middelink at polyware.nl
Tue Dec 5 15:08:10 UTC 2000
(I just subscribed to this list, so be patient pls)
Recently started using inn2.3.0 instead of 2.2.3 on a small
not-full-feed machine. Its takes about 1.7GB traffic/day.
Used cfs for *jobs* and *bin* and put the rest in tradspool.
Left it to run for a week...
Oh, havegroupbaseexpire is on.
Expire takes now more than 24 hours to complete. (~48 hours)
Expireover determines in appr. 1.5 hours what to delete,
expirerm deletes it in about 3 hours and than expire takes
Looking at the strace from it, it is access(2)ing every article
in the spool (1.5mil) which takes a long time, since inn itself
is also banging away on the disk...
why is expire doing this? I assumed it want to determine which
articles got removed by expireover to update the history. Can't
it just read and use expire.rm? (The file expireover neatly
created for fastrm?)
Or am I missing something obvious?
Met vriendelijke groet,
GPG Key fingerprint = 2D5B 87A7 DDA6 0378 5DEA BD3B 9A50 B416 E2D0 C3C2
For more details look at my website http://www.polyware.nl/~middelink
More information about the inn-workers