will be the TCP/IP address you want, |
2955 | + and "hostname" is the name you wish to give the host. This kind of |
2956 | + setup is relatively complex and the admin should have a full |
2957 | + understanding of how DHCP works before attempting such a setup. |
2958 | + For more information, check the Internet. |
2959 | + </para> |
2960 | + </chapter> |
2961 | + <chapter id='dhcp-loadbalance'> |
2962 | + <title>DHCP failover load balancing</title> |
2963 | + <para> |
2964 | + Another common method of load balancing is to use DHCP |
2965 | + load balancing. There's an excellent writeup on the topic at: |
2966 | + <ulink url="https://wiki.edubuntu.org/EdubuntuDHCPload-balancingFailover"> |
2967 | + https://wiki.edubuntu.org/EdubuntuDHCPload-balancingFailover |
2968 | + </ulink> |
2969 | + </para> |
2970 | + <sect1> |
2971 | + <!-- FIXME: Should mention KDE's lockdown, and not go into so much |
2972 | + detail. Generalize a bit. |
2973 | + --> |
2974 | + <title/> |
2975 | + <sect2> |
2976 | + <title>Lockdown with Sabayon (user profile manager) and |
2977 | + Pessulus (lockdown editor) |
2978 | + </title> |
2979 | + <para> |
2980 | + A common requirement in both schools and businesses is |
2981 | + having the ability to lock down the desktop and provide |
2982 | + certain default configurations. |
2983 | + </para> |
2984 | + <para> |
2985 | + In LTSP, the applications you'll want to use are Sabayon |
2986 | + and Pessulus. You'll want to add them from the package |
2987 | + manager. |
2988 | + </para> |
2989 | + <para> |
2990 | + The Sabayon user profile editor looks like a window that |
2991 | + contains a smaller sized picture of your desktop. Within |
2992 | + this window, you can create a default layout: add icons to |
2993 | + panels and the desktop, lock down the panels so they can't |
2994 | + be modified, remove access to the command line, etc. |
2995 | + </para> |
2996 | + <para> |
2997 | + Once you're done, you can save your profile. You have |
2998 | + the option of applying your profile to either individual |
2999 | + users, or all users on the system. Please consult the |
3000 | + manual included with Sabayon for all the details. |
3001 | + </para> |
3002 | + <para> |
3003 | + More information is available here:</para> |
3004 | + <para> |
3005 | + <ulink url="http://live.gnome.org/PythonSabayon"> |
3006 | + http://live.gnome.org/PythonSabayon |
3007 | + </ulink> |
3008 | + </para> |
3009 | + <para> |
3010 | + <ulink url="http://www.gnome.org/~seth/blog/sabayon"> |
3011 | + http://www.gnome.org/~seth/blog/sabayon |
3012 | + </ulink> |
3013 | + </para> |
3014 | + <para> |
3015 | + http://www.gnome.org/projects/sabayon/</para> |
3016 | + </sect2> |
3017 | + </sect1> |
3018 | + <sect1> |
3019 | + <title>Replication of desktop profiles</title> |
3020 | + <para> |
3021 | + If you customize user's desktop, then custom desktop profiles |
3022 | + should be copied to every server. Gnome desktop profiles |
3023 | + created with Sabayon are located in |
3024 | + <filename>/etc/desktop-profiles</filename> |
3025 | + </para> |
3026 | + </sect1> |
3027 | + </chapter> |
3028 | + <chapter id='managing-clients'> |
3029 | + <!-- FIXME: Same. Too Gnome specific, and not really part of LTSP. |
3030 | + Should just provide a pointer to a wiki page. |
3031 | + --> |
3032 | + <title>Managing the thin client</title> |
3033 | + <para> |
3034 | + Previously, there was a program called TCM or thin client |
3035 | + manager, which was responsible for checking what was happening on |
3036 | + the various thin terminals, messaging between them, locking, or |
3037 | + generally offering support from a master terminal. This has now |
3038 | + been replaced by the use of Italc, which must be separately |
3039 | + installed depending on your distribution. |
3040 | + </para> |
3041 | + <sect1> |
3042 | + <title>Lockdown Editor</title> |
3043 | + <para> |
3044 | + By choosing a single user and right clicking on that users |
3045 | + name, you will open up the context menu. From here you can |
3046 | + choose "Lockdown", which will allow you to set options to |
3047 | + restrict a particular user. Clicking this menu item will |
3048 | + invoke the "Pessulus" program, which is the Gnome lockdown |
3049 | + editor. Ticking and unticking options in Pessulus will enable |
3050 | + and disable certain functions for that particular user. There |
3051 | + is a padlock next to each option in Pessulus. Ticking this |
3052 | + will make the option unchangeable by the user. This is called |
3053 | + a mandatory setting. For further help with Pessulus, please |
3054 | + refer to the Pessulus documentation. |
3055 | + </para> |
3056 | + </sect1> |
3057 | + </chapter> |
3058 | + <chapter id='updating-chroot'> |
3059 | + <!-- FIXME: Distro specific. Mention using a package manager to |
3060 | + update, and point them to a wiki page with more specific information. |
3061 | + --> |
3062 | + <title>Updating your LTSP chroot</title> |
3063 | + <para> |
3064 | + At some point in the future, updates will become available for |
3065 | + your LTSP server. You must remember that although you may have |
3066 | + applied all the updates to the server itself, as in the |
3067 | + instructions....HERE it is likely that the LTSP chroot will also |
3068 | + need updating. To do this you must open up a terminal and use the |
3069 | + following commands. |
3070 | + </para> |
3071 | + <para> |
3072 | + First make sure the Client environment has the same Package |
3073 | + lists as the Server, to achieve that, you will copy the |
3074 | + /etc/apt/sources.list (on Debian and Ubuntu) or the |
3075 | + /etc/yum.repos.d/fedora.repo file from the Server to the Client |
3076 | + environment. |
3077 | + </para> |
3078 | + <para> |
3079 | + Now issue the command below.</para> |
3080 | + <screen>sudo chroot /opt/ltsp/<arch></screen> |
3081 | + <para>(replace <arch> with the architecture |
3082 | + you are working with.) |
3083 | + </para> |
3084 | + <para> |
3085 | + This will change your root directory to be the LTSP clients root |
3086 | + directory. In essence, anything you now do inside here, will be |
3087 | + applied to the LTSP clients root. This is a separate small set of |
3088 | + files that are used to boot the clients into a usable, and enable |
3089 | + them to contact the LTSP server. Once inside this shell, we must |
3090 | + type the following command to obtain the latest list of packages |
3091 | + from the apt/yum servers. |
3092 | + </para> |
3093 | + <screen>apt-get get update</screen> |
3094 | + <para> |
3095 | + on Debian and Ubuntu |
3096 | + </para> |
3097 | + <para> |
3098 | + You need to mount <filename>/proc</filename> in the chroot before beginning, as some of |
3099 | + the packages you install may need resources in <filename>/proc</filename> to install correctly. |
3100 | + </para> |
3101 | + <screen>mount -t proc proc /proc</screen> |
3102 | + <para> |
3103 | + To be sure no deamons are started do the following:</para> |
3104 | + <screen>export LTSP_HANDLE_DAEMONS=false</screen> |
3105 | + <para> |
3106 | + Once this has completed you will have to upgrade the software in |
3107 | + the chroot by running the following command: |
3108 | + </para> |
3109 | + <screen>apt-get upgrade</screen> |
3110 | + <para>(on Debian and Ubuntu)</para> |
3111 | + <para> |
3112 | + or |
3113 | + </para> |
3114 | + <screen>yum update</screen> |
3115 | + <para>(on Fedora)</para> |
3116 | + <para> |
3117 | + Just in case <filename>/proc</filename> is still mounted when you exit the chroot, |
3118 | + unmount it first by doing:</para> |
3119 | + <screen>umount /proc</screen> |
3120 | + |
3121 | + <para> |
3122 | + Once you're done, you must leave the chroot by either typing |
3123 | + <emphasis>exit</emphasis> |
3124 | + or by using the key combination Ctrl+D. This will return you to |
3125 | + the root of the server. |
3126 | + </para> |
3127 | + <para> |
3128 | + If your kernel has been upgraded you must run the LTSP kernel |
3129 | + upgrade script, to ensure that your LTSP chroot uses the latest |
3130 | + version. This is performed by running the command below: |
3131 | + </para> |
3132 | + <screen>ltsp-update-kernels</screen> |
3133 | + <para> |
3134 | + All of your clients will now use the latest kernel upon their |
3135 | + next reboot. |
3136 | + </para> |
3137 | + <para> |
3138 | + Finally, you must remember to rebuild the NBD boot image from |
3139 | + your chroot with the following command: |
3140 | + </para> |
3141 | + <screen>ltsp-update-image</screen> |
3142 | + <para>(add architecture using -arch= addition)</para> |
3143 | + <para> |
3144 | + Be advised that this may take a few minutes, depending on the |
3145 | + speed of your server. |
3146 | + </para> |
3147 | + </chapter> |
3148 | + <chapter id='changing-server-ip'> |
3149 | + <title>Changing the IP of your LTSP server</title> |
3150 | + <para> |
3151 | + At some point in time, it may become necessary to change the IP |
3152 | + address of your LTSP server. Normally this does not present an |
3153 | + issue, but LTSP servers and clients communicate over and encrypted |
3154 | + channel and require all SSL certificates to be updated. Without |
3155 | + this update, <emphasis>no LTSP clients will be able to log in</emphasis>. |
3156 | + This is done by simply opening a terminal and running the following command. |
3157 | + </para> |
3158 | + <screen> |
3159 | +sudo ltsp-update-sshkeys |
3160 | +sudo ltsp-update-image |
3161 | + </screen> |
3162 | + </chapter> |
3163 | + <chapter id='appendix'> |
3164 | + <title>Appendix I</title> |
3165 | + <para> |
3166 | + Here you can find some solutions to common questions and |
3167 | + problems. |
3168 | + </para> |
3169 | + <sect1> |
3170 | + <title>Using NFS instead of NBD</title> |
3171 | + <para> |
