[squid-users] Interpreting DiskD Stats

Hi Folks:

I'm running a squid-2.4.STABLE6 with security related patches release on a
redhat 7.3 box. It's using a single 8GB DiskD cache_dir, that is
hardware striped. I was looking at the diskd statistics and I'm seeing
the following

sent_count: 5889702
recv_count: 5889702
max_away: 7
max_shmuse: 7
open_fail_queue_len: 0
block_queue_len: 9370

             OPS SUCCESS FAIL
   open 367700 367700 0
 create 588864 588864 0
  close 956564 956564 0
 unlink 283164 66827 216337
   read 948716 948716 0
  write 2744694 2744694 0

I don't see any errors in cache.log, what would explain the extremely
large number of unlink failures? And would this be normal?

As a quick side question, would renicing a squid process make it perform
significantly better, and would that renice be required for the unlinkd
process and any associated child processes?

Thanks.

Brian Seppanen
seppy@chartermi.net
906-228-4226 ext 23
Received on Mon Dec 16 2002 - 09:57:32 MST