crash whilst expiring the overviewdb

[Peter Gradwell]

Hello,

We have a problem where our overview db's keeps filling up because
our nightly expire job is crashing.

We're using cyclicbuffers for our overview db [ from  buffindexed.conf ]:

0:/var/spool/news/overview/OV1:1536000
1:/var/spool/news/overview/OV2:1536000
2:/var/spool/news/overview/OV3:1536000
3:/var/spool/news/overview/OV4:1536000
4:/var/spool/news/overview/OV5:1536000
5:/var/spool/news/overview/OV6:1536000
6:/var/spool/news/overview/OV7:1536000

We get the following in the daily usenet report:

buffindexed.c:1590 Can't malloc 701472768 bytes: Cannot allocate
memoryExpire messages:
expireover start Wed Feb 14 03:08:03 GMT 2001: ( -z/var/log/news/expire.rm
-Z/var/log/news/expire.lowmark)
Article lines processed 21807
Articles dropped 0
Overview index dropped 0

Upgrading to INN 2.3.1 (on FreeBSD 4.1.1) hasn't helped, and I can't see
much reference to this sort of problem in these archives. Which is making
me wonder why no one else is seeing this problem. I don't see much change
in this area in the snap shots I checked out last week either.

The current theory is that no one else is running buffindexed on such a
large spool. We have a 50.5 gig cyclic text only spool. - But I didn't think
50.5 gig was that much news <g>

Any way, has anyone else experienced this problem? Is it a (known) bug?
Is there a patch? Or, does anyone have any pointers so that I can go
and persuade some one with copious free time to fix it!

At the moment it takes out our news server once a month (usually at
3am) which is getting a bit tedious.

thanks

peter

-- 
peter gradwell. gradwell dot com Ltd. http://www.gradwell.com/