3172 | + Using NBD instead of NFS has several advantages:</para> |
3173 | + <orderedlist> |
3174 | + <listitem> |
3175 | + <para> |
3176 | + Using a squashfs image we can now merge that |
3177 | + together in a unionfs to get writeable access which is |
3178 | + a lot faster during bootup. |
3179 | + </para> |
3180 | + </listitem> |
3181 | + <listitem> |
3182 | + <para> |
3183 | + A squashed root filesystem uses less network |
3184 | + bandwidth. |
3185 | + </para> |
3186 | + </listitem> |
3187 | + <listitem> |
3188 | + <para> |
3189 | + Many users and administrators have asked us to |
3190 | + eliminate NFS, for reasons of site policy. Since the |
3191 | + squashed image is now served out by nbd-server, which |
3192 | + is an entirely userspace program, and is started as |
3193 | + the user nobody, this should help to eliminate |
3194 | + concerns over NFS shares. |
3195 | + </para> |
3196 | + </listitem> |
3197 | + </orderedlist> |
3198 | + <para> |
3199 | + However, some people still want to use NFS. Fortunately, |
3200 | + it's easy to switch back to NFS, if it's so desired: |
3201 | + </para> |
3202 | + <orderedlist> |
3203 | + <listitem> |
3204 | + <para> |
3205 | + On the server, use the chroot command to maintain |
3206 | + the LTSP chroot:</para> |
3207 | + <screen>sudo chroot /opt/ltsp/<arch></screen> |
3208 | + </listitem> |
3209 | + <listitem> |
3210 | + <para> |
3211 | + Now edit /etc/default/ltsp-client-setup and change |
3212 | + the value of the root_write_method variable to use |
3213 | + bind mounts instead of unionfs, it should look like |
3214 | + this afterwards: |
3215 | + </para> |
3216 | + <screen>root_write_method="bind_mounts"</screen> |
3217 | + </listitem> |
3218 | + <listitem> |
3219 | + <para> |
3220 | + Next, create the file |
3221 | + <filename>/etc/initramfs-tools/conf.d/ltsp</filename> and add the following |
3222 | + line (set the value of the BOOT variable to nfs): |
3223 | + </para> |
3224 | + <screen>BOOT=nfs</screen> |
3225 | + </listitem> |
3226 | + <listitem> |
3227 | + <para> |
3228 | + Regenerate the initramfs: </para> |
3229 | + <screen>update-initramfs -u</screen> |
3230 | + </listitem> |
3231 | + <listitem> |
3232 | + <para> |
3233 | + Hit CTRL-D to exit the chroot now. Make sure LTSP |
3234 | + uses the new initramfs to boot: |
3235 | + </para> |
3236 | + <screen>sudo ltsp-update-kernels</screen> |
3237 | + </listitem> |
3238 | + </orderedlist> |
3239 | + </sect1> |
3240 | + <sect1> |
3241 | + <title>Enabling dual monitors</title> |
3242 | + <para> |
3243 | + First, I am going to start with a couple assumptions:</para> |
3244 | + <itemizedlist> |
3245 | + <listitem> |
3246 | + <para> |
3247 | + I will assume that you are operating thin clients |
3248 | + with an NBD file system in this write-up. |
3249 | + </para> |
3250 | + </listitem> |
3251 | + <listitem> |
3252 | + <para> |
3253 | + I will assume that you are running Ubuntu 8.04.1</para> |
3254 | + </listitem> |
3255 | + <listitem> |
3256 | + <para> |
3257 | + I will assume that you are running LTSP 5</para> |
3258 | + </listitem> |
3259 | + <listitem> |
3260 | + <para> |
3261 | + I will assume that you are replacing a running |
3262 | + image that has been properly tested, and is working. |
3263 | + </para> |
3264 | + </listitem> |
3265 | + </itemizedlist> |
3266 | + <para> |
3267 | + Create a |
3268 | + new image to ensure your configuration is congruent with my |
3269 | + successfully tested configuration. |
3270 | + </para> |
3271 | + <screen> |
3272 | +sudo ltsp-build-client --copy-sourceslist --arch i386 |
3273 | + </screen> |
3274 | + <para> |
3275 | + (note the --arch i386 command is required for my system |
3276 | + because it's running an amd64 kernel. It may not be required |
3277 | + for individuals running a 32-bit kernel) |
3278 | + </para> |
3279 | + <para> |
3280 | + Download the pertinent VIA unichrome driver for your chipset |
3281 | + from this web site: <ulink url="http://linux.via.com.tw/support/downloadFiles.action"> |
3282 | + http://linux.via.com.tw/support/downloadFiles.action |
3283 | + </ulink> |
3284 | + Be sure to select the proper OS as well. The installation |
3285 | + script is set up specifically for the directory structure of |
3286 | + each OS, and will error out if the wrong OS release is |
3287 | + installed. Next we need to move the downloaded file to the |
3288 | + image directory |
3289 | + </para> |
3290 | + <screen> |
3291 | +cp /home/<username/Desktop/chrome9.83-242-sl10.1.tar.gz /opt/ltsp/i386 |
3292 | + </screen> |
3293 | + <para> |
3294 | + After that, we need to chroot to the same image directory.</para> |
3295 | + <screen> |
3296 | +sudo chroot /opt/ltsp/i386/ |
3297 | + </screen> |
3298 | + <para> |
3299 | + Unpack the driver in the root directory </para> |
3300 | + <screen> |
3301 | +tar -zxvf chrome9.83-242-sl10.1.tar.gz |
3302 | + </screen> |
3303 | + <para> |
3304 | + After unpacking, enter the directory:</para> |
3305 | + <screen> |
3306 | +cd chrome9.83-242-sl10.1/ |
3307 | + </screen> |
3308 | + <para> |
3309 | + Run the file contained inside to start the driver installation |
3310 | + </para> |
3311 | + <screen> |
3312 | +./vinstall ..................done! Original X config file |
3313 | +was saved as /etc/X11/xorg.conf.viabak |
3314 | + </screen> |
3315 | + <para>(The following error: "VIAERROR:The /etc/X11/xorg.conf is |
3316 | + missing!" Can be ignored. We will be replacing the xorg.conf |
3317 | + anyway, and the drivers are still installed properly.) |
3318 | + </para> |
3319 | + <para> |
3320 | + Next we need to put a proper xorg.conf in to the proper |
3321 | + directory. |
3322 | + </para> |
3323 | + <screen> |
3324 | +gedit /etc/X11/xorg.conf |
3325 | + </screen> |
3326 | + <para> |
3327 | + Now paste the following in to the empty file: |
3328 | + </para> |
3329 | + <screen> |
3330 | +Section "Module" |
3331 | + Load "extmod" |
3332 | + Load "dbe" |
3333 | + Load "dri" |
3334 | + Load "glx" |
3335 | + Load "freetype" |
3336 | + Load "type1" |
3337 | +EndSection |
3338 | + |
3339 | +Section "Files" |
3340 | + RgbPath "/usr/X11R6/lib/X11/rgb" |
3341 | + FontPath "/usr/X11R6/lib/X11/fonts/misc/" |
3342 | + FontPath "/usr/X11R6/lib/X11/fonts/75dpi/:unscaled" |
3343 | + FontPath "/usr/X11R6/lib/X11/fonts/75dpi/" |
3344 | + FontPath "/usr/X11R6/lib/X11/fonts/Type1" |
3345 | + FontPath "/usr/X11R6/lib/X11/fonts/TTF" |
3346 | +EndSection |
3347 | + |
3348 | +Section "ServerFlags" |
3349 | + Option "Dont Zoom" |
3350 | + Option "AllowMouseOpenFail" "Yes" |
3351 | + Option "BlankTime" "20" |
3352 | + Option "StandbyTime" "0" |
3353 | + Option "SuspendTime" "0" |
3354 | + Option "OffTime" "0" |
3355 | + Option "Xinerama" "on" |
3356 | +EndSection |
3357 | + |
3358 | +Section "InputDevice" |
3359 | + Identifier "Keyboard1" |
3360 | + Driver "Keyboard" |
3361 | + Driver "keyboard" |
3362 | + Option "AutoRepeat" "500 30" |
3363 | + Option "XkbRules" "xfree86" |
3364 | + Option "XkbModel" "pc105" |
3365 | + Option "XkbLayout" "en_US,en_US" |
3366 | + Option "XkbOptions" "grp:alt_shift_toggle,grp_led:scroll" EndSection |
3367 | +EndSection |
3368 | + |
3369 | +Section "InputDevice" |
3370 | + Identifier "USBMouse" |
3371 | + Driver "mouse" |
3372 | + Option "Protocol" "IMPS/2" |
3373 | + Option "Device" "/dev/input/mice" |
3374 | + Option "ZAxisMapping" "4 5" |
3375 | + Option "Buttons" "5" |
3376 | +EndSection |
3377 | + |
3378 | +Section "InputDevice" |
3379 | + Identifier "Mouse1" |
3380 | + Driver "mouse" |
3381 | + Option "Protocol" "Auto" |
3382 | + Option "Device" "/dev/psaux" |
3383 | + Option "ZAxisMapping" "4 5" |
3384 | + Option "Buttons" "5" |
3385 | +EndSection |
3386 | + |
3387 | +Section "Monitor" |
3388 | + Identifier "Monitor0" |
3389 | + HorizSync 31.5-48.5 |
3390 | + VertRefresh 60 |
3391 | + Option "DPMS" |
3392 | +EndSection |
3393 | + |
3394 | +Section "Device" |
3395 | + Identifier "CN700" |
3396 | + Driver "via" |
3397 | + VideoRam 16384 |
3398 | + Screen 0 |
3399 | + Option "NoDDCValue" "1" |
3400 | + Option "Simultaneous" |
3401 | + Option "DPMS" "on" |
3402 | + BusID "PCI:1:0:0" |
3403 | +EndSection |
3404 | + |
3405 | +Section "Screen" |
3406 | + Identifier "Screen0DVI" |
3407 | + Device "CN700DVI" |
3408 | + Monitor "Monitor0DVI" |
3409 | + DefaultDepth 24 |
3410 | + Subsection "Display" |
3411 | + Depth 8 |
3412 | + Modes "1280x1024" |
3413 | + ViewPort 0 0 |
3414 | + EndSubsection |
3415 | + Subsection "Display" |
3416 | + Depth 16 |
3417 | + Modes "1280x1024" |
3418 | + ViewPort 0 0 |
3419 | + EndSubsection |
3420 | + Subsection "Display" |
3421 | + Depth 24 |
3422 | + Modes "1280x1024" |
3423 | + ViewPort 0 0 |
3424 | + EndSubsection |
3425 | +EndSection |
3426 | + |
3427 | +Section "Monitor" |
3428 | + Identifier "Monitor0" |
3429 | + HorizSync 31.5-48.5 |
3430 | + VertRefresh 60 |
3431 | + Option "DPMS" |
3432 | +EndSection |
3433 | + |
3434 | +Section "Device" |
3435 | + Identifier "CN700DVI" |
3436 | + Driver "via" |
3437 | + VideoRam 16384 |
3438 | + Screen 1 |
3439 | + Option "NoDDCValue" "1" |
3440 | + Option "Simultaneous" |
3441 | + Option "DPMS" "on" |
3442 | + BusID "PCI:1:0:0" |
3443 | +EndSection |
3444 | + |
3445 | +Section "Screen" |
3446 | + Identifier "Screen0" |
3447 | + Device "CN700" |
3448 | + Monitor "Monitor0" |
3449 | + DefaultDepth 24 |
3450 | + Subsection "Display" |
3451 | + Depth 8 |
3452 | + Modes "1280x1024" |
3453 | + ViewPort 0 0 |
3454 | + EndSubsection |
3455 | + Subsection "Display" |
3456 | + Depth 16 |
3457 | + Modes "1280x1024" |
3458 | + ViewPort 0 0 |
3459 | + EndSubsection |
3460 | + Subsection "Display" |
3461 | + Depth 24 |
3462 | + Modes "1280x1024" |
3463 | + ViewPort 0 0 |
3464 | + EndSubsection |
3465 | +EndSection |
3466 | + |
3467 | +Section "ServerLayout" |
3468 | + Identifier "Simple Layout" |
3469 | + Screen "Screen0" 0 0 |
