Re: Opendkim receiving SIGABRT

From: Marcelo Vieira <marcelo_at_digirati.com.br>
Date: Tue, 04 Sep 2012 13:45:50 -0300

MSK,

I made the configuration you suggested, setting core dumps enabled and
using debug binary.
The crash did not occur for a few days, but recently we noticed it
happened again. I did not find any additional information about this,
and there was not any dump created. Am I missing something?

We also noticed that at the time de process crashes, in the syslog there
was a lot of log entries for bad singature data, Is this the case when
when the message was modified after signing and before verifying?:

     opendkim[24429]: CE60C8007E: bad signature data

I was unable to retrieve the messages that caused this warning.

The output for opendkim -V is as follows:

opendkim: OpenDKIM Filter v2.5.2
     Compiled with OpenSSL 0.9.8k 25 Mar 2009
     SMFI_VERSION 0x1000001
     libmilter version 1.0.1
     Supported signing algorithms:
         rsa-sha1
         rsa-sha256
     Supported canonicalization algorithms:
         relaxed
         simple
     Active code options:
         USE_DB
         USE_ERLANG
         USE_LUA
         USE_UNBOUND
         _FFR_REPLACE_RULES
         _FFR_SELECTOR_HEADER
         _FFR_STATS
     libopendkim 2.5.2:

Is there something I can do to find out why would this is causing the
crash of the process?


On 24-08-2012 10:34, Murray S. Kucherawy wrote:
> On Fri, 24 Aug 2012, Marcelo Vieira wrote:
>> But sometimes, I have observed that service opendkim is dying. In
>> Postfix and system logs, I found no information that would indicate
>> the source of the problem. At those moments, the performance of
>> Postfix is severely impaired. Through a 'strace' command on the
>> opendkim daemon, I get the following output:
>>
>> 2254 read(20, "", 1024) = 0
>> 2254 close(20) = 0
>> 2254 rt_sigprocmask(SIG_UNBLOCK, [ABRT], NULL, 8) = 0
>> 2254 tgkill(2213, 2254, SIGABRT) = 0
>> 2254 --- SIGABRT (Aborted) _at_ 0 (0) ---
>> 2214 +++ killed by SIGABRT +++
>> 2217 +++ killed by SIGABRT +++
>> 6858 +++ killed by SIGABRT +++
>> 6950 +++ killed by SIGABRT +++
>> 2218 +++ killed by SIGABRT +++
>> 2213 +++ killed by SIGABRT +++
>>
>> Any ideas where I should check to find what would be causing this
>> SIGABRT kill?
>
> That's almost always an assertion failure, which means something
> unexpected is happening. The most helpful thing to do is ensure you
> have core dumps enabled (set EnableCoreDumps on Linux), are running a
> debug binary (pass --enable-debug to ./configure), and when it crashes
> use gdb to extract a stack trace so we can figure out where it was
> executing when this happened. Also valuable is a copy of a message
> that causes this failure to occur.
>
> For starters, what's the output of "opendkim -V"?
>
> -MSK
Received on Tue Sep 04 2012 - 16:46:09 PST

This archive was generated by hypermail 2.3.0 : Mon Oct 29 2012 - 23:20:43 PST