[SQU] squid dies on "storeDirWriteCleanLogs"

Hello people,

I have cache_dirs of about 5GB with 200MB free on steady state.
Each machine has 1G of RAM to spend, but now only about 600MB
are in effective use.

In the last 2 days, squid dies off with the following error:

2000/09/19 22:53:20| storeDirWriteCleanLogs: \
  /var/spool/squid2/swap.state.clean: write: (28) No space left on
device
2000/09/19 22:53:20| storeDirWriteCleanLogs: \
  Current swap logfile not replaced.

How much free space must I have on each cache_dir?

Is it a mixture of the swap.state file and RAM used?

thanks,
jp

PS: I'm using 2.4.DEVEL2

--
FCCN - Fundação para a Computação Científica Nacional - Tel:
+351-218440100
--
To unsubscribe, see http://www.squid-cache.org/mailing-lists.html