3470 | + Screen 1 "Screen0DVI" LeftOf "Screen0" |
3471 | + InputDevice "Mouse1" "CorePointer" |
3472 | + InputDevice "Keyboard1" "CoreKeyboard" |
3473 | + InputDevice "USBMouse" "AlwaysCore" |
3474 | +EndSection |
3475 | + </screen> |
3476 | + <para> |
3477 | + <emphasis>IMPORTANT NOTE</emphasis> IN THE ABOVE SECTION PASTED INTO THE XORG.CONF, NOTICE |
3478 | + THAT THERE ARE RESOLUTIONS SPECIFIED PER MONITOR. PLEASE |
3479 | + ENSURE THAT YOU HAVE THE PROPER RESOLUTIONS FOR YOU YOUR |
3480 | + MONITOR ENTERED ON THOSE AREAS. Be sure to save the file as |
3481 | + xorg.conf and exit out of your chroot'd image. <CTRL+D or |
3482 | + "exit"> next we need to put in an addendum in the |
3483 | + lts.conf |
3484 | + </para> |
3485 | + <screen> |
3486 | +gedit /var/lib/tfpboot/ltsp/i386/lts.conf |
3487 | + </screen> |
3488 | + <para> |
3489 | + Feel |
3490 | + free to comment out anything that you need to in the lts.conf. |
3491 | + I will include my full lts.conf as an example: |
3492 | + </para> |
3493 | + <screen> |
3494 | +[DEFAULT] |
3495 | + #X_COLOR_DEPTH = "16" |
3496 | + #X_MODE_0 = "1680x1050" |
3497 | + #X_VERTREFRESH = "43-61" |
3498 | + #X_HORZSYNC = "28-85" |
3499 | + #X_OPTION_01 = "\"ForcePanel\" \"True\"" |
3500 | + #X_OPTION_01 = "\"NoPanel\" \"true\"" |
3501 | + #SCREEN_02 = shell |
3502 | + #SCREEN_07 = ldm |
3503 | + X_CONF = /etc/X11/xorg.conf |
3504 | + </screen> |
3505 | + <para> |
3506 | + Feel free to copy-paste this in it's |
3507 | + entirety if you want, but you will only need the last line. |
3508 | + After you add the X_CONF line, save & exit. Now we need to |
3509 | + make the changes that we have made take effect in the image |
3510 | + </para> |
3511 | + <screen> |
3512 | +# sudo ltsp-update-image --arch i386 |
3513 | + </screen> |
3514 | + <para>(again, the --arch i386 |
3515 | + will not be required for most, but I am putting it in just in |
3516 | + case a user has a x64 installation on their server.) That |
3517 | + should do it! Boot up the client and you should be good to go. |
3518 | + </para> |
3519 | + </sect1> |
3520 | + </chapter> |
3521 | +</book> |
3522 | |
3523 | === added file 'Makefile.in' |
3524 | --- Makefile.in 1970-01-01 00:00:00 +0000 |
3525 | +++ Makefile.in 2017-02-24 11:47:08 +0000 |
3526 | @@ -0,0 +1,111 @@ |
3527 | +# @configure_input@ |
3528 | + |
3529 | +# info about where we are |
3530 | +buildir=@builddir@ |
3531 | +abs_builddir=@abs_builddir@ |
3532 | +top_builddir=@top_builddir@ |
3533 | +srcdir=@srcdir@ |
3534 | +abs_srcdir=@abs_srcdir@ |
3535 | +top_srcdir=@top_srcdir@ |
3536 | +abs_top_srcdir=@abs_top_srcdir@ |
3537 | +datarootdir=@datarootdir@ |
3538 | + |
3539 | +# install stuff |
3540 | +prefix=@prefix@ |
3541 | +exec_prefix=@exec_prefix@ |
3542 | +bindir=@bindir@ |
3543 | +datadir=@datadir@ |
3544 | +mandir=@mandir@ |
3545 | +includedir=@includedir@ |
3546 | +infodir=@infodir@ |
3547 | +libdir=@libdir@ |
3548 | +libexecdir=@libexecdir@ |
3549 | +localstatedir=@localstatedir@ |
3550 | +oldincludedir=@oldincludedir@ |
3551 | +sbindir=@sbindir@ |
3552 | +sharedstatedir=@sharedstatedir@ |
3553 | +sysconfdir=@sysconfdir@ |
3554 | + |
3555 | +PDF_CMD=@PDF_CMD@ |
3556 | +XMLTO_CMD=@XMLTO_CMD@ |
3557 | +PDF_CMD_OPTS=-P latex.class.options=letterpaper,11pt,twoside,openright |
3558 | +XMLTO_CMD_OPTS= |
3559 | +INSTALL=@INSTALL@ |
3560 | +INSTALL_DATA=$(INSTALL) -D -m 0644 |
3561 | +INSTALL_DIR=$(INSTALL) -d |
3562 | + |
3563 | +# for pkgwrite |
3564 | +DESTDIR= |
3565 | +MANDIR=@mandir@ |
3566 | +DATADIR=@datadir@ |
3567 | + |
3568 | +########## |
3569 | +# Targets |
3570 | +########## |
3571 | +MANUALS=LTSPManual |
3572 | +MANPAGES=lts.conf |
3573 | + |
3574 | +########## |
3575 | +# Sources |
3576 | +########## |
3577 | +LTSPManual_SRC=LTSPManual.xml |
3578 | +MAN5=lts.conf.5 # config files - refsects in docbook |
3579 | +MAN8= # system commands |
3580 | + |
3581 | +# man pages aren't working yet |
3582 | +all: pdfs html man |
3583 | + |
3584 | +######################################### |
3585 | +# Do Not Edit Below This Line: |
3586 | +# (unless you know what you're doing...) |
3587 | +######################################### |
3588 | +.SECONDEXPANSION: |
3589 | + |
3590 | +pdfs: $(addsuffix .pdf, $(MANUALS)) |
3591 | +html: $(addsuffix .html, $(MANUALS)) |
3592 | +# txt: $(addsuffix .txt, $(MANUALS)) |
3593 | + |
3594 | +# man pages |
3595 | +man: $(MAN5) $(MAN8) |
3596 | + |
3597 | +%.pdf: $$($$*_SRC) |
3598 | + @echo "building $@...." |
3599 | + @$(PDF_CMD) $(PDF_CMD_OPTS) ${@:.pdf=.xml} |
3600 | + |
3601 | +%.html: $$($$*_SRC) |
3602 | + @echo "Building $@...." |
3603 | + @$(XMLTO_CMD) html-nochunks $(XML_CMD_OPTS) $(@:.html=.xml) |
3604 | + |
3605 | +$(MAN5) $(MAN8): $(MAN5:.5=.xml) $(MAN8:.8=.xml) |
3606 | + @echo "building man pages" |
3607 | + @echo "$(addsuffix .xml, $(MANPAGES))" | xargs -n1 $(XMLTO_CMD) man $(DOC_CMD_OPTS) |
3608 | + |
3609 | +# %.txt: $$($$*_SRC) |
3610 | +# @echo "building $@...." |
3611 | +# @$(XMLTO_CMD) txt $(XMLTO_CMD_OPTS) $(@:.txt=.xml) |
3612 | + |
3613 | +none: |
3614 | + |
3615 | +install: all |
3616 | + $(INSTALL_DIR) $(DESTDIR)$(MANDIR)/man5 |
3617 | + $(INSTALL_DATA) $(MAN5) $(DESTDIR)$(MANDIR)/man5/ |
3618 | + $(INSTALL_DIR) $(DESTDIR)$(datadir)/doc/ltsp |
3619 | + $(INSTALL_DATA) *.pdf $(DESTDIR)$(datadir)/doc/ltsp/. |
3620 | + $(INSTALL_DATA) *.html $(DESTDIR)$(datadir)/doc/ltsp/ |
3621 | + $(INSTALL_DIR) $(DESTDIR)$(datadir)/gnome/help/ltsp/C |
3622 | + $(INSTALL_DATA) *.xml $(DESTDIR)$(datadir)/gnome/help/ltsp/C/ |
3623 | + $(INSTALL_DIR) $(DESTDIR)$(datadir)/omf/ltsp |
3624 | + $(INSTALL_DATA) *.omf $(DESTDIR)$(datadir)/omf/ltsp/ |
3625 | + |
3626 | +clean: |
3627 | + @bash -c "rm -f *.{5,8,html,doc,pdf,txt,rtf,links,refs,tex,aux}" |
3628 | + #@bash -c "rm -f ../*.{5,8,html,doc,pdf,txt,rtf}" |
3629 | + @bash -c "rm -f ../*.pdf" |
3630 | + @bash -c "rm -rf $(MANUALS)" |
3631 | + @bash -c "rm -f config.log config.status" |
3632 | + |
3633 | +yelp: |
3634 | + @bash -c "yelp file:///`pwd`/LTSPManual.xml" |
3635 | + |
3636 | +distclean: clean |
3637 | + @bash -c "rm -rf autom4te.cache Makefile config.status config.log" |
3638 | |
3639 | === added file 'config.guess' |
3640 | --- config.guess 1970-01-01 00:00:00 +0000 |
3641 | +++ config.guess 2017-02-24 11:47:08 +0000 |
3642 | @@ -0,0 +1,1465 @@ |
3643 | +#! /bin/sh |
3644 | +# Attempt to guess a canonical system name. |
3645 | +# Copyright (C) 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, |
3646 | +# 2000, 2001, 2002, 2003, 2004, 2005 Free Software Foundation, Inc. |
3647 | + |
3648 | +timestamp='2005-04-22' |
3649 | + |
3650 | +# This file is free software; you can redistribute it and/or modify it |
3651 | +# under the terms of the GNU General Public License as published by |
3652 | +# the Free Software Foundation; either version 2 of the License, or |
3653 | +# (at your option) any later version. |
3654 | +# |
3655 | +# This program is distributed in the hope that it will be useful, but |
3656 | +# WITHOUT ANY WARRANTY; without even the implied warranty of |
3657 | +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
3658 | +# General Public License for more details. |
3659 | +# |
3660 | +# You should have received a copy of the GNU General Public License |
3661 | +# along with this program; if not, write to the Free Software |
3662 | +# Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. |
3663 | +# |
3664 | +# As a special exception to the GNU General Public License, if you |
3665 | +# distribute this file as part of a program that contains a |
3666 | +# configuration script generated by Autoconf, you may include it under |
3667 | +# the same distribution terms that you use for the rest of that program. |
3668 | + |
3669 | +# Originally written by Per Bothner <per@bothner.com>. |
3670 | +# Please send patches to <config-patches@gnu.org>. Submit a context |
3671 | +# diff and a properly formatted ChangeLog entry. |
3672 | +# |
3673 | +# This script attempts to guess a canonical system name similar to |
3674 | +# config.sub. If it succeeds, it prints the system name on stdout, and |
3675 | +# exits with 0. Otherwise, it exits with 1. |
3676 | +# |
3677 | +# The plan is that this can be called by configure scripts if you |
3678 | +# don't specify an explicit build system type. |
3679 | + |
3680 | +me=`echo "$0" | sed -e 's,.*/,,'` |
3681 | + |
3682 | +usage="\ |
3683 | +Usage: $0 [OPTION] |
3684 | + |
3685 | +Output the configuration name of the system \`$me' is run on. |
3686 | + |
3687 | +Operation modes: |
3688 | + -h, --help print this help, then exit |
3689 | + -t, --time-stamp print date of last modification, then exit |
3690 | + -v, --version print version number, then exit |
3691 | + |
3692 | +Report bugs and patches to <config-patches@gnu.org>." |
3693 | + |
3694 | +version="\ |
3695 | +GNU config.guess ($timestamp) |
3696 | + |
3697 | +Originally written by Per Bothner. |
3698 | +Copyright (C) 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005 |
3699 | +Free Software Foundation, Inc. |
3700 | + |
3701 | +This is free software; see the source for copying conditions. There is NO |
3702 | +warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE." |
3703 | + |
3704 | +help=" |
3705 | +Try \`$me --help' for more information." |
3706 | + |
3707 | +# Parse command line |
3708 | +while test $# -gt 0 ; do |
3709 | + case $1 in |
3710 | + --time-stamp | --time* | -t ) |
3711 | + echo "$timestamp" ; exit 0 ;; |
3712 | + --version | -v ) |
3713 | + echo "$version" ; exit 0 ;; |
3714 | + --help | --h* | -h ) |
3715 | + echo "$usage"; exit 0 ;; |
3716 | + -- ) # Stop option processing |
3717 | + shift; break ;; |
3718 | + - ) # Use stdin as input. |
3719 | + break ;; |
3720 | + -* ) |
3721 | + echo "$me: invalid option $1$help" >&2 |
3722 | + exit 1 ;; |
3723 | + * ) |
3724 | + break ;; |
3725 | + esac |
