I am running squid 24S1 on RH Linux 6.2 kernel 2.2.14-5 and it is quite
stable, my cachemgr ' Info ' look like this under heavy loads, i am running
transparent proxy through policy routing on Cisco , using ' diskd ' as store
type.
I need to know somethings from the following output:
1- My number of requests per minute is 1219.9 , can it be improved in any
way?
2- My counters for select loop called is i think ' High ', i have read the
Henrick's reply that it happens under bad transparent setups, how it can be
improved?
3- I have also read that number of requests/sec can be reached upto 100 with
a well configured Squid box, i need to know how can i reach this and what is
i am missing in order to achieve this?
4- Are the hit ratios fine or lower?
My ' df -h ' output looks like this:
Filesystem Size Used Avail Use% Mounted on
/dev/sda6 1.4G 96M 1.3G 7% /
/dev/sda1 15M 5.7M 8.6M 40% /boot
/dev/sdb1 17G 3.5G 12G 22% /cache1
/dev/sda5 1.9G 1.3G 559M 70% /usr
/dev/sda8 13G 10G 2.1G 83% /var
/dev/sdc1 34G 23G 8.7G 73% /cache2nd
My squid box is PIII 650, with 1 GB RAM , Ultra SCSI 10K rpm drives (Three
18GB IBM+18GB IBM+36GB Seagette Cheetah).
Any suggestions to improve my box , if you need any other info i am ready to
provide. I have also read Joe's article on tunning Squid boxes at
swelltech's website.
Thanks to all ' Squid Guru's ' on providing a very good support to
squid-users.
/Naeem.
Squid Object Cache: Version 2.4.STABLE1
Start Time: Wed, 09 May 2001 18:20:41 GMT
Current Time: Thu, 10 May 2001 11:03:01 GMT
Connection information for squid:
Number of clients accessing cache: 192
Number of HTTP requests received: 1222748
Number of ICP messages received: 0
Number of ICP messages sent: 0
Number of queued ICP replies: 0
Request failure ratio: 0.00%
HTTP requests per minute: 1219.9 <-----------------------------------
ICP messages per minute: 0.0
Select loop called: 9349255 times, 6.433 ms avg
<--------------------------------
Cache information for squid:
Request Hit Ratios: 5min: 35.8%, 60min: 42.1% <--------------------
Byte Hit Ratios: 5min: 39.6%, 60min: 41.8% <--------------------
Request Memory Hit Ratios: 5min: 3.8%, 60min: 2.5%
Request Disk Hit Ratios: 5min: 69.1%, 60min: 73.7%
Storage Swap size: 38288968 KB
Storage Mem size: 8128 KB
Mean Object Size: 9.96 KB
Requests given to unlinkd: 0
Median Service Times (seconds) 5 min 60 min:
HTTP Requests (All): 1.38447 1.00114
Cache Misses: 1.62803 1.71839
Cache Hits: 0.01847 0.01955
Near Hits: 1.62803 1.62803
Not-Modified Replies: 0.00865 0.00865
DNS Lookups: 0.01535 0.01464
ICP Queries: 0.00000 0.00000
Resource usage for squid:
UP Time: 60140.491 seconds
CPU Time: 20647.860 seconds
CPU Usage: 34.33%
CPU Usage, 5 minute avg: 39.93%
CPU Usage, 60 minute avg: 42.33%
Maximum Resident Size: 0 KB
Page faults with physical i/o: 44981
Memory usage for squid via mallinfo():
Total space in arena: 389011 KB
Ordinary blocks: 383919 KB 4144 blks
Small blocks: 0 KB 0 blks
Holding blocks: 3184 KB 4 blks
Free Small blocks: 0 KB
Free Ordinary blocks: 5092 KB
Total in use: 387103 KB 100%
Total free: 5092 KB 1%
Memory accounted for:
Total accounted: 300900 KB
memPoolAlloc calls: 162941971
memPoolFree calls: 151332375
File descriptor usage for squid:
Maximum number of file descriptors: 8192
Largest file desc currently in use: 1196
Number of file desc currently in use: 1060
Files queued for open: 0
Available number of file descriptors: 7132
Reserved number of file descriptors: 100
Store Disk files open: 0
Internal Data Structures:
3844680 StoreEntries
2378 StoreEntries with MemObjects
1089 Hot Object Cache Items
3843346 on-disk objects
Received on Thu May 10 2001 - 05:38:35 MDT
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:59:54 MST