buffindexed expireover
Katsuhiro Kondou
kondou at nec.co.jp
Tue Oct 12 04:45:11 UTC 1999
In article <199910112239.RAA23598 at darkside.norand.com>,
Heath Kehoe <hakehoe at norand.com> wrote;
} Is anyone else having a problem with expireover taking a long time
} to run with buffindexed? On my server with 24G of cnfs spool and 4G
On my testing server(solaris2.5.1) it takes about 2 hours.
expire begin Tue Oct 12 01:03:39 JST 1999: (-v1 -z/usr/lib/news/log/expire.rm)
Article lines processed 1752108
Articles retained 1184403
Entries expired 567705
Articles dropped 0
Old entries dropped 566546
Old entries retained 106
expire end Tue Oct 12 01:08:46 JST 1999
all done Tue Oct 12 01:08:46 JST 1999
expireover start Tue Oct 12 01:08:49 JST 1999
expireover end Tue Oct 12 02:53:37 JST 1999
lowmarkrenumber begin Tue Oct 12 02:53:37 JST 1999: (/usr/lib/news/log/expire.lowmark)
lowmarkrenumber end Tue Oct 12 02:53:42 JST 1999
But I have another server(solaris2.5.1) which uses RAID5
for article and buffindexed spool and it takes too much
longer.
expire begin Tue Oct 12 03:18:21 JST 1999: (-v1 -z/usr/lib/news/log/expire.rm)
Article lines processed 1956085
Articles retained 1944815
Entries expired 11270
Articles dropped 0
Old entries dropped 77654
Old entries retained 149
expire end Tue Oct 12 03:26:35 JST 1999
all done Tue Oct 12 03:26:35 JST 1999
expireover start Tue Oct 12 03:26:35 JST 1999
expireover end Tue Oct 12 10:23:36 JST 1999
lowmarkrenumber begin Tue Oct 12 10:23:56 JST 1999: (/usr/lib/news/log/expire.lowmark)
lowmarkrenumber end Tue Oct 12 10:24:05 JST 1999
} There are two major problems here... one is that this process seems
} inefficient. Could we make it so that instead of re-writing the records,
} it makes modifications in-place? So that expired article records are
} removed, but the remaining records are left in-place.
But buffindexed keeps room for expired article in the
case. How do you think it will be purged?
--
Katsuhiro Kondou
More information about the inn-workers
mailing list