Merge lp:~pfalcon/linaro-android-gerrit-support/cronjob-deploy into lp:linaro-android-gerrit-support
- cronjob-deploy
- Merge into gerrit-support
Status: | Merged |
---|---|
Merged at revision: | 64 |
Proposed branch: | lp:~pfalcon/linaro-android-gerrit-support/cronjob-deploy |
Merge into: | lp:linaro-android-gerrit-support |
Diff against target: |
234 lines (+36/-145) 7 files modified
README (+13/-0) android.git.linaro.org/README (+2/-0) android.git.linaro.org/cron-runner.sh (+11/-0) android.git.linaro.org/crontab (+4/-0) linaro-aosp-projects.txt (+0/-142) mirror.conf (+5/-3) re-remote.sh (+1/-0) |
To merge this branch: | bzr merge lp:~pfalcon/linaro-android-gerrit-support/cronjob-deploy |
Related bugs: |
Reviewer | Review Type | Date Requested | Status |
---|---|---|---|
Paul Sokolovsky | Approve | ||
Review via email: mp+84309@code.launchpad.net |
Commit message
Description of the change
These are changes for last WI "Deploy as cronjob" of
https:/
So, these are changes needed to actually run automatic upstream mirror on android.
- 70. By Paul Sokolovsky
-
cd into tool dir to access mirror.conf
Deepti B. Kalakeri (deeptik) wrote : | # |
Paul Sokolovsky (pfalcon) wrote : | # |
Hello,
On Mon, 05 Dec 2011 10:04:23 -0000
"Deepti B. Kalakeri" <email address hidden> wrote:
[]
> > === modified file 'mirror.conf'
> > --- mirror.conf 2011-11-23 18:53:52 +0000
> > +++ mirror.conf 2011-12-02 18:08:23 +0000
> > @@ -1,7 +1,9 @@
> > [DESTINATION]
> > # Usecase 1: target is Gerrit
> > -$repo_root = ssh://<email address hidden>:29418
> > -$gerrit_host = <email address hidden>
> > +#$repo_root = ssh://<email address hidden>:29418
> > +#$gerrit_host = <email address hidden>
>
> Was it intentional to leave the above commented than removed?
Yes, for debugging any issues and being able to run easily at
alternative location (android-build).
>
> >
> > +$repo_root = ssh://linaro-
> > +$gerrit_host = linaro-
> > $gerrit_port = 29418
> >
>
> we can move the gerrit_port declaration above and can use it in the
> repo_root command above.
No, those are different settings for different services (git vs
gerrit). They happen to coincide in this case, but that's it -
coincidence.
--
Best Regards,
Paul
Linaro.org | Open source software for ARM SoCs
Follow Linaro: http://
http://
Deepti B. Kalakeri (deeptik) wrote : | # |
On Mon, Dec 5, 2011 at 4:31 PM, Paul Sokolovsky
<email address hidden>wrote:
> Hello,
>
> On Mon, 05 Dec 2011 10:04:23 -0000
> "Deepti B. Kalakeri" <email address hidden> wrote:
>
> []
> > > === modified file 'mirror.conf'
> > > --- mirror.conf 2011-11-23 18:53:52 +0000
> > > +++ mirror.conf 2011-12-02 18:08:23 +0000
> > > @@ -1,7 +1,9 @@
> > > [DESTINATION]
> > > # Usecase 1: target is Gerrit
> > > -$repo_root = ssh://<email address hidden>:29418
> > > -$gerrit_host = <email address hidden>
> > > +#$repo_root = ssh://<email address hidden>:29418
> > > +#$gerrit_host = <email address hidden>
> >
> > Was it intentional to leave the above commented than removed?
>
> Yes, for debugging any issues and being able to run easily at
> alternative location (android-build).
>
> >
> > >
> > > +$repo_root = ssh://linaro-
> > > +$gerrit_host = linaro-
> > > $gerrit_port = 29418
> > >
> >
> > we can move the gerrit_port declaration above and can use it in the
> > repo_root command above.
>
> No, those are different settings for different services (git vs
> gerrit). They happen to coincide in this case, but that's it -
> coincidence.
>
> ah! ok.
+1 looks fine.
>
> --
> Best Regards,
> Paul
>
> Linaro.org | Open source software for ARM SoCs
> Follow Linaro: http://
> http://
>
>
> https:/
> Your team Linaro Infrastructure is requested to review the proposed merge
> of lp:~pfalcon/linaro-android-gerrit-support/cronjob-deploy into
> lp:linaro-android-gerrit-support.