3726 | +done |
3727 | + |
3728 | +if test $# != 0; then |
3729 | + echo "$me: too many arguments$help" >&2 |
3730 | + exit 1 |
3731 | +fi |
3732 | + |
3733 | +trap 'exit 1' 1 2 15 |
3734 | + |
3735 | +# CC_FOR_BUILD -- compiler used by this script. Note that the use of a |
3736 | +# compiler to aid in system detection is discouraged as it requires |
3737 | +# temporary files to be created and, as you can see below, it is a |
3738 | +# headache to deal with in a portable fashion. |
3739 | + |
3740 | +# Historically, `CC_FOR_BUILD' used to be named `HOST_CC'. We still |
3741 | +# use `HOST_CC' if defined, but it is deprecated. |
3742 | + |
3743 | +# Portable tmp directory creation inspired by the Autoconf team. |
3744 | + |
3745 | +set_cc_for_build=' |
3746 | +trap "exitcode=\$?; (rm -f \$tmpfiles 2>/dev/null; rmdir \$tmp 2>/dev/null) && exit \$exitcode" 0 ; |
3747 | +trap "rm -f \$tmpfiles 2>/dev/null; rmdir \$tmp 2>/dev/null; exit 1" 1 2 13 15 ; |
3748 | +: ${TMPDIR=/tmp} ; |
3749 | + { tmp=`(umask 077 && mktemp -d -q "$TMPDIR/cgXXXXXX") 2>/dev/null` && test -n "$tmp" && test -d "$tmp" ; } || |
3750 | + { test -n "$RANDOM" && tmp=$TMPDIR/cg$$-$RANDOM && (umask 077 && mkdir $tmp) ; } || |
3751 | + { tmp=$TMPDIR/cg-$$ && (umask 077 && mkdir $tmp) && echo "Warning: creating insecure temp directory" >&2 ; } || |
3752 | + { echo "$me: cannot create a temporary directory in $TMPDIR" >&2 ; exit 1 ; } ; |
3753 | +dummy=$tmp/dummy ; |
3754 | +tmpfiles="$dummy.c $dummy.o $dummy.rel $dummy" ; |
3755 | +case $CC_FOR_BUILD,$HOST_CC,$CC in |
3756 | + ,,) echo "int x;" > $dummy.c ; |
3757 | + for c in cc gcc c89 c99 ; do |
3758 | + if ($c -c -o $dummy.o $dummy.c) >/dev/null 2>&1 ; then |
3759 | + CC_FOR_BUILD="$c"; break ; |
3760 | + fi ; |
3761 | + done ; |
3762 | + if test x"$CC_FOR_BUILD" = x ; then |
3763 | + CC_FOR_BUILD=no_compiler_found ; |
3764 | + fi |
3765 | + ;; |
3766 | + ,,*) CC_FOR_BUILD=$CC ;; |
3767 | + ,*,*) CC_FOR_BUILD=$HOST_CC ;; |
3768 | +esac ;' |
3769 | + |
3770 | +# This is needed to find uname on a Pyramid OSx when run in the BSD universe. |
3771 | +# (ghazi@noc.rutgers.edu 1994-08-24) |
3772 | +if (test -f /.attbin/uname) >/dev/null 2>&1 ; then |
3773 | + PATH=$PATH:/.attbin ; export PATH |
3774 | +fi |
3775 | + |
3776 | +UNAME_MACHINE=`(uname -m) 2>/dev/null` || UNAME_MACHINE=unknown |
3777 | +UNAME_RELEASE=`(uname -r) 2>/dev/null` || UNAME_RELEASE=unknown |
3778 | +UNAME_SYSTEM=`(uname -s) 2>/dev/null` || UNAME_SYSTEM=unknown |
3779 | +UNAME_VERSION=`(uname -v) 2>/dev/null` || UNAME_VERSION=unknown |
3780 | + |
3781 | +# Note: order is significant - the case branches are not exclusive. |
3782 | + |
3783 | +case "${UNAME_MACHINE}:${UNAME_SYSTEM}:${UNAME_RELEASE}:${UNAME_VERSION}" in |
3784 | + *:NetBSD:*:*) |
3785 | + # NetBSD (nbsd) targets should (where applicable) match one or |
3786 | + # more of the tupples: *-*-netbsdelf*, *-*-netbsdaout*, |
3787 | + # *-*-netbsdecoff* and *-*-netbsd*. For targets that recently |
3788 | + # switched to ELF, *-*-netbsd* would select the old |
3789 | + # object file format. This provides both forward |
3790 | + # compatibility and a consistent mechanism for selecting the |
3791 | + # object file format. |
3792 | + # |
3793 | + # Note: NetBSD doesn't particularly care about the vendor |
3794 | + # portion of the name. We always set it to "unknown". |
3795 | + sysctl="sysctl -n hw.machine_arch" |
3796 | + UNAME_MACHINE_ARCH=`(/sbin/$sysctl 2>/dev/null || \ |
3797 | + /usr/sbin/$sysctl 2>/dev/null || echo unknown)` |
3798 | + case "${UNAME_MACHINE_ARCH}" in |
3799 | + armeb) machine=armeb-unknown ;; |
3800 | + arm*) machine=arm-unknown ;; |
3801 | + sh3el) machine=shl-unknown ;; |
3802 | + sh3eb) machine=sh-unknown ;; |
3803 | + *) machine=${UNAME_MACHINE_ARCH}-unknown ;; |
3804 | + esac |
3805 | + # The Operating System including object format, if it has switched |
3806 | + # to ELF recently, or will in the future. |
3807 | + case "${UNAME_MACHINE_ARCH}" in |
3808 | + arm*|i386|m68k|ns32k|sh3*|sparc|vax) |
3809 | + eval $set_cc_for_build |
3810 | + if echo __ELF__ | $CC_FOR_BUILD -E - 2>/dev/null \ |
3811 | + | grep __ELF__ >/dev/null |
3812 | + then |
3813 | + # Once all utilities can be ECOFF (netbsdecoff) or a.out (netbsdaout). |
3814 | + # Return netbsd for either. FIX? |
3815 | + os=netbsd |
3816 | + else |
3817 | + os=netbsdelf |
3818 | + fi |
3819 | + ;; |
3820 | + *) |
3821 | + os=netbsd |
3822 | + ;; |
3823 | + esac |
3824 | + # The OS release |
3825 | + # Debian GNU/NetBSD machines have a different userland, and |
3826 | + # thus, need a distinct triplet. However, they do not need |
3827 | + # kernel version information, so it can be replaced with a |
3828 | + # suitable tag, in the style of linux-gnu. |
3829 | + case "${UNAME_VERSION}" in |
3830 | + Debian*) |
3831 | + release='-gnu' |
3832 | + ;; |
3833 | + *) |
3834 | + release=`echo ${UNAME_RELEASE}|sed -e 's/[-_].*/\./'` |
3835 | + ;; |
3836 | + esac |
3837 | + # Since CPU_TYPE-MANUFACTURER-KERNEL-OPERATING_SYSTEM: |
3838 | + # contains redundant information, the shorter form: |
3839 | + # CPU_TYPE-MANUFACTURER-OPERATING_SYSTEM is used. |
3840 | + echo "${machine}-${os}${release}" |
3841 | + exit 0 ;; |
3842 | + amd64:OpenBSD:*:*) |
3843 | + echo x86_64-unknown-openbsd${UNAME_RELEASE} |
3844 | + exit 0 ;; |
3845 | + amiga:OpenBSD:*:*) |
3846 | + echo m68k-unknown-openbsd${UNAME_RELEASE} |
3847 | + exit 0 ;; |
3848 | + cats:OpenBSD:*:*) |
3849 | + echo arm-unknown-openbsd${UNAME_RELEASE} |
3850 | + exit 0 ;; |
3851 | + hp300:OpenBSD:*:*) |
3852 | + echo m68k-unknown-openbsd${UNAME_RELEASE} |
3853 | + exit 0 ;; |
3854 | + luna88k:OpenBSD:*:*) |
3855 | + echo m88k-unknown-openbsd${UNAME_RELEASE} |
3856 | + exit 0 ;; |
3857 | + mac68k:OpenBSD:*:*) |
3858 | + echo m68k-unknown-openbsd${UNAME_RELEASE} |
3859 | + exit 0 ;; |
3860 | + macppc:OpenBSD:*:*) |
3861 | + echo powerpc-unknown-openbsd${UNAME_RELEASE} |
3862 | + exit 0 ;; |
3863 | + mvme68k:OpenBSD:*:*) |
3864 | + echo m68k-unknown-openbsd${UNAME_RELEASE} |
3865 | + exit 0 ;; |
3866 | + mvme88k:OpenBSD:*:*) |
3867 | + echo m88k-unknown-openbsd${UNAME_RELEASE} |
3868 | + exit 0 ;; |
3869 | + mvmeppc:OpenBSD:*:*) |
3870 | + echo powerpc-unknown-openbsd${UNAME_RELEASE} |
3871 | + exit 0 ;; |
3872 | + sgi:OpenBSD:*:*) |
3873 | + echo mips64-unknown-openbsd${UNAME_RELEASE} |
3874 | + exit 0 ;; |
3875 | + sun3:OpenBSD:*:*) |
3876 | + echo m68k-unknown-openbsd${UNAME_RELEASE} |
3877 | + exit 0 ;; |
3878 | + *:OpenBSD:*:*) |
3879 | + echo ${UNAME_MACHINE}-unknown-openbsd${UNAME_RELEASE} |
3880 | + exit 0 ;; |
3881 | + *:ekkoBSD:*:*) |
3882 | + echo ${UNAME_MACHINE}-unknown-ekkobsd${UNAME_RELEASE} |
3883 | + exit 0 ;; |
3884 | + macppc:MirBSD:*:*) |
3885 | + echo powerppc-unknown-mirbsd${UNAME_RELEASE} |
3886 | + exit 0 ;; |
3887 | + *:MirBSD:*:*) |
3888 | + echo ${UNAME_MACHINE}-unknown-mirbsd${UNAME_RELEASE} |
3889 | + exit 0 ;; |
3890 | + alpha:OSF1:*:*) |
3891 | + case $UNAME_RELEASE in |
3892 | + *4.0) |
3893 | + UNAME_RELEASE=`/usr/sbin/sizer -v | awk '{print $3}'` |
3894 | + ;; |
3895 | + *5.*) |
3896 | + UNAME_RELEASE=`/usr/sbin/sizer -v | awk '{print $4}'` |
3897 | + ;; |
3898 | + esac |
3899 | + # According to Compaq, /usr/sbin/psrinfo has been available on |
3900 | + # OSF/1 and Tru64 systems produced since 1995. I hope that |
3901 | + # covers most systems running today. This code pipes the CPU |
3902 | + # types through head -n 1, so we only detect the type of CPU 0. |
3903 | + ALPHA_CPU_TYPE=`/usr/sbin/psrinfo -v | sed -n -e 's/^ The alpha \(.*\) processor.*$/\1/p' | head -n 1` |
3904 | + case "$ALPHA_CPU_TYPE" in |
3905 | + "EV4 (21064)") |
3906 | + UNAME_MACHINE="alpha" ;; |
3907 | + "EV4.5 (21064)") |
3908 | + UNAME_MACHINE="alpha" ;; |
3909 | + "LCA4 (21066/21068)") |
3910 | + UNAME_MACHINE="alpha" ;; |
3911 | + "EV5 (21164)") |
3912 | + UNAME_MACHINE="alphaev5" ;; |
3913 | + "EV5.6 (21164A)") |
3914 | + UNAME_MACHINE="alphaev56" ;; |
3915 | + "EV5.6 (21164PC)") |
3916 | + UNAME_MACHINE="alphapca56" ;; |
3917 | + "EV5.7 (21164PC)") |
3918 | + UNAME_MACHINE="alphapca57" ;; |
3919 | + "EV6 (21264)") |
3920 | + UNAME_MACHINE="alphaev6" ;; |
3921 | + "EV6.7 (21264A)") |
3922 | + UNAME_MACHINE="alphaev67" ;; |
3923 | + "EV6.8CB (21264C)") |
3924 | + UNAME_MACHINE="alphaev68" ;; |
3925 | + "EV6.8AL (21264B)") |
3926 | + UNAME_MACHINE="alphaev68" ;; |
3927 | + "EV6.8CX (21264D)") |
3928 | + UNAME_MACHINE="alphaev68" ;; |
3929 | + "EV6.9A (21264/EV69A)") |
3930 | + UNAME_MACHINE="alphaev69" ;; |
3931 | + "EV7 (21364)") |
3932 | + UNAME_MACHINE="alphaev7" ;; |
3933 | + "EV7.9 (21364A)") |
3934 | + UNAME_MACHINE="alphaev79" ;; |
3935 | + esac |
3936 | + # A Pn.n version is a patched version. |
3937 | + # A Vn.n version is a released version. |
3938 | + # A Tn.n version is a released field test version. |
3939 | + # A Xn.n version is an unreleased experimental baselevel. |
3940 | + # 1.2 uses "1.2" for uname -r. |
3941 | + echo ${UNAME_MACHINE}-dec-osf`echo ${UNAME_RELEASE} | sed -e 's/^[PVTX]//' | tr 'ABCDEFGHIJKLMNOPQRSTUVWXYZ' 'abcdefghijklmnopqrstuvwxyz'` |
3942 | + exit 0 ;; |
3943 | + Alpha\ *:Windows_NT*:*) |
3944 | + # How do we know it's Interix rather than the generic POSIX subsystem? |
3945 | + # Should we change UNAME_MACHINE based on the output of uname instead |
3946 | + # of the specific Alpha model? |
3947 | + echo alpha-pc-interix |
3948 | + exit 0 ;; |
3949 | + 21064:Windows_NT:50:3) |
3950 | + echo alpha-dec-winnt3.5 |
3951 | + exit 0 ;; |
3952 | + Amiga*:UNIX_System_V:4.0:*) |
3953 | + echo m68k-unknown-sysv4 |
3954 | + exit 0;; |
3955 | + *:[Aa]miga[Oo][Ss]:*:*) |
3956 | + echo ${UNAME_MACHINE}-unknown-amigaos |
3957 | + exit 0 ;; |
3958 | + *:[Mm]orph[Oo][Ss]:*:*) |
3959 | + echo ${UNAME_MACHINE}-unknown-morphos |
3960 | + exit 0 ;; |
3961 | + *:OS/390:*:*) |
3962 | + echo i370-ibm-openedition |
