debugging high volume caches

From: tom minchin <tom@dont-contact.us>
Date: Sun, 8 Aug 1999 00:46:26 +1000

Hi,
        I've got two "crash and restart" type conditions in Squid 2.2S4 on
Linux that I want to debug so I can send off the logs to squid-bugs. However,
the volume on the caches is such that:
        1) I generate about 4 gig of logs an hour
        2) debug logging slows Squid down to a near halt

        Does anyone have any suggestions on how go about debugging in this
situation? I've partially solved 1) by cp /dev/null the cache.log every 10
mins and when RunCache restarts squid after a crash it copies the current
cache.log into a safe place - it's 2) that kills me though. The errors usually
occur a day or so apart so I can't leave debugging on with 30-60 second
response times.

        The error conditions are:
        1) good old segmentation violation - Squid writes out the log and
quits
        2) assertion failed in http.c - Squid quits

tom@interact.net.au
Received on Sat Aug 07 1999 - 08:26:15 MDT

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:47:53 MST