>
--
Thanks and Regards,
Deepti
Infrastructure Team Member, Linaro Platform Teams
Linaro.org | Open source software for ARM SoCs
Follow Linaro: http://
http://
Guilherme Salgado (salgado) wrote : | # |
I don't quite like the commented code left behind nor the hard-coded paths. The former is not a big deal but we should at least try and fix the latter
Paul Sokolovsky (pfalcon) wrote : | # |
> I don't quite like the commented code left behind nor the hard-coded paths.
> The former is not a big deal but we should at least try and fix the latter
Well, all the changes here are essentially configuration for running specific instance of service on specific machine, so yes, all paths are specific to that machine. We could have two configs for production vs reserve/test machine, but then they would need to be updated in parallel, so as they differ in just 3-4 lines, it's easier to keep them in one file, to avoid head-scratching trying to understand what's where in 6 months ;-).
Guilherme Salgado (salgado) wrote : | # |
> > I don't quite like the commented code left behind nor the hard-coded paths.
> > The former is not a big deal but we should at least try and fix the latter
>
> Well, all the changes here are essentially configuration for running specific
> instance of service on specific machine, so yes, all paths are specific to
> that machine. We could have two configs for production vs reserve/test
> machine, but then they would need to be updated in parallel, so as they differ
> in just 3-4 lines, it's easier to keep them in one file, to avoid head-
> scratching trying to understand what's where in 6 months ;-).
runner.sh doesn't look like configuration to me; it looks like a script which developers may want to run on their machine and in that case they'll have to manually change the script. I see that sync-linaro.sh also has a hard-coded path in it, so if I wanted to test linaro-
As you said, those paths are really just configuration so I think they should at least be moved to a config file so that they're not scattered all around and developers have a single place to change when they want to run it on their machine.
Paul Sokolovsky (pfalcon) wrote : | # |
Hello,
On Wed, 07 Dec 2011 13:38:32 -0000
Guilherme Salgado <email address hidden> wrote:
> > > I don't quite like the commented code left behind nor the
> > > hard-coded paths. The former is not a big deal but we should at
> > > least try and fix the latter
> >
> > Well, all the changes here are essentially configuration for
> > running specific instance of service on specific machine, so yes,
> > all paths are specific to that machine. We could have two configs
> > for production vs reserve/test machine, but then they would need to
> > be updated in parallel, so as they differ in just 3-4 lines, it's
> > easier to keep them in one file, to avoid head- scratching trying
> > to understand what's where in 6 months ;-).
>
> runner.sh doesn't look like configuration to me; it looks like a
> script which developers may want to run on their machine and in that
> case they'll have to manually change the script.
Well, that's typical pattern of setting up a cronjob: instead of
putting too long lines into crontab itself, you call a wrapper script
which sets up environment/
to run. So, it's kinda of "executable configuration". runner.sh is not
intended to be run by developer, except for the case of simulating
running from cron (that's another case why this pattern of wrapper
scripts is useful). So, maybe the name is confusing, let's rename it to
cron-runner.sh and add corresponding comment if that answers your
concerns.
> I see that
> sync-linaro.sh also has a hard-coded path in it, so if I wanted to
Well, that's a deprecated script ;-). It really should be removed some
time later, once new system is up and running for some buffer time.
> test linaro-
> at least these two places and I'd have to find a way to keep those
> changes locally while I'm trying l-a-g-s, but not commit them back.
> We used to have a similar problem in Offspring and I can tell you
> it's very annoying.
Yes, I understand your concerns. But those are indeed just one-off
wrapper scripts. Except that one "one-off" is special - the production
one. It could have been a separate script lying elsewhere on a target
host, but we really want to capture all details of the production
service setup, to be able to re-setup it without contingencies in case
of need. The ideal situation to handle that would be to keep generic
tool vs specific service config in two different repositories, and I
would happily do that if LP supported >1 repos per project. But
creating a new LP project for each such case doesn't scale at all IMHO.
Anyway, if you think that separating generic tool vs specific config
into different dirs in repo will be helpful, let's do that.
Finally, answering your concern how to deal with that locally - you'd
need just to copy runner.sh to your one-off wrapper, runner-local.sh,
and change it, that's the idea.
> As you said, those paths are really just configuration so I think
> they should at least be moved to a config file so that they're not
> scattered all around and developers have a single place to change
> when they want to run it on their machine.
