My Squid (2.4S1 on Sun Solaris 8) failed recently with a disk full message
when it rotated the squid logs. The OS reported the same error even though
there was plently of free space and inodes according to 'df'!
Anyway, I found a note in the FAQ with a problem unique to SUN's ufs file
system and this problem. The solution according to the FAQ was to optimize
the partition for space instead of time. I've rebuilt my two cache
partitions using this technique and Squid is up and running again.
Can any Solaris/Squid users out there verify that this is indeed a
solution to the problem as opposed to just delaying it longer?
-Mike
Received on Mon Oct 22 2001 - 18:50:04 MDT
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 17:03:03 MST