Merge ~mirespace/ubuntu/+source/corosync:sru-corosync-bionic-lp1677684-lp1437359 into ubuntu/+source/corosync:ubuntu/bionic-devel
Status: | Superseded |
---|---|
Proposed branch: | ~mirespace/ubuntu/+source/corosync:sru-corosync-bionic-lp1677684-lp1437359 |
Merge into: | ubuntu/+source/corosync:ubuntu/bionic-devel |
Diff against target: |
47 lines (+16/-1) 3 files modified
debian/changelog (+12/-0) debian/control (+3/-1) debian/libtotem-pg5.symbols (+1/-0) |
Related bugs: |
Reviewer | Review Type | Date Requested | Status |
---|---|---|---|
Athos Ribeiro (community) | Approve | ||
Ubuntu Core Development Team | Pending | ||
Review via email: mp+409319@code.launchpad.net |
This proposal has been superseded by a proposal from 2021-10-18.
Description of the change
Hi,
PPA for this is ppa:mirespace/
I have some doubts about the SRU templates (especially for the LP #1437359) and because the build shows a lintian error, so I prefer to show them so we can discuss the way to handle it (I mean, I'm asking for advice on this).
With this, two bugs for corosync are going to be fixed in Bionic: LP1677684 & LP1437359. Both fixes have been cherry-picked from the work of Rafael Tinoco and Jorge Niedbalski but for Focal series:
[Jorge Niedbalski] - dd471ac791ee8f5
* d/control: corosync binary depends on libqb-dev (LP: #1677684)
[Rafael David Tinoco] - 16a37d42582913c
* debian/
For the first one, we can test that with the fix we obtain a correct answer:
ubuntu@
Dumping the contents of /var/lib/
[debug] shm size:8392704; real_size:8392704; rb->word_
[debug] read total of: 8392724
Ringbuffer:
->NORMAL
->write_pt [2866]
->read_pt [0]
->size [2098176 words]
=>free [8381236 bytes]
=>used [11464 bytes]
debug Sep 28 20:21:50 totempg_
debug Sep 28 20:21:50 totemsrp_
debug Sep 28 20:21:50 totemsrp_
debug Sep 28 20:21:50 totemsrp_
debug Sep 28 20:21:50 totemsrp_
debug Sep 28 20:21:50 totemsrp_
debug Sep 28 20:21:50 totemsrp_
debug Sep 28 20:21:50 totemsrp_
debug Sep 28 20:21:50 totemsrp_
debug Sep 28 20:21:50 totemsrp_
debug Sep 28 20:21:50 totemsrp_
debug Sep 28 20:21:50 totemsrp_
debug Sep 28 20:21:50 totemsrp_
debug Sep 28 20:21:50 totemsrp_
debug Sep 28 20:21:50 totemsrp_
debug Sep 28 20:21:50 totemsrp_
debug Sep 28 20:21:50 totemsrp_
debug Sep 28 20:21:50 totemsrp_
notice Sep 28 20:21:50 totemnet_
notice Sep 28 20:21:50 init_nss(688):14: Initializing transmit/receive security (NSS) crypto: none hash: none
debug Sep 28 20:21:50 totemudp_
debug Sep 28 20:21:50 totemudp_
debug Sep 28 20:21:50 totemudp_
debug Sep 28 20:21:50 totemudp_
trace Sep 28 20:21:50 qb_loop_
trace Sep 28 20:21:50 qb_loop_
trace Sep 28 20:21:50 qb_loop_
notice Sep 28 20:21:50 timer_function_
debug Sep 28 20:21:50 main_iface_
info Sep 28 20:21:50 qb_ipcs_
trace Sep 28 20:21:50 qb_loop_
info Sep 28 20:21:50 qb_ipcs_
trace Sep 28 20:21:50 qb_loop_
info Sep 28 20:21:50 qb_ipcs_
trace Sep 28 20:21:50 qb_loop_
info Sep 28 20:21:50 qb_ipcs_
trace Sep 28 20:21:50 qb_loop_