So, there'r...
- 71. By Paul Sokolovsky
-
Rename runner.sh => cron-runner.sh to avoid confusion.
- 72. By Paul Sokolovsky
-
Remove stale file.
- 73. By Paul Sokolovsky
-
Move older utilities to old/ .
Keep them around for some time yet.
- 74. By Paul Sokolovsky
-
Add description for re-remote.sh
- 75. By Paul Sokolovsky
-
Move host-specific cron config to the corresponding dir.
Paul Sokolovsky (pfalcon) wrote : | # |
Ok, made changes along the lines of what was described above. This wasn't intended to be done for this merge request (not related to cronjob set up per se), so now there're too many changes, so it make sense to look at the end result: http://
Guilherme Salgado (salgado) wrote : | # |
On Thu, 2011-12-08 at 12:17 +0000, Paul Sokolovsky wrote:
> Hello,
>
> On Wed, 07 Dec 2011 13:38:32 -0000
> Guilherme Salgado <email address hidden> wrote:
>
> > > > I don't quite like the commented code left behind nor the
> > > > hard-coded paths. The former is not a big deal but we should at
> > > > least try and fix the latter
> > >
> > > Well, all the changes here are essentially configuration for
> > > running specific instance of service on specific machine, so yes,
> > > all paths are specific to that machine. We could have two configs
> > > for production vs reserve/test machine, but then they would need to
> > > be updated in parallel, so as they differ in just 3-4 lines, it's
> > > easier to keep them in one file, to avoid head- scratching trying
> > > to understand what's where in 6 months ;-).
> >
> > runner.sh doesn't look like configuration to me; it looks like a
> > script which developers may want to run on their machine and in that
> > case they'll have to manually change the script.
>
> Well, that's typical pattern of setting up a cronjob: instead of
> putting too long lines into crontab itself, you call a wrapper script
> which sets up environment/
> to run. So, it's kinda of "executable configuration". runner.sh is not
> intended to be run by developer, except for the case of simulating
> running from cron (that's another case why this pattern of wrapper
> scripts is useful). So, maybe the name is confusing, let's rename it to
> cron-runner.sh and add corresponding comment if that answers your
> concerns.
>
> > I see that
> > sync-linaro.sh also has a hard-coded path in it, so if I wanted to
>
> Well, that's a deprecated script ;-). It really should be removed some
> time later, once new system is up and running for some buffer time.
>
> > test linaro-
> > at least these two places and I'd have to find a way to keep those
> > changes locally while I'm trying l-a-g-s, but not commit them back.
> > We used to have a similar problem in Offspring and I can tell you
> > it's very annoying.
>
> Yes, I understand your concerns. But those are indeed just one-off
> wrapper scripts. Except that one "one-off" is special - the production
What do you mean with one-off here? To me a one-off script is something that
we'll run once and throw away, but here it's something that will run as a
cronjob in production, so it definitely doesn't fall on that category.
> one. It could have been a separate script lying elsewhere on a target
> host, but we really want to capture all details of the production
> service setup, to be able to re-setup it without contingencies in case
Ok, if that's not expected to be needed on developer machines it's probably
fine to keep the paths hard-coded, but it's a good idea to move it to a
separate directory, which you've already did. :)
> of need. The ideal situation to handle that would be to keep generic
> tool vs specific service config in two different repositories, and I
> would happily do that if LP supported >1 repos per project. But
> creating a new LP project for each such case doesn't...
- 76. By Paul Sokolovsky
-
Add READMEs.
Paul Sokolovsky (pfalcon) wrote : | # |
Hello Guilherme,
On Thu, 08 Dec 2011 18:22:24 -0000
Guilherme Salgado <email address hidden> wrote:
[]
> > Yes, I understand your concerns. But those are indeed just one-off
> > wrapper scripts. Except that one "one-off" is special - the
> > production
>
> What do you mean with one-off here? To me a one-off script is
> something that we'll run once and throw away, but here it's something
> that will run as a cronjob in production, so it definitely doesn't
> fall on that category.
Yep, I meant more along the lines "configured once for specific
usecase, and can be easily forked and reconfigured for another usecase".
