[squid-users] excessive DNS queries

Hello, I am running a Redhat 7.2 box with the latest Redhat RPM (Squid
Cache: Version 2.4.STABLE3), and I am seeing what appears to be the internal
DNS resolver not doing any caching, or at least not using the nscd daemon
how I would expect it to be used. I downloaded the source for 2.4STABLE6,
and saw the same behavior from it.

From the research into it that I have been able to do so far, it appears
that the squid host sends multiple queries that contain the exact same
contents, for a particular host. There appears to be no caching happening.
Also, it seems that it is not an error timeout type of thing, as the queries
happen almost all at the same time.

The parts of my config file that may matter:

ipcache_size 2048
fqdncache_size 2048
debug_options ALL,1
log_fqdn on
dns_retransmit_interval 8 seconds
positive_dns_ttl 4 hours
negative_dns_ttl 5 minutes

Has anyone else experienced this, or is it an expected behavior?

thanks,

Jim
Received on Wed Apr 24 2002 - 10:30:54 MDT