What is your memcached version? Are you using persistent or single-shot connections to the memcached server? You could also look at the socket state on your servers with netstat etc))
Cheers, Trond On Feb 17, 2010, at 9:48 AM, Niro wrote: > I'm running an EC2/Scalr small app instance with the Scalr Apache. I > added APC to cache php and run the pre- installed memcache server on > the local machine. > > after a few million hits, the memcache fails with error : "Can't > connect to 127.0.0.1:11211, Connection refused (111) " and then many > "cannot find connection identifier" and occasional "Can't connect to > 127.0.0.1:11211, Connection refused (111) ". > > The server remains in this situation and never recovers unless I > restart memcached deamon. > > Any idea on how I can find the root cause and solve it is highly > appreciated. > > Server info: > from phpinfo: Apache/2.2.8 (Ubuntu) PHP/5.2.4-2ubuntu5.10 with Suhosin- > Patch mod_ssl/2.2.8 OpenSSL/0.9.8g > From cat /etc/*-release > DISTRIB_ID=Ubuntu > DISTRIB_RELEASE=8.04 > DISTRIB_CODENAME=hardy > DISTRIB_DESCRIPTION="Ubuntu 8.04.1" > >
