On 11Nov2012 11:59, Frantisek Hanzlik <franta@xxxxxxxxxxx> wrote:
| squid very often load CPU at near 100%, with cca 200 users and 4000
| connections (~2000 to users, 2000 to internet). Removing delay pool
| configuration has no big effect.
[...]
| It this CPU load normal, or can be there done some performance
| tunnning for it?
I can't tell you how normal it is, but last time we had a squid behaving
that badly it had corrupted its cache somehow (or it had simply got into
some pathological state). We removed the cache (with rm -r) and remade it
with squid -z, and it has been happy since. Obviously we shut the squid
down around that operation:-)
This was a reverse proxy squid, but that may be irrelevant.
--
Cameron Simpson <cs@xxxxxxxxxx>
Truly, the ultimate demonstration of the computer's utility is that it
continues to be indispensable in spite of those who run the things.
- Steve Glass, in e-mail
--
users mailing list
users@xxxxxxxxxxxxxxxxxxxxxxx
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org
