Evolution stays running in the background when mailto: link used

Thanks for your bug. What value has the "/desktop/gnome/url-handlers/mailto/command" gconf key? Does running the same command with "mailto:<email address hidden>" from a command line does the same? Do you have anything about that to ~/.xsession-errors? Do you use the standard GNOME desktop?

Your bug lacks information we would need to investigate further. We
are now going to close the bug - please reopen if you have more
information at hand.

(NB: this bug shows how broken the Ubuntu bug reporting system is. The original reporter left a perfectly good bug report, and someone set it to invalid without fixing it. Two and a half years later, we still have the same bug!)

I confirmed this problem on Intrepid Alpha, evolution 2.23.91-0ubuntu1. When evolution is started as a mailto: URI handler (from Firefox, for example) it never exits. Not only that, but after it fails to exit, it chews infinite amounts of CPU. Here's my ps axwww listing:

jwb 21470 1 1 15:27 ? 00:00:18 evolution mailto:<email address hidden>?subject=IBM%20X60%20Thinkpad%20w/Verizon%20EVDO%20&%3b%20many%20extras!%20-%20$749%20(noe%20valley)

You can see that evolution was started to handle a mailto: URI. You can also see that it has 18 CPU seconds charged against it, going up constantly. Let's use strace to see what it's doing:

$ strace -s 1024 -f -p 21470
Process 21583 attached with 3 threads - interrupt to quit
[pid 21583] restart_syscall(<... resuming interrupted call ...> <unfinished ...>
[pid 21470] restart_syscall(<... resuming interrupted call ...> <unfinished ...>
[pid 21581] restart_syscall(<... resuming interrupted call ...>

Nothing in particular. Maybe GDB can tell us something:

(gdb) info threads
  3 Thread 0xb2cf9b90 (LWP 21581) 0xb7f7e430 in __kernel_vsyscall ()
  2 Thread 0xb24f8b90 (LWP 21583) 0xb7f7e430 in __kernel_vsyscall ()
  1 Thread 0xb633c940 (LWP 21470) 0xb7f7e430 in __kernel_vsyscall ()
(gdb) thread 1
[Switching to thread 1 (Thread 0xb633c940 (LWP 21470))]#0 0xb7f7e430 in __kernel_vsyscall ()
(gdb) where
#0 0xb7f7e430 in __kernel_vsyscall ()
#1 0xb695fde7 in poll () from /lib/tls/i686/cmov/libc.so.6
#2 0xb6a24c32 in ?? () from /usr/lib/libglib-2.0.so.0
#3 0xb6a252c2 in g_main_loop_run () from /usr/lib/libglib-2.0.so.0
#4 0xb7516cf3 in bonobo_main () from /usr/lib/libbonobo-2.so.0
#5 0x0805ed13 in main ()
(gdb) thread 2
[Switching to thread 2 (Thread 0xb24f8b90 (LWP 21583))]#0 0xb7f7e430 in __kernel_vsyscall ()
(gdb) where
#0 0xb7f7e430 in __kernel_vsyscall ()
#1 0xb695fde7 in poll () from /lib/tls/i686/cmov/libc.so.6
#2 0xb6a24c32 in ?? () from /usr/lib/libglib-2.0.so.0
#3 0xb6a252c2 in g_main_loop_run () from /usr/lib/libglib-2.0.so.0
#4 0xb74c87f0 in ?? () from /usr/lib/libORBit-2.so.0
#5 0xb6a4c05f in ?? () from /usr/lib/libglib-2.0.so.0
#6 0xb78f74ff in start_thread () from /lib/tls/i686/cmov/libpthread.so.0
#7 0xb696a64e in clone () from /lib/tls/i686/cmov/libc.so.6
(gdb) thread 3
[Switching to thread 3 (Thread 0xb2cf9b90 (LWP 21581))]#0 0xb7f7e430 in __kernel_vsyscall ()
(gdb) where
#0 0xb7f7e430 in __kernel_vsyscall ()
#1 0xb695fde7 in poll () from /lib/tls/i686/cmov/libc.so.6
#2 0xb6a24c32 in ?? () from /usr/lib/libglib-2.0.so.0
#3 0xb6a252c2 in g_main_loop_run () from /usr/lib/libglib-2.0.so.0
#4 0xb7ae6ee2 in ?? () from /usr/lib/libebook-1.2.so.9
#5 0xb6a4c05f in ?? () from /usr/lib/libglib-2.0.so.0
#6 0xb78f74ff in start_thread () from /lib/tls/i686/cmov/libpthread.so.0
#7 0xb696a64e in clone () from /lib/tls/i686/cmov/libc.so.6

Not too informative. Now to answer your...

After the mail is sent (and the Evolution compose window disappears) xwininfo -root -all still shows that Evolution continues to have an open window:

$ xwininfo -root -all | grep -i evo
     0x180003b "evolution": () 10x10+-100+-100 +-100+-100
     0x1800001 "evolution": ("evolution" "Evolution") 10x10+10+10 +10+10

So evolution has started a 10x10 window which is off-screen at -100 x -100. This is why it keeps running.

$ xkill -id 0x180003b
xkill: killing creator of resource 0x180003b

This causes evolution to finally exit.

setting to triaged. Thank you, Jeffrey, for opening the upstream bug.

This will be fixed when Evolution got rid of Bonobo in GNOME 2.29.*

Good to know! Does that mean there's something wrong with the launchpad bug watcher? Upstream closed this bug 20 days ago.

Well, it's not the bugwatcher directly, but since there are thousands of bugs that all needs to be synced from the GNOME bugtracking system, the watcher often gets server timeouts. This sometimes results in broken bugwatchers.

marking this fixed for now as 2.30 is in Maverick.