Re: squid rebuilds cache for three days

From: Pavel P. Zabortsev <[email protected]>
Date: Fri, 11 Jun 1999 10:18:43 +0400

Hi Alex,

>It looks like you have either misconfigured your kernel or stressing the
>box too much. FreeBSD will crash with the "Out of mbuf clusters" message if
>you run out of memory allocated for mbufs (via NMBCLUSTERS or maxusers
>kernel options). You can use netstat, including the -m option, to see what
>is going on with your memory allocated for network use.

netstat -m:
642 mbufs in use:
        484 mbufs allocated to data
        127 mbufs allocated to packet headers
        28 mbufs allocated to protocol control blocks
        3 mbufs allocated to socket names and addresses
461/618 mbuf clusters in use
1316 Kbytes allocated to network (76% in use)
0 requests for memory denied
0 requests for memory delayed
0 calls to protocol drain routines

By the way: squid is still rebuilding cache!
In cache.log:
1999/06/11 00:00:53| Not currently OK to rewrite swap log.
1999/06/11 00:00:53| storeDirWriteCleanLogs: Operation aborted.
1999/06/11 00:00:53| storeLogRotate: Rotating.
1999/06/11 00:00:53| accessLogRotate: Rotating
1999/06/11 04:06:30| 851968 Entries read so far.
[...skip...]

>
>$0.02,
>
>Alex.
>

Yours sincerely,
Pavel

----------------------------------------------------------------
Pavel P. Zabortsev e-mail: ppz@cdu.elektra.ru
Software engineer, Postmaster voice: 927-9523, 26-23 (corp)
CDO UPS of Russia fax: 220-6542
Moscow, Russia ICQ: 15371542
----------------------------------------------------------------
Received on Fri Jun 11 1999 - 00:06:13 MDT

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:46:50 MST