info Sep 28 20:21:50 qb_ipcs_
trace Sep 28 20:21:50 qb_loop_
debug Sep 28 20:21:50 memb_state_
debug Sep 28 20:21:50 memb_state_
debug Sep 28 20:21:50 old_ring_
debug Sep 28 20:21:50 memb_state_
debug Sep 28 20:21:50 message_
debug Sep 28 20:21:50 memb_state_
debug Sep 28 20:21:50 memb_state_
debug Sep 28 20:21:50 memb_state_
debug Sep 28 20:21:50 memb_state_
debug Sep 28 20:21:50 memb_state_
debug Sep 28 20:21:50 message_
debug Sep 28 20:21:50 message_
trace Sep 28 20:21:50 totemsrp_
debug Sep 28 20:21:50 message_
debug Sep 28 20:21:50 message_
debug Sep 28 20:21:50 message_
debug Sep 28 20:21:50 message_
debug Sep 28 20:21:50 message_
debug Sep 28 20:21:50 message_
debug Sep 28 20:21:50 message_
debug Sep 28 20:21:50 message_
debug Sep 28 20:21:50 message_
debug Sep 28 20:21:50 old_ring_
debug Sep 28 20:21:50 deliver_
trace Sep 28 20:21:50 memb_state_
debug Sep 28 20:21:50 totempg_
debug Sep 28 20:21:50 memb_state_
notice Sep 28 20:21:50 memb_state_
trace Sep 28 20:21:50 totemsrp_
trace Sep 28 20:21:50 messages_
trace Sep 28 20:21:50 messages_
trace Sep 28 20:21:50 messages_
trace Sep 28 20:21:50 totemsrp_
trace Sep 28 20:21:50 messages_
trace Sep 28 20:21:50 messages_
trace Sep 28 20:21:50 totemsrp_
trace Sep 28 20:21:50 totemsrp_
trace Sep 28 20:21:50 messages_
trace Sep 28 20:21:50 messages_
trace Sep 28 20:21:50 messages_
trace Sep 28 20:21:50 messages_
trace Sep 28 20:21:50 totemsrp_
trace Sep 28 20:21:50 messages_
trace Sep 28 20:21:50 messages_
trace Sep 28 20:21:50 messages_
debug Sep 28 20:21:50 totempg_
trace Sep 28 20:21:50 messages_
trace Sep 28 20:21:50 messages_
trace Sep 28 20:22:03 qb_loop_
debug Sep 28 20:22:03 handle_
debug Sep 28 20:22:03 qb_ipcs_
debug Sep 28 20:22:03 qb_rb_open_
debug Sep 28 20:22:03 qb_rb_open_
debug Sep 28 20:22:03 qb_rb_open_
trace Sep 28 20:22:03 qb_loop_
debug Sep 28 20:22:03 qb_ipcs_
debug Sep 28 20:22:03 qb_ipcs_
trace Sep 28 20:22:03 qb_rb_close(290):9: ENTERING qb_rb_close()
debug Sep 28 20:22:03 qb_rb_close_
trace Sep 28 20:22:03 my_posix_
trace Sep 28 20:22:03 qb_rb_close(290):9: ENTERING qb_rb_close()
debug Sep 28 20:22:03 qb_rb_close_
trace Sep 28 20:22:03 my_posix_
trace Sep 28 20:22:03 qb_rb_close(290):9: ENTERING qb_rb_close()
debug Sep 28 20:22:03 qb_rb_close_
trace Sep 28 20:22:03 my_posix_
debug Sep 28 20:22:03 handle_
debug Sep 28 20:22:03 qb_ipcs_
debug Sep 28 20:22:03 qb_rb_open_
debug Sep 28 20:22:03 qb_rb_open_
debug Sep 28 20:22:03 qb_rb_open_
ERROR: qb_rb_chunk_read failed: Connection timed out
[trace] ENTERING qb_rb_close()
[debug] Free'ing ringbuffer: /dev/shm/
For the second one I didn't see any check (I suppose for the bug dependency with sysV)... I can force it and check the PID's change (in fact, I didn't write the SRU template for that depending on your opinion).
About the SRUs template, there are a kind-of previous one for LP: #1677684, and no one for the PID bug... still thinking about the steps to reproduce for this last one (as I said before).
