This turned out to be the situation. I'm going to implement the "free"
solution and see if that fixes it.
Otherwise, time to look at Veritas...
Rik
jredrado@edunet.es on 07/14/99 09:04:27 AM
To: rik.flor@abbott.com @ INTERNET
cc: squid-users@ircache.net @ INTERNET
Subject: RE: Strange Error -- Out of disk space?
Your file system's fragmentation is high.
I found this problem on SPARC Solaris 2.5.
Read this.
http://squid.nlanr.net/Squid/FAQ/FAQ-14.html#ss14.1
Jeszs Redrado
EduNet
-----Mensaje original-----
De: Flor,Riklef <rik.flor@abbott.com>
Para: squid-users@ircache.net <squid-users@ircache.net>
Fecha: Wednesday, July 14, 1999 3:53 PM
Asunto: Strange Error -- Out of disk space?
>This morning, I came in the office to find Squid looping trying to restart
>itself. Looking in cache.log, I find the following entries,
>repeated constantly as Squid tries to restart (see below).
>
>Thing is...if I do a "df -k", both my cache directories still had 800 MB
free
>each...so why is Squid complaining that it has run out
>of disk space?
>
>In order to get it back up and running, I ended up blowing away the cache
>directories and starting from scratch.
>
>Thanks for any suggestions...
>Rik
>
>1999/07/14 07:21:06| Starting Squid Cache version 2.2.STABLE3 for
>i386-pc-solari
>s2.6...
>1999/07/14 07:21:06| Process ID 13775
>1999/07/14 07:21:06| With 1024 file descriptors available
>1999/07/14 07:21:06| Performing DNS Tests...
>1999/07/14 07:21:06| Successful DNS name lookup tests...
>1999/07/14 07:21:06| helperOpenServers: Starting 10 'dnsserver' processes
>1999/07/14 07:21:06| Unlinkd pipe opened on FD 16
>1999/07/14 07:21:06| Swap maxSize 8192000 KB, estimated 630153 objects
>1999/07/14 07:21:06| Target number of buckets: 12603
>1999/07/14 07:21:06| Using 16384 Store buckets, replacement runs every 5
>seconds
>1999/07/14 07:21:06| Max Mem size: 131072 KB
>1999/07/14 07:21:06| Max Swap size: 8192000 KB
>1999/07/14 07:21:06| Rebuilding storage in Cache Dir #0 (DIRTY)
>1999/07/14 07:21:06| Rebuilding storage in Cache Dir #1 (DIRTY)
>1999/07/14 07:21:06| Set Current Directory to /usr/local/squid/cache
>1999/07/14 07:21:06| Loaded Icons.
>1999/07/14 07:21:06| Accepting HTTP connections on port 80, FD 41.
>1999/07/14 07:21:06| Accepting ICP messages on port 3130, FD 42.
>1999/07/14 07:21:06| Ready to serve requests.
>1999/07/14 07:21:06| Configuring Parent weblock.cmis.abbott.com/80/7
>1999/07/14 07:21:08| 65536 Entries read so far.
>1999/07/14 07:21:10| 131072 Entries read so far.
>1999/07/14 07:21:13| 196608 Entries read so far.
>1999/07/14 07:21:15| 262144 Entries read so far.
>1999/07/14 07:21:18| 327680 Entries read so far.
>1999/07/14 07:21:21| 393216 Entries read so far.
>1999/07/14 07:21:21| diskHandleWrite: FD 17: disk write error: (28) No
space
>lef
>t on device
>FATAL: Write failure -- check your disk space and cache.log
>Squid Cache (Version 2.2.STABLE3): Terminated abnormally.
>CPU Usage: 14.210 seconds
>Maximum Resident Size: 0 KB
>Page faults with physical i/o: 0
>Memory usage for squid via mallinfo():
> total space in arena: 40705 KB
> Ordinary blocks: 2987 KB 1608 blks
> Small blocks: 37343 KB 797300 blks
> Holding blocks: 186 KB 7973 blks
> Free Small blocks: 127 KB
> Free Ordinary blocks: 61 KB
> Total in use: 40517 KB 100%
> Total free: 188 KB 0%
>
Received on Wed Jul 14 1999 - 08:33:51 MDT
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:47:23 MST