> > one. It could have been a separate script lying elsewhere on a
> > target host, but we really want to capture all details of the
> > production service setup, to be able to re-setup it without
> > contingencies in case
>
> Ok, if that's not expected to be needed on developer machines it's
> probably fine to keep the paths hard-coded, but it's a good idea to
> move it to a separate directory, which you've already did. :)
>
> > of need. The ideal situation to handle that would be to keep generic
> > tool vs specific service config in two different repositories, and I
> > would happily do that if LP supported >1 repos per project. But
> > creating a new LP project for each such case doesn't scale at all
> > IMHO.
>
> You can have as many branches as you want under a given project in LP
> and those branches may contain distinct code bases, so you could do
> something like:
>
> mkdir cronscripts
> cd cronscripts
> bzr init
> cp /path/to/
> bzr add *
> bzr commit -m "..."
> bzr push
> lp:~linaro-infrastructure/linaro-android-gerrit-support/cronscripts
>
> And then when deploying you fetch that branch as well.
Well, yeah, I'm aware that one can have completely different content in
two branches and git, and you point that it can be doable for bzr too.
But is it natural use pattern for bzr? For git, that would be quite a
hack. And does LP has good support for such usage (e.g. does it allow
to specify descriptions for each such branch)? Anyway, I'll keep that
option in mind, and on next occasion will figure those questions.
>
> > Anyway, if you think that separating generic tool vs specific config
> > into different dirs in repo will be helpful, let's do that.
> >
> > Finally, answering your concern how to deal with that locally -
> > you'd need just to copy runner.sh to your one-off wrapper,
> > runner-local.sh, and change it, that's the idea.
>
> Yes, I know there are many ways a developer can deal with that, but
> my point is that they shouldn't have to change any code to run (or
> test) our software, and if they do need to change configs and such,
> it should be documented.
Yes, improving docs is a known TODO, I added
https:/
track case of this specific tool/service.
Thanks,
Paul
Linaro.org | Open source software for ARM SoCs
Follow Linaro: http://
http://
Preview Diff
1 | === added file 'README' |
2 | --- README 1970-01-01 00:00:00 +0000 |
3 | +++ README 2011-12-09 15:10:28 +0000 |
4 | @@ -0,0 +1,13 @@ |
5 | +git-gerrit-mirror is a generic tool to mirror set of git repositories. It |
6 | +can mirror one set of remote repositories ("upstreams") into set of |
7 | +corresponding remote repositories ("targets"), by keeping local working |
8 | +copy to perform fetch/push operations. It also has optional support for |
9 | +Gerrit services associated with upstream and target repositories, which |
10 | +can be used to automatically discover new projects in upstream and |
11 | +create corresponding projects in target. |
12 | + |
13 | +More documenation is available at: |
14 | +http://wiki.linaro.org/Platform/Android/GerritMirroring |
15 | + |
16 | +Project home page: |
17 | +https://launchpad.net/linaro-android-gerrit-support |
18 | |
19 | === added directory 'android.git.linaro.org' |
20 | === added file 'android.git.linaro.org/README' |
21 | --- android.git.linaro.org/README 1970-01-01 00:00:00 +0000 |
22 | +++ android.git.linaro.org/README 2011-12-09 15:10:28 +0000 |
23 | @@ -0,0 +1,2 @@ |
24 | +This directory contains configuration and helper scripts to run Gerrit |
25 | +mirroring cronjob on android.git.linaro.org host. |
26 | |
27 | === added file 'android.git.linaro.org/cron-runner.sh' |
28 | --- android.git.linaro.org/cron-runner.sh 1970-01-01 00:00:00 +0000 |
