First try to reduce the amount of I/O threads to a more sane value.
Somewhere around 6-10 per cache_dir should be quite sufficient.
Also see your cache.log in case something interesting gets logged there.
If not, next time Squid hangs, run "squid -k debug" to see if there is
any clue on what Squid is doing when it hangs..
-- Henrik Nordstrom Squid hacker Mike Richards wrote: > > Folks, > Now squid running with snmp and 255 threads is > periodically hanging...I poll it every 10 minutes > via mrtg for stats..... how do I get around this HANG!!! > > thanx, > Mike > > ----- Original Message ----- > > > Henrik, > > > > No, I'd forgotten to do that....I'll change it > > and monitor the squid tomorrow... > > > > thanx, > > Mike > > > > ----- Original Message ----- > > > > > Mike Richards wrote: > > > > > > > > Henrik, > > > > > > > > squid 2.3r4 compiled with > > > > # ./configure --enable-snmp --enable-poll --enable-async-io=255 > > > > > > And have you configured squid.conf to make use of async-io? (asyncufs > > > store type). > > > > > > ps. In the current configuration with no/small cache you do not seem to > > > be short of memory. > > > > > > -- > > > Henrik Nordstrom > > > Squid hacker > > > > > -- > To unsubscribe, see http://www.squid-cache.org/mailing-lists.html -- To unsubscribe, see http://www.squid-cache.org/mailing-lists.htmlReceived on Tue Mar 06 2001 - 15:33:46 MST
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:58:32 MST