Expire crashes after eating memory

Daniele Orlandi daniele at orlandi.com
Mon Aug 21 11:10:29 UTC 2000



Hello,

I'm using squid-BETA-20000819, on a Linux 2.0 machine configured with
CNFS and BuffIndexed overview. I receive a full feed (120 GB/day) on a
300 GB spool.

It has been up for a month or so, but in the last three days, expired
crashed every time after eating a lot of memory leaving me with
history.n and a continuously growing history (1.3 GB).

  PID USER     PRI  NI  SIZE  RSS SHARE STAT  LIB %CPU %MEM   TIME
COMMAND
14913 news      16   0  411M 408M 69476 R       0 63.4 81.0  18:27
expire

expire.log doesn't appear to say anything useful:

expireover start Mon Aug 21 03:03:35 CEST 2000: (
-z/opt/inn/log/expire.rm -Z/opt/inn/log/expire.lowmark)
expireover end Mon Aug 21 05:16:34 CEST 2000
lowmarkrenumber begin Mon Aug 21 05:16:34 CEST 2000:
(/opt/inn/log/expire.lowmark)
lowmarkrenumber end Mon Aug 21 05:16:35 CEST 2000
expire begin Mon Aug 21 05:17:06 CEST 2000: (-v1)
    Article lines processed  8050036
    Articles retained        6261573
    Entries expired          1788463
    Old entries dropped      6020897
    Old entries retained     4916249
expire end Mon Aug 21 12:15:43 CEST 2000
        all done Mon Aug 21 12:15:43 CEST 2000

I suppose that something in my spool is corrupted and makes expire
crash, but how can I fix this without loosing my spool ?

Thanks in advance!

-- 
 Daniele

-------------------------------------------------------------------------------
 Daniele Orlandi - Utility Line Italia - http://www.orlandi.com
 Via Mezzera 29/A - 20030 - Seveso (MI) - Italy
-------------------------------------------------------------------------------



More information about the inn-workers mailing list