29 | +++ android.git.linaro.org/cron-runner.sh 2011-12-09 15:10:28 +0000 |
30 | @@ -0,0 +1,11 @@ |
31 | +#!/bin/bash |
32 | +# This script is intended to be run from crontab |
33 | + |
34 | +MIRROR_DIR=/srv/build_engineer_scratch/gerrit-mirror/ |
35 | +LOG_DIR=/srv/build_engineer_scratch/gerrit-mirror-logs/ |
36 | + |
37 | +TSTAMP=`date '+%Y%m%dT%H%M'` |
38 | +MYPATH=. |
39 | +LOG=$LOG_DIR/$TSTAMP-$1.log |
40 | + |
41 | +(time $MYPATH/git-gerrit-mirror --mirror-dir=$MIRROR_DIR $* >$LOG 2>&1) 2>>$LOG |
42 | |
43 | === added file 'android.git.linaro.org/crontab' |
44 | --- android.git.linaro.org/crontab 1970-01-01 00:00:00 +0000 |
45 | +++ android.git.linaro.org/crontab 2011-12-09 15:10:28 +0000 |
46 | @@ -0,0 +1,4 @@ |
47 | +# This is a sample crontab for Linaro Gerrit Mirror Service |
48 | + |
49 | +0 */2 * * * (cd ~/linaro-android-gerrit-support; android.git.linaro.org/cron-runner.sh fetch) |
50 | +30 */2 * * * (cd ~/linaro-android-gerrit-support; android.git.linaro.org/cron-runner.sh push) |
51 | |
52 | === removed file 'linaro-aosp-projects.txt' |
53 | --- linaro-aosp-projects.txt 2011-07-23 19:42:48 +0000 |
54 | +++ linaro-aosp-projects.txt 1970-01-01 00:00:00 +0000 |
55 | @@ -1,142 +0,0 @@ |
56 | -platform/bionic |
57 | -platform/bootable/bootloader/legacy |
58 | -platform/bootable/diskinstaller |
59 | -platform/bootable/recovery |
60 | -platform/cts |
61 | -platform/development |
62 | -device/common |
63 | -device/sample |
64 | -device/samsung/crespo |
65 | -platform/external/apache-http |
66 | -platform/external/astl |
67 | -platform/external/bison |
68 | -platform/external/blktrace |
69 | -platform/external/bluetooth/bluez |
70 | -platform/external/bluetooth/glib |
71 | -platform/external/bluetooth/hcidump |
72 | -platform/external/bouncycastle |
73 | -platform/external/bsdiff |
74 | -platform/external/bzip2 |
75 | -platform/external/chromium |
76 | -platform/external/clearsilver |
77 | -platform/external/dbus |
78 | -platform/external/dhcpcd |
79 | -platform/external/dnsmasq |
80 | -platform/external/dropbear |
81 | -platform/external/e2fsprogs |
82 | -platform/external/easymock |
83 | -platform/external/elfutils |
84 | -platform/external/embunit |
85 | -platform/external/emma |
86 | -platform/external/esd |
87 | -platform/external/expat |
88 | -platform/external/fdlibm |
89 | -platform/external/freetype |
90 | -platform/external/fsck_msdos |
91 | -platform/external/genext2fs |
92 | -platform/external/giflib |
93 | -platform/external/grub |
94 | -platform/external/gtest |
95 | -platform/external/guava |
96 | -platform/external/icu4c |
97 | -platform/external/iproute2 |
98 | -platform/external/ipsec-tools |
99 | -platform/external/iptables |
100 | -platform/external/jdiff |
101 | -platform/external/jhead |
102 | -platform/external/jpeg |
103 | -platform/external/jsr305 |
104 | -platform/external/junit |
105 | -platform/external/kernel-headers |
106 | -platform/external/libffi |
107 | -platform/external/libgsm |
108 | -platform/external/libnfc-nxp |
109 | -platform/external/libpcap |
110 | -platform/external/libpng |
111 | -platform/external/libvpx |
112 | -platform/external/libxml2 |
113 | -platform/external/mtpd |
114 | -platform/external/netcat |
115 | -platform/external/netperf |
116 | -platform/external/neven |
117 | -platform/external/nist-sip |
118 | -platform/external/oauth |
119 | -platform/external/openssl |
120 | -platform/external/oprofile |
121 | -platform/external/ping |
122 | -platform/external/ppp |
123 | -platform/external/proguard |
124 | -platform/external/protobuf |
125 | -platform/external/qemu |
126 | -platform/external/quake |
127 | -platform/external/safe-iop |
128 | -platform/external/sonivox |
129 | -platform/external/speex |
130 | -platform/external/sqlite |
