After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 602481 - POP3 receive hangs: running under GDB shows camel crash
POP3 receive hangs: running under GDB shows camel crash
Status: RESOLVED OBSOLETE
Product: evolution
Classification: Applications
Component: Mailer
3.0.x (obsolete)
Other Linux
: Normal normal
: ---
Assigned To: evolution-mail-maintainers
Evolution QA team
evolution[pop]
Depends on:
Blocks:
 
 
Reported: 2009-11-20 09:27 UTC by Jeremy Bennett
Modified: 2011-10-14 08:24 UTC
See Also:
GNOME target: ---
GNOME version: ---


Attachments
GDB output running evolution with CAMEL_DEBUG=all (536.73 KB, text/plain)
2009-11-20 09:27 UTC, Jeremy Bennett
Details
Evolution running under GDB without Camel debug (29.79 KB, text/plain)
2009-11-20 09:28 UTC, Jeremy Bennett
Details

Description Jeremy Bennett 2009-11-20 09:27:21 UTC
Created attachment 148158 [details]
GDB output running evolution with CAMEL_DEBUG=all

I find that fairly frequently POP3 email receive hangs.

The curious thing is that if it hangs (once or twice a day), it invariably hangs when fetching message 12. It doesn't always hang -sometimes it will receive large numbers of messages with not problem. This has been going on for some time, including with older versions of Evolution under Fedora 9.

I can request the transfer to be cancelled, and it will show "failed at message 12 of xxx" (but still show "Fetching Mail (...)", or it will eventually time out. Once it has timed out, send & receive will retrieve the rest of the mail (possibly hanging at message 12 again).

I've run Evolution under GDB to try to capture this bug. I have observed a failure as a broken pipe. I attach two GDB traces, one with plain evolution, one with CAMEL_DEBUG=all

  CAMEL_DEBUG=all gdb evolution 2>&1 | tee evolution-camel-debug-gdb-output.txt
  gdb evolution 2>&1 | tee evolution-gdb-output.txt

I observed that failure happened very quickly with CAMEL_DEBUG enabled. Without I had to wait 24 hours.

Let me know if I can provide any further debug information to track down this problem.

I am running Evolution 2.26.3 under 32-bit Fedora 11 Linux on a Core2 Duo PC with 2GB RAM.
Comment 1 Jeremy Bennett 2009-11-20 09:28:21 UTC
Created attachment 148159 [details]
Evolution running under GDB without Camel debug
Comment 2 André Klapper 2009-11-20 16:37:34 UTC
All the output at the beginning
warning: difference appears to be caused by prelink, adjusting expectations
already show that there's some other issues with your system.
Comment 3 Jeremy Bennett 2009-11-20 18:41:12 UTC
Acknowledged. But this just is a side-effect of the Evolution debuginfo, which GDB corrects and has no effect on the running program.

The warning shows that prelink (used for efficiency) has modified relocation information, so it doesn't match the standalone binary, with which the debuginfo is associated.

GDB sees that things are not as expected, but understands why and adjusts accordingly. The warning should only be a cause of alarm if the binary has not been prelinked.

The correct solution is for Evolution to have a comprehensive debuginfo which matches its prelinked image. However, Evolution's debuginfo is not comprehensive, and various standard debuginfos have to be installed by hand. It is some of these which cause the warnings.

It's not particularly an Evolution thing. All sorts of programs have been caught out by the arrival of prelink, and this warning occurs all over the place.

This is a standard Fedora 11 system with its standard Evolution 2.26.3. From 'uname -a':

Linux thomas 2.6.30.9-96.fc11.i686.PAE #1 SMP Tue Nov 3 23:41:33 EST 2009 i686 i686 i386 GNU/Linux

HTH
Comment 4 Jeremy Bennett 2010-03-31 07:39:15 UTC
I observe that this problem goes away if POP3 extensions are disabled.
Comment 5 Akhil Laddha 2011-10-13 10:00:34 UTC
Can you please check again whether this issue still happens in Evolution 3.0.3 or 3.2.0 and update this report by adding a comment and changing the "Version"
field? Thanks a lot.
Comment 6 Jeremy Bennett 2011-10-14 08:22:46 UTC
I can report that the problem has not been seen in Evolution 3.0.3, which I have been running for some time.

This issue can be closed.

Thanks
Comment 7 Akhil Laddha 2011-10-14 08:24:08 UTC
Thanks.