tcygne wrote:
There is no "FATAL: assertion failed " in cache.log. I think I'm
misrepresenting the problem when I say "crashing". I don't think the squid
server is crashing. I think it just stops handing out data to the clients.
From the client point of view it looks as if the websites never load. It
just continues to spin. A ps -A shows squid still up. An /etc/init.d/squid
stop and then squid -z and then /etc/init.d/squid start will cause the
server to start handing out data again, for a little while (no more than 10
minutes), and clients get web pages. But then stops again. I've had to take
this server off line (it is still up, but has no clients going through it)
until I can figure out what went wrong.
I set up 2 of these web cache proxies at the exact same time with the exact
same hardware specs. And the exact same configuration. (other than IP
information) One is on the 10.60.0.0 subnet and one on the 10.80.0.0 subnet.
2 different locations on the WAN. One of them is still chugging along great.
Okay. Something about that instance then.
If you know or can find out how to use a utility called strace. It can
probably show exactly what Squid is doing during this lost time.
It's usually some sort of garbage collection or cache rebuilding. Though
there are a few security bugs long fixed which can result in hanging
when processing certain protocol requests/replies.
Amos
--
Please be using
Current Stable Squid 2.7.STABLE7 or 3.0.STABLE21
Current Beta Squid 3.1.0.15
