GNOME Bugzilla – Bug 679319
Evolution freezes after resume on saving/sending mail
Last modified: 2013-03-14 14:47:20 UTC
just using it on a notebook taking with me. When I come back and resume to docking station, I can start working with evolution mail normally, till I try to send a mail or save one as draft. Than it is state something like "Saving state of the user interface" forever in the main window. The message window is doing the task (sending or saving) forever without finishing. I'm unable to close the mail window. The main window can be closed. Also I can not even kill evolution at all. Not even kill -9 is working. I have at least to kill the whole X session. Mostly I reboot the notebook to solve this. Best regards, xcomm P.S.: Note is a Dell, Gnome is 3.4.x on Debian sid i386 BTW: That even kill -9 is not working really makes me a little bit angry (as well as the long time Network-Manager dependency of Evolution). ;-)
As it happend again I did some strace on it, maybe it helps. This is from the 5 stace threads: output.28081 read(48, "\2\0\0\0\2\0\0\0\0\0\0\0\20\0\0\0output.28081\0\0\0\0", 1024) = 32 read(48, "\2\0\0\0\2\0\0\0\0\0\0\0\20\0\0\0output.28081\0\0\0\0", 1024) = 32 read(48, "\2\0\0\0\2\0\0\0\0\0\0\0\20\0\0\0output.28081\0\0\0\0", 1024) = 32 read(48, "\2\0\0\0\2\0\0\0\0\0\0\0\20\0\0\0output.28081\0\0\0\0", 1024) = 32 read(48, "\2\0\0\0\2\0\0\0\0\0\0\0\20\0\0\0output.28081\0\0\0\0", 1024) = 32 read(48, "\2\0\0\0\2\0\0\0\0\0\0\0\20\0\0\0output.28081\0\0\0\0", 1024) = 32 read(48, "\2\0\0\0\2\0\0\0\0\0\0\0\20\0\0\0output.28081\0\0\0\0", 1024) = 32 read(48, "\2\0\0\0\2\0\0\0\0\0\0\0\20\0\0\0output.28081\0\0\0\0", 1024) = 32 read(48, "\2\0\0\0\2\0\0\0\0\0\0\0\20\0\0\0output.28081\0\0\0\0", 1024) = 32 output.29385 restart_syscall(<... resuming interrupted call ...>) = -1 ETIMEDOUT (Connection timed out) futex(0xa265480, FUTEX_WAKE_PRIVATE, 1) = 0 madvise(0xa54ff000, 8372224, MADV_DONTNEED) = 0 _exit(0) = ? ==> output.28084 <== restart_syscall(<... resuming interrupted call ...> ==> output.28083 <== restart_syscall(<... resuming interrupted call ...> ==> output.28082 <== restart_syscall(<... resuming interrupted call ...>
kill -9 28081 strace -ff -o output -p 28081 PANIC: handle_group_exit: 28084 leader 28081 PANIC: handle_group_exit: 28083 leader 28081 PANIC: handle_group_exit: 28082 leader 28081 ==> output.28081 <== gettimeofday({1341475737, 859469}, NULL) = 0 gettimeofday({1341475737, 859486}, NULL) = 0 gettimeofday({1341475737, 859504}, NULL) = 0 gettimeofday({1341475737, 859521}, NULL) = 0 gettimeofday({1341475737, 859539}, NULL) = 0 gettimeofday({1341475737, 859556}, NULL) = 0 gettimeofday({1341475738, 23372}, NULL) = 0 gettimeofday({1341475738, 23400}, NULL) = 0 gettimeofday({1341475738, 23418}, NULL) = 0 ==> output.28082 <== restart_syscall(<... resuming interrupted call ...>+++ killed by SIGKILL +++ ==> output.28083 <== restart_syscall(<... resuming interrupted call ...>+++ killed by SIGKILL +++ ==> output.28084 <== restart_syscall(<... resuming interrupted call ...>+++ killed by SIGKILL +++ ps wauxm | grep 28081 xcomm 28081 3.6 0.0 0 0 ? - 09:42 1:00 [evolution] <defunct> ...
Thanks for a bug report. This should be addressed by bug #551788, and possible others as well, as 3.2.x is quite old now. I'm marking this as a duplicate of the other bug. *** This bug has been marked as a duplicate of bug 551788 ***
Hello Milan, It had went away for some times as you reported it closed. Anyway it is now back to me. It happens every time I did take my notebook in suspend with me and I resume it, without had it resumed in between. Every time I try to send an email this message window gets frozen and cannot be killed anymore. Despite the main Evolution window can be closed. I'm now on: ii evolution 3.4.4-1 i386 groupware suite with mail client and organizer This is Debian GNU/Linux unstable sid. Maybe the error is coming back with upstream? Kind regards, Jan
Thanks for the update. Could you get backtrace of running evolution to see where it is stuck, please? You can do that with command like this: $ gdb --batch --ex "t a a bt" -pid=PID &>bt.txt where PID is a process ID of running evolution (ps ax | grep evolution). Please make sure you've installed debug info packages at least for evolution-data-server and evolution, and that they are of the same version as binary packages, unless the backtrace will be useless. Also make sure that the backtrace will not contain any private information, like server addresses, email addresses or passwords (I usually search for "pass" at least). Thanks in advance.
Closing this bug report as no further information has been provided. Please feel free to reopen this bug if you can provide the information asked for. Thanks!