You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@spamassassin.apache.org by bu...@bugzilla.spamassassin.org on 2004/04/17 19:35:28 UTC

[Bug 3279] spamd crashing?

http://bugzilla.spamassassin.org/show_bug.cgi?id=3279





------- Additional Comments From dallase@nmgi.com  2004-04-17 10:35 -------
this looks identical to the problem i was seeing in 2.63 that i was not having
in 3.0 last i tested (r9952)

from the sa-talk list a couple days ago...

################
I'm having a problem with mail containing the following received header
from ever processing.   It takes 70-80 seconds for SA 2.63 to process it
which exceeds by spamc -t60 therefore it exits 74 and tempfails it.
This happens over and over all day long.   Testing this through 3.0-SVN
the message scans in just a couple seconds...  Is there an easy fix for this on
2.63, besides enabling -L for spamd (which works).

FROM 2.63
--------------------

# cat /tmp/message
Received: from e2kdfwmcs1.fhhlc.ftn.com ([10.40.34.13]) by 
e2kmemsmtp1.ftbco.ftn.com with Microsoft SMTPSVC(5.0.2195.6713); Thu,  15 Apr
2004 11:27:07 -0500

Hi

[root@mailgw tmp]# spamassassin -V
SpamAssassin version 2.63

[root@mailgw tmp]# time spamc -t60 < /tmp/message
Received: from e2kdfwmcs1.fhhlc.ftn.com ([10.40.34.13]) by 
e2kmemsmtp1.ftbco.ftn.com with Microsoft SMTPSVC(5.0.2195.6713); Thu,  15 Apr
2004 11:27:07 -0500

Hi

real    1m46.338s
user    0m0.000s
sys     0m0.000s

FROM 3.0 SVN
--------------------

[root@spamd4 tmp]# spamassassin -V
SpamAssassin version 3.0.0-r9952

[root@spamd4 tmp]# time spamc -t60 < /tmp/message
Received: from e2kdfwmcs1.fhhlc.ftn.com ([10.40.34.13]) by 
e2kmemsmtp1.ftbco.ftn.com with Microsoft SMTPSVC(5.0.2195.6713); Thu,  15 Apr
2004 11:27:07 -0500
X-Spam-Level: **
X-Spam-Status: No, hits=2.9 required=5.0

Hi

real    0m3.819s
user    0m0.000s
sys     0m0.000s

--------------------

when i was seeing the problem, I believe it is related to a non-existent domain
or a domain that doesnt not have DNS servers answering requests, therefore spamd
was waiting for a DNS timeout from the OS..  

At least when I was experiencing the problem with my received header hanging (by
watching spamd debug log), i could not do any lookups on NS's for ftn.com,
nmap'ing port 53udp on their NS's said filtered.  It looks like they might have
a valid NS in there now.

Is putting an eval {};  on the Received header DNS lookups in order to associate
a user-defined timeout worth having?   We have razor_timeout, pyzor_timeout,
dcc_timeout,  why not a dns_timeout?

 



------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.