131 | -platform/external/srec |
132 | -platform/external/stlport |
133 | -platform/external/strace |
134 | -platform/external/tagsoup |
135 | -platform/external/tcpdump |
136 | -platform/external/tinyxml |
137 | -platform/external/tremolo |
138 | -platform/external/v8 |
139 | -platform/external/wpa_supplicant |
140 | -platform/external/wpa_supplicant_6 |
141 | -platform/external/xmlwriter |
142 | -platform/external/yaffs2 |
143 | -platform/external/zlib |
144 | -platform/frameworks/ex |
145 | -platform/hardware/libhardware |
146 | -platform/hardware/libhardware_legacy |
147 | -platform/hardware/ril |
148 | -platform/ndk |
149 | -platform/packages/apps/AccountsAndSyncSettings |
150 | -platform/packages/apps/Bluetooth |
151 | -platform/packages/apps/Browser |
152 | -platform/packages/apps/Calculator |
153 | -platform/packages/apps/Calendar |
154 | -platform/packages/apps/Camera |
155 | -platform/packages/apps/CertInstaller |
156 | -platform/packages/apps/Contacts |
157 | -platform/packages/apps/DeskClock |
158 | -platform/packages/apps/Email |
159 | -platform/packages/apps/Gallery |
160 | -platform/packages/apps/Gallery3D |
161 | -platform/packages/apps/HTMLViewer |
162 | -platform/packages/apps/Launcher2 |
163 | -platform/packages/apps/Mms |
164 | -platform/packages/apps/Music |
165 | -platform/packages/apps/Nfc |
166 | -platform/packages/apps/PackageInstaller |
167 | -platform/packages/apps/Phone |
168 | -platform/packages/apps/Protips |
169 | -platform/packages/apps/QuickSearchBox |
170 | -platform/packages/apps/Settings |
171 | -platform/packages/apps/SoundRecorder |
172 | -platform/packages/apps/SpeechRecorder |
173 | -platform/packages/apps/Stk |
174 | -platform/packages/apps/Tag |
175 | -platform/packages/apps/VoiceDialer |
176 | -platform/packages/experimental |
177 | -platform/packages/inputmethods/LatinIME |
178 | -platform/packages/inputmethods/OpenWnn |
179 | -platform/packages/inputmethods/PinyinIME |
180 | -platform/packages/providers/ApplicationsProvider |
181 | -platform/packages/providers/CalendarProvider |
182 | -platform/packages/providers/ContactsProvider |
183 | -platform/packages/providers/DownloadProvider |
184 | -platform/packages/providers/DrmProvider |
185 | -platform/packages/providers/GoogleContactsProvider |
186 | -platform/packages/providers/MediaProvider |
187 | -platform/packages/providers/TelephonyProvider |
188 | -platform/packages/providers/UserDictionaryProvider |
189 | -platform/packages/wallpapers/Basic |
190 | -platform/packages/wallpapers/LivePicker |
191 | -platform/packages/wallpapers/MagicSmoke |
192 | -platform/packages/wallpapers/MusicVisualization |
193 | -platform/prebuilt |
194 | -platform/sdk |
195 | -platform/system/bluetooth |
196 | -platform/system/extras |
197 | -platform/system/netd |
198 | |
199 | === modified file 'mirror.conf' |
200 | --- mirror.conf 2011-11-23 18:53:52 +0000 |
201 | +++ mirror.conf 2011-12-09 15:10:28 +0000 |
202 | @@ -1,7 +1,9 @@ |
203 | [DESTINATION] |
204 | # Usecase 1: target is Gerrit |
205 | -$repo_root = ssh://pfalcon@android.git.linaro.org:29418 |
206 | -$gerrit_host = pfalcon@android.git.linaro.org |
207 | +#$repo_root = ssh://pfalcon@android.git.linaro.org:29418 |
208 | +#$gerrit_host = pfalcon@android.git.linaro.org |
209 | +$repo_root = ssh://linaro-aosp-sync-bot@gerrit-git:29418 |
210 | +$gerrit_host = linaro-aosp-sync-bot@gerrit-git |
211 | $gerrit_port = 29418 |
212 | |
213 | # Usecase 2: target is a remote tree |
214 | @@ -26,5 +28,5 @@ |
215 | #$active = false |
216 | kernel/omap = kernel/omap-omapzoom |
217 | |
218 | -[git://github.com] |
219 | +[https://github.com] |
220 | msg555/PowerTutor = platform/packages/apps/PowerTutor |
221 | |
222 | === added directory 'old' |
