slon periodically leaks memory

[Stuart Bishop]

There are a few memory leaks in the slon process because some results of
PQexec() calls are not always cleared, in remoteListenThread_main()
(remote_listen.c) and in copy_set() (remote_worker.c) functions.

[Dquiroga (dquiroga)]

Created an attachment (id=16)
Fix-memory-leak-in-remote_worker.c-3032

[Dquiroga (dquiroga)]

Created an attachment (id=17)
Fix-memory-leak-in-remote_worker.c-3067

[Dquiroga (dquiroga)]

Created an attachment (id=18)
Fix-memory-leak-in-remote_worker.c-3111

[Cbbrowne-ca (cbbrowne-ca)]

Committed to the 1.2 branch...

http://lists.slony.info/pipermail/slony1-commit/2008-June/002305.html

I compared the CVS HEAD code; these changes do not seem necessary there, as the HEAD code has already been changed more extensively.

[Dquiroga (dquiroga)]

The leaks on remote_worker.c were fixed but the one on remote_listen.c was not.

[Stuart Bishop (stub)]

http://slony.info claims this bug was fixed in 1.2.15. Should this bug report be closed now, or are there still outstanding issues?

[Stuart Bishop (stub)]

There are a few memory leaks in the slon process because some results of
PQexec() calls are not always cleared, in remoteListenThread_main()
(remote_listen.c) and in copy_set() (remote_worker.c) functions.

[Stuart Bishop (stub)]

According to http://slony.info, this is fixed in 1.2.15 (although the upstream bug report has not yet been closed).

[Frank-huddler (frank-huddler)]

I've been running 1.2.15 on a development server since it was released, and I'm still seeing memory leaks, which leads me to believe that Damian's comment #5 is correct -- not all of the leaks have been resolved.

[Steve Langasek (vorlon)]

slony 1.2.15 is in jaunty, so marking this bug as fixed per the previous comment.

[Alex Valavanis (valavanisalex)]

Intrepid Ibex reached end-of-life on 30 April 2010 so I am closing the
report. The bug has been fixed in newer releases of Ubuntu.

[Ssinger (ssinger)]

Some more memory leak fixes have been committed to 2.0.4 and 1.2 after .21 (1.2.22 has not yet been tagged/released).

Please retest with the memory leak fixes Ulrich Weber sent in.

[Jinkuihao (jinkuihao)]

I did not find that they say the bug 52 was fixed in the release note of 1.2.22. but when i did a test with 1.2.21 and 1.2.22 in same condition, i can see the usage of memory of slony process (1.2.21) increase slowly while i can not see the same phenomenon in 1.2.22---the usage of memory was kept in the same value.

so i wonder if i made a mistake or maybe they fixed the bug in 1.2.22. is there anybody can tell me? thank you.

[Ssinger (ssinger)]

(In reply to comment #9)
> I did not find that they say the bug 52 was fixed in the release note of
> 1.2.22. but when i did a test with 1.2.21 and 1.2.22 in same condition, i can
> see the usage of memory of slony process (1.2.21) increase slowly while i can
> not see the same phenomenon in 1.2.22---the usage of memory was kept in the
> same value.
>
> so i wonder if i made a mistake or maybe they fixed the bug in 1.2.22. is there
> anybody can tell me? thank you.

You are correct.
http://git.postgresql.org/gitweb?p=slony1-engine.git;a=commit;h=f518c0636f12db60d041e2d61642a26756071954 contains additional memory leak fixes that are in 1.2.22 but not 1.2.21. This seems to have been missed from the release notes

[Jinkuihao (jinkuihao)]

Thank you for your answer. then it means the #52 can be closed?

[Rolf Leggewie (r0lf)]

Hardy has seen the end of its life and is no longer receiving any updates. Marking the Hardy task for this ticket as "Won't Fix".