Feizhou wrote:
To prevent random OOM Killer Action in the case something goes wrong :D
That's never worked in the past.
Well...for a fast memory chewing bug I guess not...
But people who have stuff in place to monitor memory and swap usage will
have a chance to catch things for those that slowly chew up memory.
Mozilla chews memory slowly. The system either gets slower until it's
too painful, or it gets slower until oom goes mad.
It was pretty bad on RHL 7.3 which I used to run in 128 mbytes, and on
occasion I found Moz "using" 700 Mbytes plus.
It still seems to happen, but I generally kill if for some other reason
(eg it tends to loop and take all CPU).
Now, if I knew how to make postgresql take a little more. It's been
running, loading a table, for almost two days on my laptop. Over 1 Gbyte
RAM, pg's only using 11m res, about 20mbytes all up.
--
Cheers
John
-- spambait
1aaaaaaa@xxxxxxxxxxxxxxxx Z1aaaaaaa@xxxxxxxxxxxxxxxx
Please do not reply off-list
_______________________________________________
CentOS mailing list
CentOS@xxxxxxxxxx
http://lists.centos.org/mailman/listinfo/centos