3963 | + exit 0 ;; |
3964 | + *:z/VM:*:*) |
3965 | + echo s390-ibm-zvmoe |
3966 | + exit 0 ;; |
3967 | + *:OS400:*:*) |
3968 | + echo powerpc-ibm-os400 |
3969 | + exit 0 ;; |
3970 | + arm:RISC*:1.[012]*:*|arm:riscix:1.[012]*:*) |
3971 | + echo arm-acorn-riscix${UNAME_RELEASE} |
3972 | + exit 0;; |
3973 | + SR2?01:HI-UX/MPP:*:* | SR8000:HI-UX/MPP:*:*) |
3974 | + echo hppa1.1-hitachi-hiuxmpp |
3975 | + exit 0;; |
3976 | + Pyramid*:OSx*:*:* | MIS*:OSx*:*:* | MIS*:SMP_DC-OSx*:*:*) |
3977 | + # akee@wpdis03.wpafb.af.mil (Earle F. Ake) contributed MIS and NILE. |
3978 | + if test "`(/bin/universe) 2>/dev/null`" = att ; then |
3979 | + echo pyramid-pyramid-sysv3 |
3980 | + else |
3981 | + echo pyramid-pyramid-bsd |
3982 | + fi |
3983 | + exit 0 ;; |
3984 | + NILE*:*:*:dcosx) |
3985 | + echo pyramid-pyramid-svr4 |
3986 | + exit 0 ;; |
3987 | + DRS?6000:unix:4.0:6*) |
3988 | + echo sparc-icl-nx6 |
3989 | + exit 0 ;; |
3990 | + DRS?6000:UNIX_SV:4.2*:7* | DRS?6000:isis:4.2*:7*) |
3991 | + case `/usr/bin/uname -p` in |
3992 | + sparc) echo sparc-icl-nx7 && exit 0 ;; |
3993 | + esac ;; |
3994 | + sun4H:SunOS:5.*:*) |
3995 | + echo sparc-hal-solaris2`echo ${UNAME_RELEASE}|sed -e 's/[^.]*//'` |
3996 | + exit 0 ;; |
3997 | + sun4*:SunOS:5.*:* | tadpole*:SunOS:5.*:*) |
3998 | + echo sparc-sun-solaris2`echo ${UNAME_RELEASE}|sed -e 's/[^.]*//'` |
3999 | + exit 0 ;; |
4000 | + i86pc:SunOS:5.*:*) |
4001 | + echo i386-pc-solaris2`echo ${UNAME_RELEASE}|sed -e 's/[^.]*//'` |
4002 | + exit 0 ;; |
4003 | + sun4*:SunOS:6*:*) |
4004 | + # According to config.sub, this is the proper way to canonicalize |
4005 | + # SunOS6. Hard to guess exactly what SunOS6 will be like, but |
4006 | + # it's likely to be more like Solaris than SunOS4. |
4007 | + echo sparc-sun-solaris3`echo ${UNAME_RELEASE}|sed -e 's/[^.]*//'` |
4008 | + exit 0 ;; |
4009 | + sun4*:SunOS:*:*) |
4010 | + case "`/usr/bin/arch -k`" in |
4011 | + Series*|S4*) |
4012 | + UNAME_RELEASE=`uname -v` |
4013 | + ;; |
4014 | + esac |
4015 | + # Japanese Language versions have a version number like `4.1.3-JL'. |
4016 | + echo sparc-sun-sunos`echo ${UNAME_RELEASE}|sed -e 's/-/_/'` |
4017 | + exit 0 ;; |
4018 | + sun3*:SunOS:*:*) |
4019 | + echo m68k-sun-sunos${UNAME_RELEASE} |
4020 | + exit 0 ;; |
4021 | + sun*:*:4.2BSD:*) |
4022 | + UNAME_RELEASE=`(sed 1q /etc/motd | awk '{print substr($5,1,3)}') 2>/dev/null` |
4023 | + test "x${UNAME_RELEASE}" = "x" && UNAME_RELEASE=3 |
4024 | + case "`/bin/arch`" in |
4025 | + sun3) |
4026 | + echo m68k-sun-sunos${UNAME_RELEASE} |
4027 | + ;; |
4028 | + sun4) |
4029 | + echo sparc-sun-sunos${UNAME_RELEASE} |
4030 | + ;; |
4031 | + esac |
4032 | + exit 0 ;; |
4033 | + aushp:SunOS:*:*) |
4034 | + echo sparc-auspex-sunos${UNAME_RELEASE} |
4035 | + exit 0 ;; |
4036 | + # The situation for MiNT is a little confusing. The machine name |
4037 | + # can be virtually everything (everything which is not |
4038 | + # "atarist" or "atariste" at least should have a processor |
4039 | + # > m68000). The system name ranges from "MiNT" over "FreeMiNT" |
4040 | + # to the lowercase version "mint" (or "freemint"). Finally |
4041 | + # the system name "TOS" denotes a system which is actually not |
4042 | + # MiNT. But MiNT is downward compatible to TOS, so this should |
4043 | + # be no problem. |
4044 | + atarist[e]:*MiNT:*:* | atarist[e]:*mint:*:* | atarist[e]:*TOS:*:*) |
4045 | + echo m68k-atari-mint${UNAME_RELEASE} |
4046 | + exit 0 ;; |
4047 | + atari*:*MiNT:*:* | atari*:*mint:*:* | atarist[e]:*TOS:*:*) |
4048 | + echo m68k-atari-mint${UNAME_RELEASE} |
4049 | + exit 0 ;; |
4050 | + *falcon*:*MiNT:*:* | *falcon*:*mint:*:* | *falcon*:*TOS:*:*) |
4051 | + echo m68k-atari-mint${UNAME_RELEASE} |
4052 | + exit 0 ;; |
4053 | + milan*:*MiNT:*:* | milan*:*mint:*:* | *milan*:*TOS:*:*) |
4054 | + echo m68k-milan-mint${UNAME_RELEASE} |
4055 | + exit 0 ;; |
4056 | + hades*:*MiNT:*:* | hades*:*mint:*:* | *hades*:*TOS:*:*) |
4057 | + echo m68k-hades-mint${UNAME_RELEASE} |
4058 | + exit 0 ;; |
4059 | + *:*MiNT:*:* | *:*mint:*:* | *:*TOS:*:*) |
4060 | + echo m68k-unknown-mint${UNAME_RELEASE} |
4061 | + exit 0 ;; |
4062 | + m68k:machten:*:*) |
4063 | + echo m68k-apple-machten${UNAME_RELEASE} |
4064 | + exit 0 ;; |
4065 | + powerpc:machten:*:*) |
4066 | + echo powerpc-apple-machten${UNAME_RELEASE} |
4067 | + exit 0 ;; |
4068 | + RISC*:Mach:*:*) |
4069 | + echo mips-dec-mach_bsd4.3 |
4070 | + exit 0 ;; |
4071 | + RISC*:ULTRIX:*:*) |
4072 | + echo mips-dec-ultrix${UNAME_RELEASE} |
4073 | + exit 0 ;; |
4074 | + VAX*:ULTRIX*:*:*) |
4075 | + echo vax-dec-ultrix${UNAME_RELEASE} |
4076 | + exit 0 ;; |
4077 | + 2020:CLIX:*:* | 2430:CLIX:*:*) |
4078 | + echo clipper-intergraph-clix${UNAME_RELEASE} |
4079 | + exit 0 ;; |
4080 | + mips:*:*:UMIPS | mips:*:*:RISCos) |
4081 | + eval $set_cc_for_build |
4082 | + sed 's/^ //' << EOF >$dummy.c |
4083 | +#ifdef __cplusplus |
4084 | +#include <stdio.h> /* for printf() prototype */ |
4085 | + int main (int argc, char *argv[]) { |
4086 | +#else |
4087 | + int main (argc, argv) int argc; char *argv[]; { |
4088 | +#endif |
4089 | + #if defined (host_mips) && defined (MIPSEB) |
4090 | + #if defined (SYSTYPE_SYSV) |
4091 | + printf ("mips-mips-riscos%ssysv\n", argv[1]); exit (0); |
4092 | + #endif |
4093 | + #if defined (SYSTYPE_SVR4) |
4094 | + printf ("mips-mips-riscos%ssvr4\n", argv[1]); exit (0); |
4095 | + #endif |
4096 | + #if defined (SYSTYPE_BSD43) || defined(SYSTYPE_BSD) |
4097 | + printf ("mips-mips-riscos%sbsd\n", argv[1]); exit (0); |
4098 | + #endif |
4099 | + #endif |
4100 | + exit (-1); |
4101 | + } |
4102 | +EOF |
4103 | + $CC_FOR_BUILD -o $dummy $dummy.c \ |
4104 | + && $dummy `echo "${UNAME_RELEASE}" | sed -n 's/\([0-9]*\).*/\1/p'` \ |
4105 | + && exit 0 |
4106 | + echo mips-mips-riscos${UNAME_RELEASE} |
4107 | + exit 0 ;; |
4108 | + Motorola:PowerMAX_OS:*:*) |
4109 | + echo powerpc-motorola-powermax |
4110 | + exit 0 ;; |
4111 | + Motorola:*:4.3:PL8-*) |
4112 | + echo powerpc-harris-powermax |
4113 | + exit 0 ;; |
4114 | + Night_Hawk:*:*:PowerMAX_OS | Synergy:PowerMAX_OS:*:*) |
4115 | + echo powerpc-harris-powermax |
4116 | + exit 0 ;; |
4117 | + Night_Hawk:Power_UNIX:*:*) |
4118 | + echo powerpc-harris-powerunix |
4119 | + exit 0 ;; |
4120 | + m88k:CX/UX:7*:*) |
4121 | + echo m88k-harris-cxux7 |
4122 | + exit 0 ;; |
4123 | + m88k:*:4*:R4*) |
4124 | + echo m88k-motorola-sysv4 |
4125 | + exit 0 ;; |
4126 | + m88k:*:3*:R3*) |
4127 | + echo m88k-motorola-sysv3 |
4128 | + exit 0 ;; |
4129 | + AViiON:dgux:*:*) |
4130 | + # DG/UX returns AViiON for all architectures |
4131 | + UNAME_PROCESSOR=`/usr/bin/uname -p` |
4132 | + if [ $UNAME_PROCESSOR = mc88100 ] || [ $UNAME_PROCESSOR = mc88110 ] |
4133 | + then |
4134 | + if [ ${TARGET_BINARY_INTERFACE}x = m88kdguxelfx ] || \ |
4135 | + [ ${TARGET_BINARY_INTERFACE}x = x ] |
4136 | + then |
4137 | + echo m88k-dg-dgux${UNAME_RELEASE} |
4138 | + else |
4139 | + echo m88k-dg-dguxbcs${UNAME_RELEASE} |
4140 | + fi |
4141 | + else |
4142 | + echo i586-dg-dgux${UNAME_RELEASE} |
4143 | + fi |
4144 | + exit 0 ;; |
4145 | + M88*:DolphinOS:*:*) # DolphinOS (SVR3) |
4146 | + echo m88k-dolphin-sysv3 |
4147 | + exit 0 ;; |
4148 | + M88*:*:R3*:*) |
4149 | + # Delta 88k system running SVR3 |
4150 | + echo m88k-motorola-sysv3 |
4151 | + exit 0 ;; |
4152 | + XD88*:*:*:*) # Tektronix XD88 system running UTekV (SVR3) |
4153 | + echo m88k-tektronix-sysv3 |
4154 | + exit 0 ;; |
4155 | + Tek43[0-9][0-9]:UTek:*:*) # Tektronix 4300 system running UTek (BSD) |
4156 | + echo m68k-tektronix-bsd |
4157 | + exit 0 ;; |
4158 | + *:IRIX*:*:*) |
4159 | + echo mips-sgi-irix`echo ${UNAME_RELEASE}|sed -e 's/-/_/g'` |
4160 | + exit 0 ;; |
4161 | + ????????:AIX?:[12].1:2) # AIX 2.2.1 or AIX 2.1.1 is RT/PC AIX. |
4162 | + echo romp-ibm-aix # uname -m gives an 8 hex-code CPU id |
4163 | + exit 0 ;; # Note that: echo "'`uname -s`'" gives 'AIX ' |
4164 | + i*86:AIX:*:*) |
4165 | + echo i386-ibm-aix |
4166 | + exit 0 ;; |
4167 | + ia64:AIX:*:*) |
4168 | + if [ -x /usr/bin/oslevel ] ; then |
4169 | + IBM_REV=`/usr/bin/oslevel` |
4170 | + else |
4171 | + IBM_REV=${UNAME_VERSION}.${UNAME_RELEASE} |
4172 | + fi |
4173 | + echo ${UNAME_MACHINE}-ibm-aix${IBM_REV} |
4174 | + exit 0 ;; |
4175 | + *:AIX:2:3) |
4176 | + if grep bos325 /usr/include/stdio.h >/dev/null 2>&1; then |
4177 | + eval $set_cc_for_build |
4178 | + sed 's/^ //' << EOF >$dummy.c |
4179 | + #include <sys/systemcfg.h> |
4180 | + |
4181 | + main() |
4182 | + { |
4183 | + if (!__power_pc()) |
4184 | + exit(1); |
4185 | + puts("powerpc-ibm-aix3.2.5"); |
4186 | + exit(0); |
4187 | + } |
4188 | +EOF |
4189 | + $CC_FOR_BUILD -o $dummy $dummy.c && $dummy && exit 0 |
4190 | + echo rs6000-ibm-aix3.2.5 |
4191 | + elif grep bos324 /usr/include/stdio.h >/dev/null 2>&1; then |
4192 | + echo rs6000-ibm-aix3.2.4 |
4193 | + else |
4194 | + echo rs6000-ibm-aix3.2 |
4195 | + fi |
4196 | + exit 0 ;; |
4197 | + *:AIX:*:[45]) |
4198 | + IBM_CPU_ID=`/usr/sbin/lsdev -C -c processor -S available | sed 1q | awk '{ print $1 }'` |
4199 | + if /usr/sbin/lsattr -El ${IBM_CPU_ID} | grep ' POWER' >/dev/null 2>&1; then |
4200 | + IBM_ARCH=rs6000 |
4201 | + else |
4202 | + IBM_ARCH=powerpc |
4203 | + fi |
4204 | + if [ -x /usr/bin/oslevel ] ; then |
4205 | + IBM_REV=`/usr/bin/oslevel` |
4206 | + else |
4207 | + IBM_REV=${UNAME_VERSION}.${UNAME_RELEASE} |
4208 | + fi |
4209 | + echo ${IBM_ARCH}-ibm-aix${IBM_REV} |
4210 | + exit 0 ;; |
4211 | + *:AIX:*:*) |
4212 | + echo rs6000-ibm-aix |
4213 | + exit 0 ;; |
4214 | + ibmrt:4.4BSD:*|romp-ibm:BSD:*) |
4215 | + echo romp-ibm-bsd4.4 |
4216 | + exit 0 ;; |
4217 | + ibmrt:*BSD:*|romp-ibm:BSD:*) # covers RT/PC BSD and |
4218 | + echo romp-ibm-bsd${UNAME_RELEASE} # 4.3 with uname added to |
4219 | + exit 0 ;; # report: romp-ibm BSD 4.3 |
4220 | + *:BOSX:*:*) |
4221 | + echo rs6000-bull-bosx |
4222 | + exit 0 ;; |
