>Number: 144416 >Category: kern >Synopsis: Kernel panic on online filesystem optimization >Confidential: no >Severity: non-critical >Priority: medium >Responsible: freebsd-bugs >State: open >Quarter: >Keywords: >Date-Required: >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Tue Mar 02 08:10:01 UTC 2010 >Closed-Date: >Last-Modified: >Originator: sa >Release: FreeBSD 7.1-RELEASE >Organization: New Energy >Environment: FreeBSD hq-bsd0.hq.stroyreserv.com 7.1-RELEASE FreeBSD 7.1-RELEASE #0: Thu Jan 1 14:37:25 UTC 2009 [email protected]:/usr/obj/usr/src/sys/GENERIC i386 >Description: During online filesystem optimization system can go to kernel panic on certain circumstances.
Our case was raid 0+1, ar0: 953877MB <nVidia MediaShield RAID0+1 (stripe 64 KB)> status: READY with /var mounted on /dev/ar0s1f (ufs, NFS exported, local, soft-updates, fsid 382aad491192146c) Server role is network gateway, firewall, squid-proxy, mail server and some monitoring stuff. /var is heavily used by squid cache and mail-store. Apparently, we hit online fs optimization threshold, which is (minfree-2 percent) < 6.5 (current defrag level). In messages there are messages on fs going from TIME to SPACE optimization and it went fine, but after message fs going from SPACE to TIME there is kernel panic with following data: dev = ar0s1f, block = 29238692, fs = /var panic: ffs_blkfree: freeing free frag cpuid = 0 Uptime: 320d3h29m24s Physical memory: 2035 MB Dumping 270Mb Maybe it-s already known issue, but better repost, than forget. >How-To-Repeat: Non tried as it-s production-use server. >Fix: Avoid using online filesystem optimization by good recepies given here http://lists.freebsd.org/pipermail/freebsd-stable/2007-April/034711.html , or better use of planning file system. >Release-Note: >Audit-Trail: >Unformatted: _______________________________________________ [email protected] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-bugs To unsubscribe, send any mail to "[email protected]"
