It's been a long time since I first had this problem. I have modified,
changed, erased, added configurations but I still don't see what's the
problem. My squid consumes swap until it takes it all and the service goes
down. I have 3.0Ghz processor and 512MB of RAM to offer internet service to
more than 560 users authenticated. This is the cache configuration that I
have right now.
cache_mem 256 MB
cache_dir ufs /var/spool/squid/cache 20480 16 256
cache_access_log /var/log/squid/access.log
cache_log /var/log/squid/cache.log
cache_store_log /var/log/squid/store.log
cache_swap_low 90
cache_swap_high 95
maximum_object_size 20480 KB
maximum_object_size_in_memory 16 KB
request_body_max_size 4 MB
the last time the squid service stopped I checked the cache.log and this is
what I get:
Starting Squid Cache version 2.5.STABLE6 for i686-redhat-linux-gnu...
2007/07/13 07:46:07| Process ID 29874
2007/07/13 07:46:07| With 1024 file descriptors available
2007/07/13 07:46:07| DNS Socket created at 0.0.0.0, port 32854, FD 4
2007/07/13 07:46:07| Adding nameserver 192.168.54.3 from squid.conf
2007/07/13 07:46:07| Adding nameserver 192.168.54.4 from squid.conf
2007/07/13 07:46:07| Adding nameserver 192.168.250.1 from squid.conf
2007/07/13 07:46:07| Adding nameserver 192.168.250.3 from squid.conf
2007/07/13 07:46:07| helperOpenServers: Starting 10 'ncsa_auth' processes
2007/07/13 07:46:07| User-Agent logging is disabled.
2007/07/13 07:46:07| Referer logging is disabled.
2007/07/13 07:46:07| Unlinkd pipe opened on FD 19
2007/07/13 07:46:07| Swap maxSize 20971520 KB, estimated 1613193 objects
2007/07/13 07:46:07| Target number of buckets: 80659
2007/07/13 07:46:07| Using 131072 Store buckets
2007/07/13 07:46:07| Max Mem size: 262144 KB
2007/07/13 07:46:07| Max Swap size: 20971520 KB
2007/07/13 07:46:07| Rebuilding storage in /var/spool/squid/cache (DIRTY)
2007/07/13 07:46:07| Using Least Load store dir selection
2007/07/13 07:46:07| Set Current Directory to /var/spool/squid
2007/07/13 07:46:07| Loaded Icons.
2007/07/13 07:46:07| Accepting HTTP connections at 192.168.168.2, port 3128,
FD 21.
2007/07/13 07:46:07| Accepting HTTP connections at 192.168.54.4, port 3128,
FD 22.
2007/07/13 07:46:07| Accepting ICP messages at 0.0.0.0, port 3130, FD 23.
2007/07/13 07:46:07| WCCP Disabled.
2007/07/13 07:46:07| Configuring Parent 192.168.250.2/1080/0
2007/07/13 07:46:07| Ready to serve requests.
2007/07/13 07:46:07| Store rebuilding is 6.3% complete
2007/07/13 07:46:08| WARNING: newer swaplog entry for dirno 0, fileno
00004DFF
2007/07/13 07:46:08| Done reading /var/spool/squid/cache swaplog (66568
entries)
2007/07/13 07:46:08| Finished rebuilding storage from disk.
2007/07/13 07:46:08| 62922 Entries scanned
2007/07/13 07:46:08| 0 Invalid entries.
2007/07/13 07:46:08| 0 With invalid flags.
2007/07/13 07:46:08| 62617 Objects loaded.
2007/07/13 07:46:08| 0 Objects expired.
2007/07/13 07:46:08| 253 Objects cancelled.
2007/07/13 07:46:08| 3295 Duplicate URLs purged.
2007/07/13 07:46:08| 34 Swapfile clashes avoided.
2007/07/13 07:46:08| Took 0.8 seconds (78072.6 objects/sec).
2007/07/13 07:46:08| Beginning Validation Procedure
2007/07/13 07:46:08| Completed Validation Procedure
2007/07/13 07:46:08| Validated 59324 Entries
2007/07/13 07:46:08| store_swap_size = 890312k
2007/07/13 07:46:08| storeLateRelease: released 0 objects
2007/07/13 07:46:13| Preparing for shutdown after 0 requests
2007/07/13 07:46:13| Waiting 30 seconds for active connections to finish
2007/07/13 07:46:13| FD 21 Closing HTTP connection
2007/07/13 07:46:13| FD 22 Closing HTTP connection
2007/07/13 07:46:20| Shutting down...
2007/07/13 07:46:20| FD 23 Closing ICP connection
2007/07/13 07:46:20| Closing unlinkd pipe on FD 19
2007/07/13 07:46:20| storeDirWriteCleanLogs: Starting...
2007/07/13 07:46:20| Finished. Wrote 59326 entries.
2007/07/13 07:46:20| Took 0.0 seconds (2627834.9 entries/sec).
CPU Usage: 0.573 seconds = 0.261 user + 0.312 sys
Maximum Resident Size: 0 KB
Page faults with physical i/o: 0
Memory usage for squid via mallinfo():
total space in arena: 8052 KB
Ordinary blocks: 7934 KB 8 blks
Small blocks: 0 KB 4 blks
Holding blocks: 716 KB 2 blks
Free Small blocks: 0 KB
Free Ordinary blocks: 117 KB
Total in use: 8650 KB 107%
Total free: 117 KB 1%
I'm concerned about the swap maxsize. I don't know if I'm wrong but it says
that I have 20GB when I only have 1GB of swap available in the system
(RedHat Enterprise Linux 4).
Please I need some feedback if you can help me solve this. I don't know what
to do. My squid is consuming too much swap!!!
thanks