4223 | + DPX/2?00:B.O.S.:*:*) |
4224 | + echo m68k-bull-sysv3 |
4225 | + exit 0 ;; |
4226 | + 9000/[34]??:4.3bsd:1.*:*) |
4227 | + echo m68k-hp-bsd |
4228 | + exit 0 ;; |
4229 | + hp300:4.4BSD:*:* | 9000/[34]??:4.3bsd:2.*:*) |
4230 | + echo m68k-hp-bsd4.4 |
4231 | + exit 0 ;; |
4232 | + 9000/[34678]??:HP-UX:*:*) |
4233 | + HPUX_REV=`echo ${UNAME_RELEASE}|sed -e 's/[^.]*.[0B]*//'` |
4234 | + case "${UNAME_MACHINE}" in |
4235 | + 9000/31? ) HP_ARCH=m68000 ;; |
4236 | + 9000/[34]?? ) HP_ARCH=m68k ;; |
4237 | + 9000/[678][0-9][0-9]) |
4238 | + if [ -x /usr/bin/getconf ]; then |
4239 | + sc_cpu_version=`/usr/bin/getconf SC_CPU_VERSION 2>/dev/null` |
4240 | + sc_kernel_bits=`/usr/bin/getconf SC_KERNEL_BITS 2>/dev/null` |
4241 | + case "${sc_cpu_version}" in |
4242 | + 523) HP_ARCH="hppa1.0" ;; # CPU_PA_RISC1_0 |
4243 | + 528) HP_ARCH="hppa1.1" ;; # CPU_PA_RISC1_1 |
4244 | + 532) # CPU_PA_RISC2_0 |
4245 | + case "${sc_kernel_bits}" in |
4246 | + 32) HP_ARCH="hppa2.0n" ;; |
4247 | + 64) HP_ARCH="hppa2.0w" ;; |
4248 | + '') HP_ARCH="hppa2.0" ;; # HP-UX 10.20 |
4249 | + esac ;; |
4250 | + esac |
4251 | + fi |
4252 | + if [ "${HP_ARCH}" = "" ]; then |
4253 | + eval $set_cc_for_build |
4254 | + sed 's/^ //' << EOF >$dummy.c |
4255 | + |
4256 | + #define _HPUX_SOURCE |
4257 | + #include <stdlib.h> |
4258 | + #include <unistd.h> |
4259 | + |
4260 | + int main () |
4261 | + { |
4262 | + #if defined(_SC_KERNEL_BITS) |
4263 | + long bits = sysconf(_SC_KERNEL_BITS); |
4264 | + #endif |
4265 | + long cpu = sysconf (_SC_CPU_VERSION); |
4266 | + |
4267 | + switch (cpu) |
4268 | + { |
4269 | + case CPU_PA_RISC1_0: puts ("hppa1.0"); break; |
4270 | + case CPU_PA_RISC1_1: puts ("hppa1.1"); break; |
4271 | + case CPU_PA_RISC2_0: |
4272 | + #if defined(_SC_KERNEL_BITS) |
4273 | + switch (bits) |
4274 | + { |
4275 | + case 64: puts ("hppa2.0w"); break; |
4276 | + case 32: puts ("hppa2.0n"); break; |
4277 | + default: puts ("hppa2.0"); break; |
4278 | + } break; |
4279 | + #else /* !defined(_SC_KERNEL_BITS) */ |
4280 | + puts ("hppa2.0"); break; |
4281 | + #endif |
4282 | + default: puts ("hppa1.0"); break; |
4283 | + } |
4284 | + exit (0); |
4285 | + } |
4286 | +EOF |
4287 | + (CCOPTS= $CC_FOR_BUILD -o $dummy $dummy.c 2>/dev/null) && HP_ARCH=`$dummy` |
4288 | + test -z "$HP_ARCH" && HP_ARCH=hppa |
4289 | + fi ;; |
4290 | + esac |
4291 | + if [ ${HP_ARCH} = "hppa2.0w" ] |
4292 | + then |
4293 | + # avoid double evaluation of $set_cc_for_build |
4294 | + test -n "$CC_FOR_BUILD" || eval $set_cc_for_build |
4295 | + if echo __LP64__ | (CCOPTS= $CC_FOR_BUILD -E -) | grep __LP64__ >/dev/null |
4296 | + then |
4297 | + HP_ARCH="hppa2.0w" |
4298 | + else |
4299 | + HP_ARCH="hppa64" |
4300 | + fi |
4301 | + fi |
4302 | + echo ${HP_ARCH}-hp-hpux${HPUX_REV} |
4303 | + exit 0 ;; |
4304 | + ia64:HP-UX:*:*) |
4305 | + HPUX_REV=`echo ${UNAME_RELEASE}|sed -e 's/[^.]*.[0B]*//'` |
4306 | + echo ia64-hp-hpux${HPUX_REV} |
4307 | + exit 0 ;; |
4308 | + 3050*:HI-UX:*:*) |
4309 | + eval $set_cc_for_build |
4310 | + sed 's/^ //' << EOF >$dummy.c |
4311 | + #include <unistd.h> |
4312 | + int |
4313 | + main () |
4314 | + { |
4315 | + long cpu = sysconf (_SC_CPU_VERSION); |
4316 | + /* The order matters, because CPU_IS_HP_MC68K erroneously returns |
4317 | + true for CPU_PA_RISC1_0. CPU_IS_PA_RISC returns correct |
4318 | + results, however. */ |
4319 | + if (CPU_IS_PA_RISC (cpu)) |
4320 | + { |
4321 | + switch (cpu) |
4322 | + { |
4323 | + case CPU_PA_RISC1_0: puts ("hppa1.0-hitachi-hiuxwe2"); break; |
4324 | + case CPU_PA_RISC1_1: puts ("hppa1.1-hitachi-hiuxwe2"); break; |
4325 | + case CPU_PA_RISC2_0: puts ("hppa2.0-hitachi-hiuxwe2"); break; |
4326 | + default: puts ("hppa-hitachi-hiuxwe2"); break; |
4327 | + } |
4328 | + } |
4329 | + else if (CPU_IS_HP_MC68K (cpu)) |
4330 | + puts ("m68k-hitachi-hiuxwe2"); |
4331 | + else puts ("unknown-hitachi-hiuxwe2"); |
4332 | + exit (0); |
4333 | + } |
4334 | +EOF |
4335 | + $CC_FOR_BUILD -o $dummy $dummy.c && $dummy && exit 0 |
4336 | + echo unknown-hitachi-hiuxwe2 |
4337 | + exit 0 ;; |
4338 | + 9000/7??:4.3bsd:*:* | 9000/8?[79]:4.3bsd:*:* ) |
4339 | + echo hppa1.1-hp-bsd |
4340 | + exit 0 ;; |
4341 | + 9000/8??:4.3bsd:*:*) |
4342 | + echo hppa1.0-hp-bsd |
4343 | + exit 0 ;; |
4344 | + *9??*:MPE/iX:*:* | *3000*:MPE/iX:*:*) |
4345 | + echo hppa1.0-hp-mpeix |
4346 | + exit 0 ;; |
4347 | + hp7??:OSF1:*:* | hp8?[79]:OSF1:*:* ) |
4348 | + echo hppa1.1-hp-osf |
4349 | + exit 0 ;; |
4350 | + hp8??:OSF1:*:*) |
4351 | + echo hppa1.0-hp-osf |
4352 | + exit 0 ;; |
4353 | + i*86:OSF1:*:*) |
4354 | + if [ -x /usr/sbin/sysversion ] ; then |
4355 | + echo ${UNAME_MACHINE}-unknown-osf1mk |
4356 | + else |
4357 | + echo ${UNAME_MACHINE}-unknown-osf1 |
4358 | + fi |
4359 | + exit 0 ;; |
4360 | + parisc*:Lites*:*:*) |
4361 | + echo hppa1.1-hp-lites |
4362 | + exit 0 ;; |
4363 | + C1*:ConvexOS:*:* | convex:ConvexOS:C1*:*) |
4364 | + echo c1-convex-bsd |
4365 | + exit 0 ;; |
4366 | + C2*:ConvexOS:*:* | convex:ConvexOS:C2*:*) |
4367 | + if getsysinfo -f scalar_acc |
4368 | + then echo c32-convex-bsd |
4369 | + else echo c2-convex-bsd |
4370 | + fi |
4371 | + exit 0 ;; |
4372 | + C34*:ConvexOS:*:* | convex:ConvexOS:C34*:*) |
4373 | + echo c34-convex-bsd |
4374 | + exit 0 ;; |
4375 | + C38*:ConvexOS:*:* | convex:ConvexOS:C38*:*) |
4376 | + echo c38-convex-bsd |
4377 | + exit 0 ;; |
4378 | + C4*:ConvexOS:*:* | convex:ConvexOS:C4*:*) |
4379 | + echo c4-convex-bsd |
4380 | + exit 0 ;; |
4381 | + CRAY*Y-MP:*:*:*) |
4382 | + echo ymp-cray-unicos${UNAME_RELEASE} | sed -e 's/\.[^.]*$/.X/' |
4383 | + exit 0 ;; |
4384 | + CRAY*[A-Z]90:*:*:*) |
4385 | + echo ${UNAME_MACHINE}-cray-unicos${UNAME_RELEASE} \ |
4386 | + | sed -e 's/CRAY.*\([A-Z]90\)/\1/' \ |
4387 | + -e y/ABCDEFGHIJKLMNOPQRSTUVWXYZ/abcdefghijklmnopqrstuvwxyz/ \ |
4388 | + -e 's/\.[^.]*$/.X/' |
4389 | + exit 0 ;; |
4390 | + CRAY*TS:*:*:*) |
4391 | + echo t90-cray-unicos${UNAME_RELEASE} | sed -e 's/\.[^.]*$/.X/' |
4392 | + exit 0 ;; |
4393 | + CRAY*T3E:*:*:*) |
4394 | + echo alphaev5-cray-unicosmk${UNAME_RELEASE} | sed -e 's/\.[^.]*$/.X/' |
4395 | + exit 0 ;; |
4396 | + CRAY*SV1:*:*:*) |
4397 | + echo sv1-cray-unicos${UNAME_RELEASE} | sed -e 's/\.[^.]*$/.X/' |
4398 | + exit 0 ;; |
4399 | + *:UNICOS/mp:*:*) |
4400 | + echo craynv-cray-unicosmp${UNAME_RELEASE} | sed -e 's/\.[^.]*$/.X/' |
4401 | + exit 0 ;; |
4402 | + F30[01]:UNIX_System_V:*:* | F700:UNIX_System_V:*:*) |
4403 | + FUJITSU_PROC=`uname -m | tr 'ABCDEFGHIJKLMNOPQRSTUVWXYZ' 'abcdefghijklmnopqrstuvwxyz'` |
4404 | + FUJITSU_SYS=`uname -p | tr 'ABCDEFGHIJKLMNOPQRSTUVWXYZ' 'abcdefghijklmnopqrstuvwxyz' | sed -e 's/\///'` |
4405 | + FUJITSU_REL=`echo ${UNAME_RELEASE} | sed -e 's/ /_/'` |
4406 | + echo "${FUJITSU_PROC}-fujitsu-${FUJITSU_SYS}${FUJITSU_REL}" |
4407 | + exit 0 ;; |
4408 | + 5000:UNIX_System_V:4.*:*) |
4409 | + FUJITSU_SYS=`uname -p | tr 'ABCDEFGHIJKLMNOPQRSTUVWXYZ' 'abcdefghijklmnopqrstuvwxyz' | sed -e 's/\///'` |
4410 | + FUJITSU_REL=`echo ${UNAME_RELEASE} | tr 'ABCDEFGHIJKLMNOPQRSTUVWXYZ' 'abcdefghijklmnopqrstuvwxyz' | sed -e 's/ /_/'` |
4411 | + echo "sparc-fujitsu-${FUJITSU_SYS}${FUJITSU_REL}" |
4412 | + exit 0 ;; |
4413 | + i*86:BSD/386:*:* | i*86:BSD/OS:*:* | *:Ascend\ Embedded/OS:*:*) |
4414 | + echo ${UNAME_MACHINE}-pc-bsdi${UNAME_RELEASE} |
4415 | + exit 0 ;; |
4416 | + sparc*:BSD/OS:*:*) |
4417 | + echo sparc-unknown-bsdi${UNAME_RELEASE} |
4418 | + exit 0 ;; |
4419 | + *:BSD/OS:*:*) |
4420 | + echo ${UNAME_MACHINE}-unknown-bsdi${UNAME_RELEASE} |
4421 | + exit 0 ;; |
4422 | + *:FreeBSD:*:*) |
4423 | + echo ${UNAME_MACHINE}-unknown-freebsd`echo ${UNAME_RELEASE}|sed -e 's/[-(].*//'` |
4424 | + exit 0 ;; |
4425 | + i*:CYGWIN*:*) |
4426 | + echo ${UNAME_MACHINE}-pc-cygwin |
4427 | + exit 0 ;; |
4428 | + i*:MINGW*:*) |
4429 | + echo ${UNAME_MACHINE}-pc-mingw32 |
4430 | + exit 0 ;; |
4431 | + i*:PW*:*) |
4432 | + echo ${UNAME_MACHINE}-pc-pw32 |
4433 | + exit 0 ;; |
4434 | + x86:Interix*:[34]*) |
4435 | + echo i586-pc-interix${UNAME_RELEASE}|sed -e 's/\..*//' |
4436 | + exit 0 ;; |
4437 | + [345]86:Windows_95:* | [345]86:Windows_98:* | [345]86:Windows_NT:*) |
4438 | + echo i${UNAME_MACHINE}-pc-mks |
4439 | + exit 0 ;; |
4440 | + i*:Windows_NT*:* | Pentium*:Windows_NT*:*) |
4441 | + # How do we know it's Interix rather than the generic POSIX subsystem? |
4442 | + # It also conflicts with pre-2.0 versions of AT&T UWIN. Should we |
4443 | + # UNAME_MACHINE based on the output of uname instead of i386? |
4444 | + echo i586-pc-interix |
4445 | + exit 0 ;; |
4446 | + i*:UWIN*:*) |
4447 | + echo ${UNAME_MACHINE}-pc-uwin |
4448 | + exit 0 ;; |
4449 | + amd64:CYGWIN*:*:*) |
4450 | + echo x86_64-unknown-cygwin |
4451 | + exit 0 ;; |
4452 | + p*:CYGWIN*:*) |
4453 | + echo powerpcle-unknown-cygwin |
4454 | + exit 0 ;; |
4455 | + prep*:SunOS:5.*:*) |
4456 | + echo powerpcle-unknown-solaris2`echo ${UNAME_RELEASE}|sed -e 's/[^.]*//'` |
4457 | + exit 0 ;; |
4458 | + *:GNU:*:*) |
4459 | + # the GNU system |
4460 | + echo `echo ${UNAME_MACHINE}|sed -e 's,[-/].*$,,'`-unknown-gnu`echo ${UNAME_RELEASE}|sed -e 's,/.*$,,'` |
4461 | + exit 0 ;; |
4462 | + *:GNU/*:*:*) |
4463 | + # other systems with GNU libc and userland |
4464 | + echo ${UNAME_MACHINE}-unknown-`echo ${UNAME_SYSTEM} | sed 's,^[^/]*/,,' | tr '[A-Z]' '[a-z]'``echo ${UNAME_RELEASE}|sed -e 's/[-(].*//'`-gnu |
4465 | + exit 0 ;; |
4466 | + i*86:Minix:*:*) |
4467 | + echo ${UNAME_MACHINE}-pc-minix |
4468 | + exit 0 ;; |
4469 | + arm*:Linux:*:*) |
4470 | + echo ${UNAME_MACHINE}-unknown-linux-gnu |
4471 | + exit 0 ;; |
4472 | + cris:Linux:*:*) |
4473 | + echo cris-axis-linux-gnu |
4474 | + exit 0 ;; |
4475 | + crisv32:Linux:*:*) |
4476 | + echo crisv32-axis-linux-gnu |
4477 | + exit 0 ;; |
4478 | + frv:Linux:*:*) |
4479 | + echo frv-unknown-linux-gnu |
4480 | + exit 0 ;; |
4481 | + ia64:Linux:*:*) |
4482 | + echo ${UNAME_MACHINE}-unknown-linux-gnu |
