On Tue, 17 Jan 2006, Joost de Heer wrote:
One of our proxies has a problem which causes the response time to explode.
We've been unable to find a cause for this behaviour, but I want to implement
a workaround: when the median response time grows over 1 second (normal
behaviour is a median response time of about 100/110ms) I want to
automatically restart the proxy.
It's better to spend some time on identifying why this happens. Would give
you a much more precise tool on when Squid needs to be restarted.
Possible causes
- Memory shortage. Montior with vmstat/sar.
- I/O bottleneck. Monitor with vmstat/iostat/sar.
- Overload. Monitor CPU usage with vmstat/iostat/sar, and filedescriptor
usage in Squid using snmp polling. Also keep an eye on cache.log.
Does anyone have something like this already implemented
I have a similar thing.. restarting Squid if the event loop takes longer
than X seconds to complete. Not exacly what you ask for.
or do I need to write my own script that does this? I'd hate reinventing
the wheel, since I'm not really a script builder....
To do exacly what you ask for you probably have to write your own script.
Regards
Henrik
