Intermittent procmail error

Neil Bird neil at fnxweb.com
Thu Jul 9 14:56:08 UTC 2009


   On my F10 box I've started (some weeks ago, I think) seeing a fault 
develop which spews the following into /var/log/maillog [edited]:


2009-06-22T00:13:19.457776+01:00 host fetchmail[11220]:  not flushed
2009-06-22T00:13:19.656500+01:00 host procmail[17122]: Insufficient 
privileges to deliver to "user"
2009-06-22T00:13:19.657024+01:00 host fetchmail[11220]: MDA returned nonzero 
status 77
2009-06-22T00:13:19.657161+01:00 host fetchmail[11220]:  not flushed
2009-06-22T00:13:19.855121+01:00 host procmail[17123]: Insufficient 
privileges to deliver to "user"
2009-06-22T00:13:19.855565+01:00 host fetchmail[11220]: MDA returned nonzero 
status 77
2009-06-22T00:13:19.855670+01:00 host fetchmail[11220]:  not flushed
2009-06-22T00:13:20.249777+01:00 host procmail[17124]: Insufficient 
privileges to deliver to "user"
2009-06-22T00:13:20.250475+01:00 host fetchmail[11220]: MDA returned nonzero 
status 77
2009-06-22T00:13:20.250592+01:00 host fetchmail[11220]:  not flushed
2009-06-22T00:13:20.552370+01:00 host procmail[17127]: Insufficient 
privileges to deliver to "user"
2009-06-22T00:13:20.553004+01:00 host fetchmail[11220]: MDA returned nonzero 
status 77
2009-06-22T00:13:20.553138+01:00 host fetchmail[11220]:  not flushed
... etc.

   This prevents anything being delivered until I clear it.

   I have taken to restarting the fecthmail & sendmail services to fix this, 
and in fact now just do this as a matter of course every night with cron. 
But that's a bit naff, I want to know what's happening.


   I use fetchmail to get mail from various places, that pipes to sendmail 
which at some point invokes procmail, which I've configured to dump mail 
into /var/spool/mail.  This has worked OK for pretty much ever.

   I then have dovecot serving this via POP3 and IMAP.  The error pretty 
much always occurs for my wife's email, and the difference between hers and 
mine is that her Thunderbird is picking email up via POP3 (as it's easier to 
file locally).

   I *think* it may be happening when she leaves herself logged in with TB 
running, with a refresh poll of 1 minute;  maybe the spool gets permanently 
file-locked?  I have since reduced this to 2 min., and I *think* it's 
reduced the frequency of occurrences (my cron job also greps maillog for new 
errors).


   So ... Any ideas?

-- 
[neil at fnx ~]# rm -f .signature
[neil at fnx ~]# ls -l .signature
ls: .signature: No such file or directory
[neil at fnx ~]# exit




More information about the fedora-list mailing list