Hi!
I am trying to get req rate greater than 150req/sec. For that I
have client robots that send requests at a much higher rate. But somehow
the request rate never goes above 90-100 req/sec. Squid starts to run out
of FD, and complains about too many open files in the system, but there
are enough no. of FD. I am using Squid2.4-Stable2 with
diskd.
Can anyone tell the reason for that, and how to overcome it.
the cache manager info and a part of the cache.log is given below.
thank you
-pranav
---------------------------------------------------------------------
HTTP/1.0 200 OK
Server: Squid/2.4.STABLE2
Mime-Version: 1.0
Date: Mon, 12 Nov 2001 22:12:06 GMT
Content-Type: text/plain
Expires: Mon, 12 Nov 2001 22:12:06 GMT
Last-Modified: Mon, 12 Nov 2001 22:12:06 GMT
X-Cache: MISS from netlab16.cs.uh.edu
Proxy-Connection: close
Squid Object Cache: Version 2.4.STABLE2
Start Time: Mon, 12 Nov 2001 22:08:06 GMT
Current Time: Mon, 12 Nov 2001 22:12:06 GMT
Connection information for squid:
Number of clients accessing cache: 2
Number of HTTP requests received: 21803
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: 5445.7
ICP messages per minute: 0.0
Select loop called: 19550 times, 12.288 ms avg
Cache information for squid:
Request Hit Ratios: 5min: 12.8%, 60min: 12.8%
Byte Hit Ratios: 5min: -73.5%, 60min: -73.5%
Request Memory Hit Ratios: 5min: 79.0%, 60min: 79.0%
Request Disk Hit Ratios: 5min: 10.3%, 60min: 10.3%
Storage Swap size: 63336 KB
Storage Mem size: 45552 KB
Mean Object Size: 6.02 KB
Requests given to unlinkd: 0
Median Service Times (seconds) 5 min 60 min:
HTTP Requests (All): 4.07741 4.07741
Cache Misses: 4.30363 4.30363
Cache Hits: 0.01035 0.01035
Near Hits: 3.11263 3.11263
Not-Modified Replies: 0.00000 0.00000
DNS Lookups: 0.03868 0.03868
ICP Queries: 0.00000 0.00000
Resource usage for squid:
UP Time: 240.223 seconds
CPU Time: 161.790 seconds
CPU Usage: 67.35%
CPU Usage, 5 minute avg: 67.33%
CPU Usage, 60 minute avg: 67.33%
Maximum Resident Size: 0 KB
Page faults with physical i/o: 435
Memory usage for squid via mallinfo():
Total space in arena: 106069 KB
Ordinary blocks: 106062 KB 1 blks
Small blocks: 0 KB 0 blks
Holding blocks: 2756 KB 1 blks
Free Small blocks: 0 KB
Free Ordinary blocks: 7 KB
Total in use: 108818 KB 103%
Total free: 7 KB 0%
Memory accounted for:
Total accounted: 90879 KB
memPoolAlloc calls: 2309317
memPoolFree calls: 1979416
File descriptor usage for squid:
Maximum number of file descriptors: 16384
Largest file desc currently in use: 6793
Number of file desc currently in use: 6346
Files queued for open: 0
Available number of file descriptors: 10038
Reserved number of file descriptors: 9710
Store Disk files open: 0
Internal Data Structures:
14537 StoreEntries
12876 StoreEntries with MemObjects
9699 Hot Object Cache Items
10515 on-disk objects
--------------------------------------------------------------------------
cache.log -----------------
2001/11/12 16:08:23| storeDirWriteCleanLogs: Starting...
2001/11/12 16:08:24| Finished. Wrote 0 entries.
2001/11/12 16:08:24| Took 0.0 seconds ( 0.0 entries/sec).
2001/11/12 16:08:24| logfileRotate: /usr/local/squid/logs/access.log
2001/11/12 16:08:41| CACHEMGR: <unknown>@127.0.0.1 requesting 'info'
2001/11/12 16:08:55| CACHEMGR: <unknown>@127.0.0.1 requesting 'info'
2001/11/12 16:09:21| CACHEMGR: <unknown>@127.0.0.1 requesting 'info'
2001/11/12 16:09:31| CACHEMGR: <unknown>@127.0.0.1 requesting 'info'
2001/11/12 16:09:33| httpAccept: FD 8: accept failure: (23) Too many open
files in system
2001/11/12 16:09:33| storeSwapOutFileClosed: dirno 0, swapfile 00001B9C,
errflag=-1
(0) Success
2001/11/12 16:09:33| storeSwapOutFileClosed: dirno 0, swapfile 00001B9D,
errflag=-1
(0) Success
2001/11/12 16:09:33| storeSwapOutFileClosed: dirno 0, swapfile 00001B9F,
errflag=-1
(0) Success
2001/11/12 16:09:33| storeSwapOutFileClosed: dirno 0, swapfile 00001BA0,
errflag=-1
(0) Success
2001/11/12 16:09:33| comm_open: socket failure: (23) Too many open files
in system
2001/11/12 16:09:33| Reserved FD adjusted from 100 to 9710 due to failures
2001/11/12 16:09:33| comm_open: socket failure: (23) Too many open files
in system
2001/11/12 16:09:33| comm_open: socket failure: (23) Too many open files
in system
2001/11/12 16:09:33| storeSwapOutFileClosed: dirno 0, swapfile 00001BA1,
errflag=-1
(32) Broken pipe
2001/11/12 16:09:33| WARNING! Your cache is running out of filedescriptors
2001/11/12 16:09:34| ftpTimeout: timeout in SENT_PASV state
2001/11/12 16:09:42| ftpTimeout: timeout in SENT_PASV state
2001/11/12 16:09:49| WARNING! Your cache is running out of filedescriptors
2001/11/12 16:09:52| ftpTimeout: timeout in SENT_PASV state
2001/11/12 16:09:55| CACHEMGR: <unknown>@127.0.0.1 requesting 'info'
2001/11/12 16:09:58| ftpTimeout: timeout in SENT_PASV state
2001/11/12 16:10:05| WARNING! Your cache is running out of filedescriptors
2001/11/12 16:10:15| ftpTimeout: timeout in SENT_PASV state
2001/11/12 16:10:21| WARNING! Your cache is running out of filedescriptors
2001/11/12 16:10:37| WARNING! Your cache is running out of filedescriptors
2001/11/12 16:10:53| WARNING! Your cache is running out of filedescriptors
2001/11/12 16:11:11| WARNING! Your cache is running out of filedescriptors
2001/11/12 16:11:14| CACHEMGR: <unknown>@127.0.0.1 requesting 'info'
2001/11/12 16:11:24| comm_open: socket failure: (23) Too many open files
in system
2001/11/12 16:11:24| comm_open: socket failure: (23) Too many open files
in system
*******************************************************************
Pranav A. Desai
4309 Pease Street, Apt #6
Houston, TX - 77023
U.S.A.
Home :- (713) 926-1045
*******************************************************************
Received on Mon Nov 12 2001 - 15:24:54 MST
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 17:04:03 MST