I would bet that an strace on the master pid would reveal that you
are calling poll, not epoll.
There are several postings on the list about applying the epoll
patch, but IIRC, you need to explicitly --disable-poll --disable-
select --enable-epoll for it to work.
-Mike
On Mar 27, 2006,
This relates to the number of simultaneous requests squid is handling
-- I'm assuming you are using AUFS.
Basically, the IO threads are not processing fast enough and the io
request queue is getting long. Fast enough is a metric defined by
squid (it's in store_dir.c file I think, and there
All of the machines have had a full process restart and they are
still experiencing the same problems, so it looks as if
half_closed_clients wasn't the source of the problem.
-Mike
On Feb 22, 2006, at 1:12 PM, Mike Solomon wrote:
I added this line to the config on two of my hosts
, 2006, at 2:47 PM, Henrik Nordstrom wrote:
tis 2006-02-14 klockan 22:31 -0800 skrev Mike Solomon:
This would be fantastic, but the machines fall over after several
hours. I have 4 machines, each configured identically. They last a
few hours - they slowly consume more and more cpu, all in user space
Can you send us your squid.conf file?
Sure thing. Here is the file in its entirety.
-Mike
cache_dir aufs /cache1 24000 8 256
cache_dir aufs /mnt/drive2/cache2 24000 8 256
store_dir_select_algorithm round-robin
cache_access_log none
cache_log /home/squid/logs/cache.log
cache_store_log none
Hi all,
I've been having some problems with squid in reverse proxy mode. I
will preface this by saying that I realize I'm pushing squid to the
limits and I've been following the recent discussions on
performance. I'll try to be as clear as possible to expose the crux
of my problem, but