4483 | + exit 0 ;; |
4484 | + m32r*:Linux:*:*) |
4485 | + echo ${UNAME_MACHINE}-unknown-linux-gnu |
4486 | + exit 0 ;; |
4487 | + m68*:Linux:*:*) |
4488 | + echo ${UNAME_MACHINE}-unknown-linux-gnu |
4489 | + exit 0 ;; |
4490 | + mips:Linux:*:*) |
4491 | + eval $set_cc_for_build |
4492 | + sed 's/^ //' << EOF >$dummy.c |
4493 | + #undef CPU |
4494 | + #undef mips |
4495 | + #undef mipsel |
4496 | + #if defined(__MIPSEL__) || defined(__MIPSEL) || defined(_MIPSEL) || defined(MIPSEL) |
4497 | + CPU=mipsel |
4498 | + #else |
4499 | + #if defined(__MIPSEB__) || defined(__MIPSEB) || defined(_MIPSEB) || defined(MIPSEB) |
4500 | + CPU=mips |
4501 | + #else |
4502 | + CPU= |
4503 | + #endif |
4504 | + #endif |
4505 | +EOF |
4506 | + eval `$CC_FOR_BUILD -E $dummy.c 2>/dev/null | grep ^CPU=` |
4507 | + test x"${CPU}" != x && echo "${CPU}-unknown-linux-gnu" && exit 0 |
4508 | + ;; |
4509 | + mips64:Linux:*:*) |
4510 | + eval $set_cc_for_build |
4511 | + sed 's/^ //' << EOF >$dummy.c |
4512 | + #undef CPU |
4513 | + #undef mips64 |
4514 | + #undef mips64el |
4515 | + #if defined(__MIPSEL__) || defined(__MIPSEL) || defined(_MIPSEL) || defined(MIPSEL) |
4516 | + CPU=mips64el |
4517 | + #else |
4518 | + #if defined(__MIPSEB__) || defined(__MIPSEB) || defined(_MIPSEB) || defined(MIPSEB) |
4519 | + CPU=mips64 |
4520 | + #else |
4521 | + CPU= |
4522 | + #endif |
4523 | + #endif |
4524 | +EOF |
4525 | + eval `$CC_FOR_BUILD -E $dummy.c 2>/dev/null | grep ^CPU=` |
4526 | + test x"${CPU}" != x && echo "${CPU}-unknown-linux-gnu" && exit 0 |
4527 | + ;; |
4528 | + ppc:Linux:*:*) |
4529 | + echo powerpc-unknown-linux-gnu |
4530 | + exit 0 ;; |
4531 | + ppc64:Linux:*:*) |
4532 | + echo powerpc64-unknown-linux-gnu |
4533 | + exit 0 ;; |
4534 | + alpha:Linux:*:*) |
4535 | + case `sed -n '/^cpu model/s/^.*: \(.*\)/\1/p' < /proc/cpuinfo` in |
4536 | + EV5) UNAME_MACHINE=alphaev5 ;; |
4537 | + EV56) UNAME_MACHINE=alphaev56 ;; |
4538 | + PCA56) UNAME_MACHINE=alphapca56 ;; |
4539 | + PCA57) UNAME_MACHINE=alphapca56 ;; |
4540 | + EV6) UNAME_MACHINE=alphaev6 ;; |
4541 | + EV67) UNAME_MACHINE=alphaev67 ;; |
4542 | + EV68*) UNAME_MACHINE=alphaev68 ;; |
4543 | + esac |
4544 | + objdump --private-headers /bin/sh | grep ld.so.1 >/dev/null |
4545 | + if test "$?" = 0 ; then LIBC="libc1" ; else LIBC="" ; fi |
4546 | + echo ${UNAME_MACHINE}-unknown-linux-gnu${LIBC} |
4547 | + exit 0 ;; |
4548 | + parisc:Linux:*:* | hppa:Linux:*:*) |
4549 | + # Look for CPU level |
4550 | + case `grep '^cpu[^a-z]*:' /proc/cpuinfo 2>/dev/null | cut -d' ' -f2` in |
4551 | + PA7*) echo hppa1.1-unknown-linux-gnu ;; |
4552 | + PA8*) echo hppa2.0-unknown-linux-gnu ;; |
4553 | + *) echo hppa-unknown-linux-gnu ;; |
4554 | + esac |
4555 | + exit 0 ;; |
4556 | + parisc64:Linux:*:* | hppa64:Linux:*:*) |
4557 | + echo hppa64-unknown-linux-gnu |
4558 | + exit 0 ;; |
4559 | + s390:Linux:*:* | s390x:Linux:*:*) |
4560 | + echo ${UNAME_MACHINE}-ibm-linux |
4561 | + exit 0 ;; |
4562 | + sh64*:Linux:*:*) |
4563 | + echo ${UNAME_MACHINE}-unknown-linux-gnu |
4564 | + exit 0 ;; |
4565 | + sh*:Linux:*:*) |
4566 | + echo ${UNAME_MACHINE}-unknown-linux-gnu |
4567 | + exit 0 ;; |
4568 | + sparc:Linux:*:* | sparc64:Linux:*:*) |
4569 | + echo ${UNAME_MACHINE}-unknown-linux-gnu |
4570 | + exit 0 ;; |
4571 | + x86_64:Linux:*:*) |
4572 | + echo x86_64-unknown-linux-gnu |
4573 | + exit 0 ;; |
4574 | + i*86:Linux:*:*) |
4575 | + # The BFD linker knows what the default object file format is, so |
4576 | + # first see if it will tell us. cd to the root directory to prevent |
4577 | + # problems with other programs or directories called `ld' in the path. |
4578 | + # Set LC_ALL=C to ensure ld outputs messages in English. |
4579 | + ld_supported_targets=`cd /; LC_ALL=C ld --help 2>&1 \ |
4580 | + | sed -ne '/supported targets:/!d |
4581 | + s/[ ][ ]*/ /g |
4582 | + s/.*supported targets: *// |
4583 | + s/ .*// |
4584 | + p'` |
4585 | + case "$ld_supported_targets" in |
4586 | + elf32-i386) |
4587 | + TENTATIVE="${UNAME_MACHINE}-pc-linux-gnu" |
4588 | + ;; |
4589 | + a.out-i386-linux) |
4590 | + echo "${UNAME_MACHINE}-pc-linux-gnuaout" |
4591 | + exit 0 ;; |
4592 | + coff-i386) |
4593 | + echo "${UNAME_MACHINE}-pc-linux-gnucoff" |
4594 | + exit 0 ;; |
4595 | + "") |
4596 | + # Either a pre-BFD a.out linker (linux-gnuoldld) or |
4597 | + # one that does not give us useful --help. |
4598 | + echo "${UNAME_MACHINE}-pc-linux-gnuoldld" |
4599 | + exit 0 ;; |
4600 | + esac |
4601 | + # Determine whether the default compiler is a.out or elf |
4602 | + eval $set_cc_for_build |
4603 | + sed 's/^ //' << EOF >$dummy.c |
4604 | + #include <features.h> |
4605 | + #ifdef __ELF__ |
4606 | + # ifdef __GLIBC__ |
4607 | + # if __GLIBC__ >= 2 |
4608 | + LIBC=gnu |
4609 | + # else |
4610 | + LIBC=gnulibc1 |
4611 | + # endif |
4612 | + # else |
4613 | + LIBC=gnulibc1 |
4614 | + # endif |
4615 | + #else |
4616 | + #ifdef __INTEL_COMPILER |
4617 | + LIBC=gnu |
4618 | + #else |
4619 | + LIBC=gnuaout |
4620 | + #endif |
4621 | + #endif |
4622 | + #ifdef __dietlibc__ |
4623 | + LIBC=dietlibc |
4624 | + #endif |
4625 | +EOF |
4626 | + eval `$CC_FOR_BUILD -E $dummy.c 2>/dev/null | grep ^LIBC=` |
4627 | + test x"${LIBC}" != x && echo "${UNAME_MACHINE}-pc-linux-${LIBC}" && exit 0 |
4628 | + test x"${TENTATIVE}" != x && echo "${TENTATIVE}" && exit 0 |
4629 | + ;; |
4630 | + i*86:DYNIX/ptx:4*:*) |
4631 | + # ptx 4.0 does uname -s correctly, with DYNIX/ptx in there. |
4632 | + # earlier versions are messed up and put the nodename in both |
4633 | + # sysname and nodename. |
4634 | + echo i386-sequent-sysv4 |
4635 | + exit 0 ;; |
4636 | + i*86:UNIX_SV:4.2MP:2.*) |
4637 | + # Unixware is an offshoot of SVR4, but it has its own version |
4638 | + # number series starting with 2... |
4639 | + # I am not positive that other SVR4 systems won't match this, |
4640 | + # I just have to hope. -- rms. |
4641 | + # Use sysv4.2uw... so that sysv4* matches it. |
4642 | + echo ${UNAME_MACHINE}-pc-sysv4.2uw${UNAME_VERSION} |
4643 | + exit 0 ;; |
4644 | + i*86:OS/2:*:*) |
4645 | + # If we were able to find `uname', then EMX Unix compatibility |
4646 | + # is probably installed. |
4647 | + echo ${UNAME_MACHINE}-pc-os2-emx |
4648 | + exit 0 ;; |
4649 | + i*86:XTS-300:*:STOP) |
4650 | + echo ${UNAME_MACHINE}-unknown-stop |
4651 | + exit 0 ;; |
4652 | + i*86:atheos:*:*) |
4653 | + echo ${UNAME_MACHINE}-unknown-atheos |
4654 | + exit 0 ;; |
4655 | + i*86:syllable:*:*) |
4656 | + echo ${UNAME_MACHINE}-pc-syllable |
4657 | + exit 0 ;; |
4658 | + i*86:LynxOS:2.*:* | i*86:LynxOS:3.[01]*:* | i*86:LynxOS:4.0*:*) |
4659 | + echo i386-unknown-lynxos${UNAME_RELEASE} |
4660 | + exit 0 ;; |
4661 | + i*86:*DOS:*:*) |
4662 | + echo ${UNAME_MACHINE}-pc-msdosdjgpp |
4663 | + exit 0 ;; |
4664 | + i*86:*:4.*:* | i*86:SYSTEM_V:4.*:*) |
4665 | + UNAME_REL=`echo ${UNAME_RELEASE} | sed 's/\/MP$//'` |
4666 | + if grep Novell /usr/include/link.h >/dev/null 2>/dev/null; then |
4667 | + echo ${UNAME_MACHINE}-univel-sysv${UNAME_REL} |
4668 | + else |
4669 | + echo ${UNAME_MACHINE}-pc-sysv${UNAME_REL} |
4670 | + fi |
4671 | + exit 0 ;; |
4672 | + i*86:*:5:[78]*) |
4673 | + case `/bin/uname -X | grep "^Machine"` in |
4674 | + *486*) UNAME_MACHINE=i486 ;; |
4675 | + *Pentium) UNAME_MACHINE=i586 ;; |
4676 | + *Pent*|*Celeron) UNAME_MACHINE=i686 ;; |
4677 | + esac |
4678 | + echo ${UNAME_MACHINE}-unknown-sysv${UNAME_RELEASE}${UNAME_SYSTEM}${UNAME_VERSION} |
4679 | + exit 0 ;; |
4680 | + i*86:*:3.2:*) |
4681 | + if test -f /usr/options/cb.name; then |
4682 | + UNAME_REL=`sed -n 's/.*Version //p' </usr/options/cb.name` |
4683 | + echo ${UNAME_MACHINE}-pc-isc$UNAME_REL |
4684 | + elif /bin/uname -X 2>/dev/null >/dev/null ; then |
4685 | + UNAME_REL=`(/bin/uname -X|grep Release|sed -e 's/.*= //')` |
4686 | + (/bin/uname -X|grep i80486 >/dev/null) && UNAME_MACHINE=i486 |
4687 | + (/bin/uname -X|grep '^Machine.*Pentium' >/dev/null) \ |
4688 | + && UNAME_MACHINE=i586 |
4689 | + (/bin/uname -X|grep '^Machine.*Pent *II' >/dev/null) \ |
4690 | + && UNAME_MACHINE=i686 |
4691 | + (/bin/uname -X|grep '^Machine.*Pentium Pro' >/dev/null) \ |
4692 | + && UNAME_MACHINE=i686 |
4693 | + echo ${UNAME_MACHINE}-pc-sco$UNAME_REL |
4694 | + else |
4695 | + echo ${UNAME_MACHINE}-pc-sysv32 |
4696 | + fi |
4697 | + exit 0 ;; |
4698 | + pc:*:*:*) |
4699 | + # Left here for compatibility: |
4700 | + # uname -m prints for DJGPP always 'pc', but it prints nothing about |
4701 | + # the processor, so we play safe by assuming i386. |
4702 | + echo i386-pc-msdosdjgpp |
4703 | + exit 0 ;; |
4704 | + Intel:Mach:3*:*) |
4705 | + echo i386-pc-mach3 |
4706 | + exit 0 ;; |
4707 | + paragon:*:*:*) |
4708 | + echo i860-intel-osf1 |
4709 | + exit 0 ;; |
4710 | + i860:*:4.*:*) # i860-SVR4 |
4711 | + if grep Stardent /usr/include/sys/uadmin.h >/dev/null 2>&1 ; then |
4712 | + echo i860-stardent-sysv${UNAME_RELEASE} # Stardent Vistra i860-SVR4 |
4713 | + else # Add other i860-SVR4 vendors below as they are discovered. |
4714 | + echo i860-unknown-sysv${UNAME_RELEASE} # Unknown i860-SVR4 |
4715 | + fi |
4716 | + exit 0 ;; |
4717 | + mini*:CTIX:SYS*5:*) |
4718 | + # "miniframe" |
4719 | + echo m68010-convergent-sysv |
4720 | + exit 0 ;; |
4721 | + mc68k:UNIX:SYSTEM5:3.51m) |
4722 | + echo m68k-convergent-sysv |
4723 | + exit 0 ;; |
4724 | + M680?0:D-NIX:5.3:*) |
4725 | + echo m68k-diab-dnix |
4726 | + exit 0 ;; |
4727 | + M68*:*:R3V[5678]*:*) |
4728 | + test -r /sysV68 && echo 'm68k-motorola-sysv' && exit 0 ;; |
4729 | + 3[345]??:*:4.0:3.0 | 3[34]??A:*:4.0:3.0 | 3[34]??,*:*:4.0:3.0 | 3[34]??/*:*:4.0:3.0 | 4400:*:4.0:3.0 | 4850:*:4.0:3.0 | SKA40:*:4.0:3.0 | SDS2:*:4.0:3.0 | SHG2:*:4.0:3.0 | S7501*:*:4.0:3.0) |
4730 | + OS_REL='' |
4731 | + test -r /etc/.relid \ |
4732 | + && OS_REL=.`sed -n 's/[^ ]* [^ ]* \([0-9][0-9]\).*/\1/p' < /etc/.relid` |
4733 | + /bin/uname -p 2>/dev/null | grep 86 >/dev/null \ |
