The mailing list for listmasters using Sympa

[Olivier Salaün - CRU <address@concealed>]

Liam Kirsher wrote:

Thanks for the quick response!  I looked at the files in ~sympa/spool/tmp.

There were many entries like this:
  

[...]

Warning: Unrecognized variable: "tag"
Warning: Unrecognized variable: "tag"
Wide character in print at
/usr/lib/perl5/5.8.0/i386-linux-thread-multi/IO/Handle.pm line 395.
Wide character in print at
/usr/lib/perl5/5.8.0/i386-linux-thread-multi/IO/Handle.pm line 395.
Wide character in print at
/usr/lib/perl5/5.8.0/i386-linux-thread-multi/IO/Handle.pm line 395.
Wide character in print at
/usr/lib/perl5/5.8.0/i386-linux-thread-multi/IO/Handle.pm line 395.
Wide character in print at /home/sympa/bin/mail.pm line 621.
Wide character in print at
/usr/lib/perl5/5.8.0/i386-linux-thread-multi/IO/Handle.pm line 395.
Wide character in print at /home/sympa/bin/mail.pm line 621.
Wide character in print at
/usr/lib/perl5/5.8.0/i386-linux-thread-multi/IO/Handle.pm line 395.
Wide character in print at
/usr/lib/perl5/5.8.0/i386-linux-thread-multi/IO/Handle.pm line 395.
    

These warnings must come from the archived.pl process (I guess that  because of the first warning message regarding "tag"). Therefore it has nothing to do with sympa.pl which is the process responsible for message distribution.

If you want to learn more about those "Wide character" warnings, check http://ahinea.com/en/tech/perl-unicode-struggle.html

There was also one "Out of memory" error. However, I think that was related to another general problem running out of memory. It didn't
occur at the same time as the failure to process the spool.

Hey, are there supposed to be *two* sympa.pl processes running?
sympa      896   890  0 May25 ?        00:01:24 /usr/bin/perl -U /dev/fd/3//home/sympa/bin/wwsympa.fcgi
sympa      912   890  0 May25 ?        00:01:21 /usr/bin/perl -U /dev/fd/3//home/sympa/bin/wwsympa.fcgi
sympa      932   890  0 May25 ?        00:01:18 /usr/bin/perl -U /dev/fd/3//home/sympa/bin/wwsympa.fcgi
sympa    13885     1  0 May30 pts/0    00:01:45 /usr/bin/perl /home/sympa/bin/sympa.pl
sympa    13886     1  0 May30 pts/0    00:05:15 /usr/bin/perl /home/sympa/bin/sympa.pl
sympa    13897     1  0 May30 pts/0    00:00:03 /usr/bin/perl /home/sympa/bin/archived.pl
sympa    13922     1  0 May30 pts/0    00:01:32 /usr/bin/perl /home/sympa/bin/bounced.pl
sympa    13952     1  0 May30 pts/0    00:01:18 /usr/bin/perl /home/sympa/bin/task_manager.pl
  

The ps shows that 2 sympa.pl are running fine. Since you have 2 processes running, I guess that you have the distribution_mode set to 'fork' (see http://www.sympa.org/wiki/manual/conf-parameters/part2#distribution_mode).

At the time where you ran the PS command did you have problems regarding message processing ?
If so, please provide a "ls -l" of your spool and you corresponding sympa log entries.