storeSwapOutFileOpened error

From: barneys <barneys@dont-contact.us>
Date: Tue, 28 Sep 99 12:28:21 +1000

We are running squid version 2.2stable3 on a machine with redhat linux 6.0
kernel version 2.2.5-15. We have only running squid for 6 weeks and and the
cache hit the cache_swap_high 95 about 2 weeks ago. A few days ago the
following error appeared in the log:

 1999/09/12 13:30:30| storeSwapOutFileOpened: Unable to open swapfile:
/scsi1/01/4E/00014EE% (27) file too large

This is message just kept repeating until the squid process crashed. We are running 2 brand new IBM 4.5gB scsi for the cache. When we reset the box linux wouldn't restart, first it told us that there was a file which was the wrong size on scsi1 and repaired it.

Then it told us that there was too many errors on scsi2 to repair and to manually run fsck on scsi2 which we did. It reported that it fixed all the errors. We then rebooted the box. Squid restarted but is now not running correctly. The cache_swap_low is 90 and the cache_swap_high is 95, but now the store swap size stays at 91%. Also the squid process used to run at 114mB, it now stays at 76mB. Whilst squid is still serving requests, it is definately not running the way it used to a week ago.

Does anyone know what went wrong and how we can fix the problem. If you
need more info please ask.

Many Thanks
Craig Morgan
Barneys Online Communications
Received on Mon Sep 27 1999 - 20:37:46 MDT

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:48:34 MST