Hi,
Been here. It's in the FAQ under System-dependent Weirdness.
There is a simple solution, which has kept our Solaris 2.6 cache running
for months.
1) Delete your cache directories.
2) Run "tunefs -o space" on each cache partition.
That fixed it for me.
Rgds
Richard Stagg
On Fri, 21 May 1999, Shigeru Matsumoto wrote:
> Dear Squid-Users,
>
> Today and the day before yesterday our Squid2.1PATCH2 broke down.
> I restarted the process but it didn't work, and then I rebooted the
> machine. However, it didn't recover, either.
>
> Error messages of cache.log is as follows:
> 1999/05/21 11:26:35| diskHandleWrite: FD 14: disk write error: (28) No
> space left on device
> FATAL: Write failure -- check your disk space and cache.log
> Squid Cache (Version 2.1.PATCH2): Terminated abnormally.
>
> And Error messages of squid.out is:
> FATAL: Write failure -- check your disk space and cache.log
> FATAL: Write failure -- check your disk space and cache.log
>
> There are enough spaces on the disk, though.
>
> Platform is Solaris2.6, and the machine is Ultra2.
>
> And this trouble happed when the traffic is very high. These days our
> Squid2.1PATCH2
> handle 690,000/day. Allocation mamory is 192 for the Squid and I set 2
> cache disks for
> juxtaposition as follows.
>
> cache_dir /export/home/squid/squid_cache 7000 16 256
> cache_dir /var/home/squid/squid_cache 1000 16 256
>
> Does anybody know the cause of the trouble?
>
> Best,
> Shiggy
> ------------------------------------------------------------------
> Shigeru Matsumoto << E-Mail: matumoto@ssd.sony.co.jp >>
> Sony Systems Design Co.Ltd.
> ------------------------------------------------------------------
>
---------------------------------
Richard Stagg
Internet Architect
squid@bae.co.uk
Received on Fri May 21 1999 - 02:19:26 MDT
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:46:22 MST