RE: [squid-users] LOTS of " WARNING: failed to unpack meta data"

From: Mark Pelkoski <[email protected]>
Date: Thu, 11 Mar 2004 08:54:39 -0700

Can ANYBODY Help me with T-shooting this???

-Mark

-----Original Message-----
From: Mark Pelkoski
Sent: Wednesday, March 10, 2004 9:18 AM
To: Elsen Marc; squid-users@squid-cache.org
Subject: RE: [squid-users] LOTS of " WARNING: failed to unpack meta
data"

>>
>> List,
>> I just built this box about a week ago. It has 4 separate
>> diskd stores for cache. I get about 20 - 30 " WARNING: failed
>> to unpack meta data" messages a minute in the cache.log file.
>> I have rebuilt the cache several times with "squid -z" that
>> did not help. I suspect I might have a faulty drive. This is
>> a production system that I can't exactly play around with.
>> What is the best way to T-shoot this? TIA
>>
>> Mark Pelkoski
>>

> - Check system (error) logging, watch for error(s) related to
> disk (sub)system.
>
> M.

My messages log is clean. I also tried commenting out each store Diskd line from my conf file and restarted Squid so it just used 3 out of the 4 disks, thinking it could be one disk that was causing the issue. It didn't matter which disk I took out of the cache pool, I still received the same error. I have a second server built EXACTLY like this one that does not have this problem. Anybody care to chime in? I need to find the root cause because my cache.log is growing very large...

-Mark
Received on Thu Mar 11 2004 - 08:54:40 MST

This archive was generated by hypermail pre-2.1.9 : Thu Apr 01 2004 - 12:00:02 MST