Tomek K. wrote:
Hello,
I have squid server running on Linux Debian Etch. Squid Version 2.6.STABLE5
First step is upgrade to the latest 2.6 STABLE17 from unstable. There
have been a lot of improvements made since stable5.
Amos
is using LDAP group authentication based on Win2003 domain controller.
The hardware configuration is:
IBM xSeries 336 with one XEON 3.4GHz, 4GB RAM, 2xHDD SCSI 146GB 15K with
RAID 1
The problem is that the squid restarts after memory leak. I tried to find
the reason, and I found that the reason is to big memory consumption by
squid parent proces which is almost the same as child memory usage which is
the main cache proces.
Below is the top informations about memory usage:
top - 22:41:04 up 1 day, 3:55, 1 user, load average: 0.00, 0.00, 0.00
Tasks: 112 total, 1 running, 111 sleeping, 0 stopped, 0 zombie
Cpu(s): 0.0%us, 0.0%sy, 0.0%ni, 99.8%id, 0.2%wa, 0.0%hi, 0.0%si,
0.0%st
Mem: 3502208k total, 3360836k used, 141372k free, 118448k buffers
Swap: 2650684k total, 0k used, 2650684k free, 165420k cached
%MEM PID USER PR NI VIRT RES SHR S %CPU TIME+ COMMAND
42.9 4181 proxy 15 0 1469m 1.4g 1512 S 0 0:11.81 squid
42.5 4175 root 25 0 1456m 1.4g 404 S 0 0:00.10 squid
This are the main squid.conf entries:
http_port 8080
icp_port 0
auth_param basic children 50
auth_param basic credentialsttl 1 minute
no_cache deny QUERY
cache_mem 48 MB
maximum_object_size 900000 KB
minimum_object_size 0 KB
request_body_max_size 900000 KB
cache_dir ufs /var/spool/squid 8192 128 256
cache_replacement_policy heap LFUDA
cache_store_log none
access_log /var/log/squid/access.log
forwarded_for off
delay_pools 1
delay_class 1 1
delay_access 1 allow all
delay_parameters 1 74000/74000
Is this memory usage by parent proces right ?
How can I change it ?
Regards
Tomasz Krawczyk
--
Please use Squid 2.6STABLE17 or 3.0STABLE1.
There are serious security advisories out on all earlier releases.