Also, I noticed an error on lintian when building (that I suppose it has to be resolved before it can be SRU-processed):
E: libtotem-pg5: symbols-
E: Lintian run failed (policy violation)
Autopackage Tests (OK):
autopkgtest [20:28:15]: test corosync: [------
+ corosync-cfgtool -s
+ grep -20 ring 0 active with no faults
Printing ring status.
Local node ID 2130706433
RING ID 0
id = 127.0.0.1
status = ring 0 active with no faults
+ corosync-quorumtool
+ grep -20 1 localhost (local)
Quorum information
------------------
Date: Tue Sep 28 18:28:16 2021
Quorum provider: corosync_votequorum
Nodes: 1
Node ID: 2130706433
Ring ID: 2130706433/4
Quorate: No
Votequorum information
-------
Expected votes: 2
Highest expected: 2
Total votes: 1
Quorum: 2 Activity blocked
Flags:
Membership information
-------
Nodeid Votes Name
2130706433 1 localhost (local)
autopkgtest [20:28:16]: test corosync: -------
autopkgtest [20:28:16]: test corosync: - - - - - - - - - - results - - - - - - - - - -
corosync PASS
autopkgtest [20:28:17]: @@@@@@@
corosync PASS
Thanks in advance for your time in reviewing this! And for your hints about what needs to be done.
On Tuesday, September 28 2021, Miriam España Acebal wrote:
> Also, I noticed an error on lintian when building (that I suppose it has to be resolved before it can be SRU-processed): file-contains- current- version- with-debian- revision on symbol crypto_ get_current_ sec_header_ size@Base
>
> E: libtotem-pg5: symbols-
> E: Lintian run failed (policy violation)
Hi Miriam,
This is not really a review, but just a reply to the question above. get_current_ sec_header_ size is not listed in the pg5.symbols file, as can be seen during the build:
This lintian error happens because the symbol
crypto_
d/libtotem-
... libtotem- pg5/DEBIAN/ symbols doesn't match completely debian/ libtotem- pg5.symbols libtotem- pg5.symbols (libtotem- pg5_2.4. 3-0ubuntu1. 2_amd64) nBsU7j 2021-09-28 20:47:44.575794305 +0000 to_nss@ Base 1.99.9 authenticate_ and_decrypt@ Base 1.99.9 encrypt_ and_sign@ Base 1.99.9 get_current_ sec_header_ size@Base 2.4.3-0ubuntu1.2 sec_header_ size@Base 1.99.9 block_len@ Base 1.99.9
dpkg-gensymbols: warning: some new symbols appeared in the symbols file: see diff output below
dpkg-gensymbols: warning: debian/
--- debian/
+++ dpkg-gensymbols
@@ -10,6 +10,7 @@
cipher_
crypto_
crypto_
+ crypto_
crypto_init@Base 1.99.9
crypto_
cypher_
...
As you can see, when a symbol is not present dpkg-gensymbols will add it Ubuntu- specific "-0ubuntu1.2" part) when determining the symbol
on-the-fly, but using the full package version (including the
Debian/
version. This is why lintian is displaying the error.
Investigating a bit more led me to the following patch:
debian/ patches/ CVE-2018- 1084-4. patch
This is the patch that actually added the new symbol. It was added in
version 2.4.3-0ubuntu1.1; this means that it's not really possible to
use just the upstream version when specifying when the symbol first
appeared. Fortunately, the lintian extended message covers this very
same scenario:
[...] If the debian revision can't be stripped because the symbol
really appeared between two specific Debian revisions, you should
postfix the version with a single "~" (example: 1.0-3~ if the symbol
appeared in 1.0-3).
Which means that, in this case, the right thing to do would be to add pg5.symbols:
the following line to d/libtotem-
crypto_ get_current_ sec_header_ size@Base 2.4.3-0ubuntu1.1~
If you were preparing an SRU just to fix this specific issue, I'd pg5.symbols file.
probably tell you that it's not worth it. In this case, however, you're
already SRU'ing some important fixes, so IMHO it's justifiable to also
include this small fix to the d/libtotem-
Thanks,
--
Sergio
GPG key ID: E92F D0B3 6B14 F1F4 D8E0 EB2F 106D A1C8 C3CB BF14