223 | === renamed file 'gerrit-create-projects.py' => 'old/gerrit-create-projects.py' |
224 | === renamed file 'ls-projects2gerrit.py' => 'old/ls-projects2gerrit.py' |
225 | === renamed file 'manifest2gerrit.py' => 'old/manifest2gerrit.py' |
226 | === renamed file 'mirror-aosp-project.sh' => 'old/mirror-aosp-project.sh' |
227 | === renamed file 'sync-aosp.sh' => 'old/sync-aosp.sh' |
228 | === renamed file 'sync-linaro.sh' => 'old/sync-linaro.sh' |
229 | === modified file 're-remote.sh' |
230 | --- re-remote.sh 2011-11-01 21:35:13 +0000 |
231 | +++ re-remote.sh 2011-12-09 15:10:28 +0000 |
232 | @@ -1,1 +1,2 @@ |
233 | +# Replace one git remote URL with another, similar to svn switch --relocate |
234 | find -path '*.git/config' | xargs -n1 sed -i.bak -e's%git://android.git.kernel.org/%https://android.googlesource.com/%' |
On Fri, Dec 2, 2011 at 11:39 PM, Paul Sokolovsky <<email address hidden>
> wrote:
> Paul Sokolovsky has proposed merging infrastructure) /code.launchpad .net/~pfalcon/ linaro- android- gerrit- support/ cronjob- deploy/ +merge/ 84309<https:/ /code.launchpad .net/%7Epfalcon /linaro- android- gerrit- support/ cronjob- deploy/ +merge/ 84309> /blueprints. launchpad. net/linaro- android- infrastructure/ +spec/linaro- android- automate- upstream- sync git.linaro. org /code.launchpad .net/~pfalcon/ linaro- android- gerrit- support/ cronjob- deploy/ +merge/ 84309<https:/ /code.launchpad .net/%7Epfalcon /linaro- android- gerrit- support/ cronjob- deploy/ +merge/ 84309> android- gerrit- support/ runner. sh fetch android- gerrit- support/ runner. sh push
> lp:~pfalcon/linaro-android-gerrit-support/cronjob-deploy into
> lp:linaro-android-gerrit-support.
>
> Requested reviews:
> Linaro Infrastructure (linaro-
>
> For more details, see:
>
> https:/
>
> These are changes for last WI "Deploy as cronjob" of
>
> https:/
>
> So, these are changes needed to actually run automatic upstream mirror on
> android.
>
> --
>
> https:/
> Your team Linaro Infrastructure is requested to review the proposed merge
> of lp:~pfalcon/linaro-android-gerrit-support/cronjob-deploy into
> lp:linaro-android-gerrit-support.
>
> === added file 'crontab'
> --- crontab 1970-01-01 00:00:00 +0000
> +++ crontab 2011-12-02 18:08:23 +0000
> @@ -0,0 +1,4 @@
> +# This is a sample crontab for Linaro Gerrit Mirror Service
> +
> +0 */2 * * * ~/linaro-
> +30 */2 * * * ~/linaro-
>
> === modified file 'mirror.conf'
> --- mirror.conf 2011-11-23 18:53:52 +0000
> +++ mirror.conf 2011-12-02 18:08:23 +0000
> @@ -1,7 +1,9 @@
> [DESTINATION]
> # Usecase 1: target is Gerrit
> -$repo_root = ssh://<email address hidden>:29418
> -$gerrit_host = <email address hidden>
> +#$repo_root = ssh://<email address hidden>:29418
> +#$gerrit_host = <email address hidden>
Was it intentional to leave the above commented than removed?
> aosp-sync- bot@gerrit- git:29418 aosp-sync- bot@gerrit- git
> +$repo_root = ssh://linaro-
> +$gerrit_host = linaro-
> $gerrit_port = 29418
>
we can move the gerrit_port declaration above and can use it in the
repo_root command above.
> omap-omapzoom /github. com] packages/ apps/PowerTutor DIR=/srv/ build_engineer_ scratch/ gerrit- mirror/ /srv/build_ engineer_ scratch/ gerrit- mirror- logs/ DIR/$TSTAMP- $1.log git-gerrit- mirror --mirror- dir=$MIRROR_ DIR $* >$LOG 2>&1)
> # Usecase 2: target is a remote tree
> @@ -26,5 +28,5 @@
> #$active = false
> kernel/omap = kernel/
>
> -[git://github.com]
> +[https:/
> msg555/PowerTutor = platform/
>
> === added file 'runner.sh'
> --- runner.sh 1970-01-01 00:00:00 +0000
> +++ runner.sh 2011-12-02 18:08:23 +0000
> @@ -0,0 +1,10 @@
> +#!/bin/bash
> +
> +MIRROR_
> +LOG_DIR=
> +TSTAMP=`date '+%Y%m%dT%H%M'`
> +
> +MYPATH=`dirname $0`
> +
> +LOG=$LOG_
> +(time $MYPATH/
> 2>>$LOG
>
>
>
--
Thanks!!!
Deepti