One more attempt: stuck processes
Sebastian Hagedorn
Hagedorn at uni-koeln.de
Thu Nov 15 05:25:19 EST 2007
--On 14. November 2007 16:39:44 -0500 Ken Murchison <murch at andrew.cmu.edu>
wrote:
> It looks to me like we are timing out the client while the client is
> IDLEing, but we get a signal from idled in the middle of shutdown(). Try
> this patch.
>
> --- imapd.c.~1.535.~ 2007-11-14 16:16:21.000000000 -0500
> +++ imapd.c 2007-11-14 16:22:59.000000000 -0500
> @@ -836,6 +836,8 @@
> {
> int i;
>
> + idle_done(imapd_mailbox);
> +
> proc_cleanup();
>
> i = 0;
Thanks. I will try this patch as soon as I can, but it's clearly not the
only issue, because the same thing happens with POP processes. Here's an
example for one:
(gdb) bt
#0 0x0096441e in __read_nocancel () from /lib/tls/libc.so.6
#1 0x00ac02f7 in BIO_new_socket () from /lib/libcrypto.so.4
#2 0x00abe2b2 in BIO_read () from /lib/libcrypto.so.4
#3 0x00fcfe13 in ssl23_read_bytes () from /lib/libssl.so.4
#4 0x00fcec51 in ssl23_get_client_hello () from /lib/libssl.so.4
#5 0x00fce712 in ssl23_accept () from /lib/libssl.so.4
#6 0x00fd2c9a in SSL_accept () from /lib/libssl.so.4
#7 0x08052cb3 in shut_down ()
#8 0x0804e513 in shut_down ()
#9 0x0804d58c in ?? ()
#10 0x00000001 in ?? ()
#11 0x08dd6848 in ?? ()
#12 0x00000000 in ?? ()
As you see it looks very much like the IMAP stack trace, but obviously the
IDLE routine is not a factor. OTOH, all the stuck IMAP processes I've
looked at so far *did* have those idle_xxx lines in their traces. I am
correct in assuming that that would only occur for clients that actually
use IDLE, right?
--
.:.Sebastian Hagedorn - RZKR-R1 (Gebäude 52), Zimmer 18.:.
Zentrum für angewandte Informatik - Universitätsweiter Service RRZK
.:.Universität zu Köln / Cologne University - ✆ +49-221-478-5587.:.
.:.:.:.Skype: shagedorn.:.:.:.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 186 bytes
Desc: not available
Url : http://lists.andrew.cmu.edu/pipermail/info-cyrus/attachments/20071115/abf9944c/attachment-0001.bin
More information about the Info-cyrus
mailing list