Hello all,
I run a large site composed of "Tier 1 Internet" and "Tier 2 Intranet"
proxies, all running Squid 2.2STABLE5. The problem I have is when I
send a HUP to squid, on the Tier 1 (parent) proxies, occasionally I get
a few of the following lines in my cache.log file.
2000/01/10 15:05:42| ipcCreate: fork: (12) Not enough space
2000/01/10 15:05:42| WARNING: Cannot run
'/usr/local/squid/bin/dnsserver' process.
The machine does not seem to have any contention for Memory, CPU, etc.,
and this ONLY happens with a HUP, and then not all processes have the
problem. There does not seem to be any rhyme or reason to when, or the
number of processes failing to start. This has also started happening
with the redirector that was recently installed. As mentioned before, I
am running a large site, so both settings in squid.conf for dnsserver
and redirector are set to "32" (I have tried settings of 16-32 -- same
result). All machines are Solaris 2.6 SPARC based.
Any help would be great!
Thanks.
Received on Mon Jan 10 2000 - 15:33:31 MST
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:50:18 MST