4734 | + && echo i486-ncr-sysv4.3${OS_REL} && exit 0 |
4735 | + /bin/uname -p 2>/dev/null | /bin/grep entium >/dev/null \ |
4736 | + && echo i586-ncr-sysv4.3${OS_REL} && exit 0 ;; |
4737 | + 3[34]??:*:4.0:* | 3[34]??,*:*:4.0:*) |
4738 | + /bin/uname -p 2>/dev/null | grep 86 >/dev/null \ |
4739 | + && echo i486-ncr-sysv4 && exit 0 ;; |
4740 | + m68*:LynxOS:2.*:* | m68*:LynxOS:3.0*:*) |
4741 | + echo m68k-unknown-lynxos${UNAME_RELEASE} |
4742 | + exit 0 ;; |
4743 | + mc68030:UNIX_System_V:4.*:*) |
4744 | + echo m68k-atari-sysv4 |
4745 | + exit 0 ;; |
4746 | + TSUNAMI:LynxOS:2.*:*) |
4747 | + echo sparc-unknown-lynxos${UNAME_RELEASE} |
4748 | + exit 0 ;; |
4749 | + rs6000:LynxOS:2.*:*) |
4750 | + echo rs6000-unknown-lynxos${UNAME_RELEASE} |
4751 | + exit 0 ;; |
4752 | + PowerPC:LynxOS:2.*:* | PowerPC:LynxOS:3.[01]*:* | PowerPC:LynxOS:4.0*:*) |
4753 | + echo powerpc-unknown-lynxos${UNAME_RELEASE} |
4754 | + exit 0 ;; |
4755 | + SM[BE]S:UNIX_SV:*:*) |
4756 | + echo mips-dde-sysv${UNAME_RELEASE} |
4757 | + exit 0 ;; |
4758 | + RM*:ReliantUNIX-*:*:*) |
4759 | + echo mips-sni-sysv4 |
4760 | + exit 0 ;; |
4761 | + RM*:SINIX-*:*:*) |
4762 | + echo mips-sni-sysv4 |
4763 | + exit 0 ;; |
4764 | + *:SINIX-*:*:*) |
4765 | + if uname -p 2>/dev/null >/dev/null ; then |
4766 | + UNAME_MACHINE=`(uname -p) 2>/dev/null` |
4767 | + echo ${UNAME_MACHINE}-sni-sysv4 |
4768 | + else |
4769 | + echo ns32k-sni-sysv |
4770 | + fi |
4771 | + exit 0 ;; |
4772 | + PENTIUM:*:4.0*:*) # Unisys `ClearPath HMP IX 4000' SVR4/MP effort |
4773 | + # says <Richard.M.Bartel@ccMail.Census.GOV> |
4774 | + echo i586-unisys-sysv4 |
4775 | + exit 0 ;; |
4776 | + *:UNIX_System_V:4*:FTX*) |
4777 | + # From Gerald Hewes <hewes@openmarket.com>. |
4778 | + # How about differentiating between stratus architectures? -djm |
4779 | + echo hppa1.1-stratus-sysv4 |
4780 | + exit 0 ;; |
4781 | + *:*:*:FTX*) |
4782 | + # From seanf@swdc.stratus.com. |
4783 | + echo i860-stratus-sysv4 |
4784 | + exit 0 ;; |
4785 | + i*86:VOS:*:*) |
4786 | + # From Paul.Green@stratus.com. |
4787 | + echo ${UNAME_MACHINE}-stratus-vos |
4788 | + exit 0 ;; |
4789 | + *:VOS:*:*) |
4790 | + # From Paul.Green@stratus.com. |
4791 | + echo hppa1.1-stratus-vos |
4792 | + exit 0 ;; |
4793 | + mc68*:A/UX:*:*) |
4794 | + echo m68k-apple-aux${UNAME_RELEASE} |
4795 | + exit 0 ;; |
4796 | + news*:NEWS-OS:6*:*) |
4797 | + echo mips-sony-newsos6 |
4798 | + exit 0 ;; |
4799 | + R[34]000:*System_V*:*:* | R4000:UNIX_SYSV:*:* | R*000:UNIX_SV:*:*) |
4800 | + if [ -d /usr/nec ]; then |
4801 | + echo mips-nec-sysv${UNAME_RELEASE} |
4802 | + else |
4803 | + echo mips-unknown-sysv${UNAME_RELEASE} |
4804 | + fi |
4805 | + exit 0 ;; |
4806 | + BeBox:BeOS:*:*) # BeOS running on hardware made by Be, PPC only. |
4807 | + echo powerpc-be-beos |
4808 | + exit 0 ;; |
4809 | + BeMac:BeOS:*:*) # BeOS running on Mac or Mac clone, PPC only. |
4810 | + echo powerpc-apple-beos |
4811 | + exit 0 ;; |
4812 | + BePC:BeOS:*:*) # BeOS running on Intel PC compatible. |
4813 | + echo i586-pc-beos |
4814 | + exit 0 ;; |
4815 | + SX-4:SUPER-UX:*:*) |
4816 | + echo sx4-nec-superux${UNAME_RELEASE} |
4817 | + exit 0 ;; |
4818 | + SX-5:SUPER-UX:*:*) |
4819 | + echo sx5-nec-superux${UNAME_RELEASE} |
4820 | + exit 0 ;; |
4821 | + SX-6:SUPER-UX:*:*) |
4822 | + echo sx6-nec-superux${UNAME_RELEASE} |
4823 | + exit 0 ;; |
4824 | + Power*:Rhapsody:*:*) |
4825 | + echo powerpc-apple-rhapsody${UNAME_RELEASE} |
4826 | + exit 0 ;; |
4827 | + *:Rhapsody:*:*) |
4828 | + echo ${UNAME_MACHINE}-apple-rhapsody${UNAME_RELEASE} |
4829 | + exit 0 ;; |
4830 | + *:Darwin:*:*) |
4831 | + UNAME_PROCESSOR=`uname -p` || UNAME_PROCESSOR=unknown |
4832 | + case $UNAME_PROCESSOR in |
4833 | + *86) UNAME_PROCESSOR=i686 ;; |
4834 | + unknown) UNAME_PROCESSOR=powerpc ;; |
4835 | + esac |
4836 | + echo ${UNAME_PROCESSOR}-apple-darwin${UNAME_RELEASE} |
4837 | + exit 0 ;; |
4838 | + *:procnto*:*:* | *:QNX:[0123456789]*:*) |
4839 | + UNAME_PROCESSOR=`uname -p` |
4840 | + if test "$UNAME_PROCESSOR" = "x86"; then |
4841 | + UNAME_PROCESSOR=i386 |
4842 | + UNAME_MACHINE=pc |
4843 | + fi |
4844 | + echo ${UNAME_PROCESSOR}-${UNAME_MACHINE}-nto-qnx${UNAME_RELEASE} |
4845 | + exit 0 ;; |
4846 | + *:QNX:*:4*) |
4847 | + echo i386-pc-qnx |
4848 | + exit 0 ;; |
4849 | + NSE-?:NONSTOP_KERNEL:*:*) |
4850 | + echo nse-tandem-nsk${UNAME_RELEASE} |
4851 | + exit 0 ;; |
4852 | + NSR-?:NONSTOP_KERNEL:*:*) |
4853 | + echo nsr-tandem-nsk${UNAME_RELEASE} |
4854 | + exit 0 ;; |
4855 | + *:NonStop-UX:*:*) |
4856 | + echo mips-compaq-nonstopux |
4857 | + exit 0 ;; |
4858 | + BS2000:POSIX*:*:*) |
4859 | + echo bs2000-siemens-sysv |
4860 | + exit 0 ;; |
4861 | + DS/*:UNIX_System_V:*:*) |
4862 | + echo ${UNAME_MACHINE}-${UNAME_SYSTEM}-${UNAME_RELEASE} |
4863 | + exit 0 ;; |
4864 | + *:Plan9:*:*) |
4865 | + # "uname -m" is not consistent, so use $cputype instead. 386 |
4866 | + # is converted to i386 for consistency with other x86 |
4867 | + # operating systems. |
4868 | + if test "$cputype" = "386"; then |
4869 | + UNAME_MACHINE=i386 |
4870 | + else |
4871 | + UNAME_MACHINE="$cputype" |
4872 | + fi |
4873 | + echo ${UNAME_MACHINE}-unknown-plan9 |
4874 | + exit 0 ;; |
4875 | + *:TOPS-10:*:*) |
4876 | + echo pdp10-unknown-tops10 |
4877 | + exit 0 ;; |
4878 | + *:TENEX:*:*) |
4879 | + echo pdp10-unknown-tenex |
4880 | + exit 0 ;; |
4881 | + KS10:TOPS-20:*:* | KL10:TOPS-20:*:* | TYPE4:TOPS-20:*:*) |
4882 | + echo pdp10-dec-tops20 |
4883 | + exit 0 ;; |
4884 | + XKL-1:TOPS-20:*:* | TYPE5:TOPS-20:*:*) |
4885 | + echo pdp10-xkl-tops20 |
4886 | + exit 0 ;; |
4887 | + *:TOPS-20:*:*) |
4888 | + echo pdp10-unknown-tops20 |
4889 | + exit 0 ;; |
4890 | + *:ITS:*:*) |
4891 | + echo pdp10-unknown-its |
4892 | + exit 0 ;; |
4893 | + SEI:*:*:SEIUX) |
4894 | + echo mips-sei-seiux${UNAME_RELEASE} |
4895 | + exit 0 ;; |
4896 | + *:DragonFly:*:*) |
4897 | + echo ${UNAME_MACHINE}-unknown-dragonfly`echo ${UNAME_RELEASE}|sed -e 's/[-(].*//'` |
4898 | + exit 0 ;; |
4899 | + *:*VMS:*:*) |
4900 | + UNAME_MACHINE=`(uname -p) 2>/dev/null` |
4901 | + case "${UNAME_MACHINE}" in |
4902 | + A*) echo alpha-dec-vms && exit 0 ;; |
4903 | + I*) echo ia64-dec-vms && exit 0 ;; |
4904 | + V*) echo vax-dec-vms && exit 0 ;; |
4905 | + esac ;; |
4906 | + *:XENIX:*:SysV) |
4907 | + echo i386-pc-xenix |
4908 | + exit 0 ;; |
4909 | +esac |
4910 | + |
4911 | +#echo '(No uname command or uname output not recognized.)' 1>&2 |
4912 | +#echo "${UNAME_MACHINE}:${UNAME_SYSTEM}:${UNAME_RELEASE}:${UNAME_VERSION}" 1>&2 |
4913 | + |
4914 | +eval $set_cc_for_build |
4915 | +cat >$dummy.c <<EOF |
4916 | +#ifdef _SEQUENT_ |
4917 | +# include <sys/types.h> |
4918 | +# include <sys/utsname.h> |
4919 | +#endif |
4920 | +main () |
4921 | +{ |
4922 | +#if defined (sony) |
4923 | +#if defined (MIPSEB) |
4924 | + /* BFD wants "bsd" instead of "newsos". Perhaps BFD should be changed, |
4925 | + I don't know.... */ |
4926 | + printf ("mips-sony-bsd\n"); exit (0); |
4927 | +#else |
4928 | +#include <sys/param.h> |
4929 | + printf ("m68k-sony-newsos%s\n", |
4930 | +#ifdef NEWSOS4 |
4931 | + "4" |
4932 | +#else |
4933 | + "" |
4934 | +#endif |
4935 | + ); exit (0); |
4936 | +#endif |
4937 | +#endif |
4938 | + |
4939 | +#if defined (__arm) && defined (__acorn) && defined (__unix) |
4940 | + printf ("arm-acorn-riscix"); exit (0); |
4941 | +#endif |
4942 | + |
4943 | +#if defined (hp300) && !defined (hpux) |
4944 | + printf ("m68k-hp-bsd\n"); exit (0); |
4945 | +#endif |
4946 | + |
4947 | +#if defined (NeXT) |
4948 | +#if !defined (__ARCHITECTURE__) |
4949 | +#define __ARCHITECTURE__ "m68k" |
4950 | +#endif |
4951 | + int version; |
4952 | + version=`(hostinfo | sed -n 's/.*NeXT Mach \([0-9]*\).*/\1/p') 2>/dev/null`; |
4953 | + if (version < 4) |
4954 | + printf ("%s-next-nextstep%d\n", __ARCHITECTURE__, version); |
4955 | + else |
4956 | + printf ("%s-next-openstep%d\n", __ARCHITECTURE__, version); |
4957 | + exit (0); |
4958 | +#endif |
4959 | + |
4960 | +#if defined (MULTIMAX) || defined (n16) |
4961 | +#if defined (UMAXV) |
4962 | + printf ("ns32k-encore-sysv\n"); exit (0); |
4963 | +#else |
4964 | +#if defined (CMU) |
4965 | + printf ("ns32k-encore-mach\n"); exit (0); |
4966 | +#else |
4967 | + printf ("ns32k-encore-bsd\n"); exit (0); |
4968 | +#endif |
4969 | +#endif |
4970 | +#endif |
4971 | + |
4972 | +#if defined (__386BSD__) |
4973 | + printf ("i386-pc-bsd\n"); exit (0); |
4974 | +#endif |
4975 | + |
4976 | +#if defined (sequent) |
4977 | +#if defined (i386) |
4978 | + printf ("i386-sequent-dynix\n"); exit (0); |
4979 | +#endif |
4980 | +#if defined (ns32000) |
4981 | + printf ("ns32k-sequent-dynix\n"); exit (0); |
4982 | +#endif |
4983 | +#endif |
4984 | + |
4985 | +#if defined (_SEQUENT_) |
4986 | + struct utsname un; |
4987 | + |
4988 | + uname(&un); |
4989 | + |
4990 | + if (strncmp(un.version, "V2", 2) == 0) { |
4991 | + printf ("i386-sequent-ptx2\n"); exit (0); |
4992 | + } |
4993 | + if (strncmp(un.version, "V1", 2) == 0) { /* XXX is V1 correct? */ |
4994 | + printf ("i386-sequent-ptx1\n"); exit (0); |
4995 | + } |
4996 | + printf ("i386-sequent-ptx\n"); exit (0); |
4997 | + |
4998 | +#endif |
4999 | + |
5000 | +#if defined (vax) |
I apologize in advance for not processing this in the correct way. If the process for submitting documentation is different I am ready to learn. :))
I have also prepared a piece on hardware for ltsp (using Alkis Georgopoulos as the source) but am confused because I see a different place where another guide has been submitted (https:/ /wiki.debian. org/DebianEdu/ Documentation/ Stretch/ HowTo/NetworkCl ients?highlight =%28ltsp% 29) which in my estimation needs editing and an unmerged document (http:// bazaar. launchpad. net/~ltsp- docwriters/ ltsp/ltsp- docs-trunk/ view/head: /LTSPManual. xml) which I assume is not to be touched.
Any suggestions are welcome.
Richard