Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 05 Jan 2011 16:21:27 +0000
From:      Johannes Totz <jtotz@imperial.ac.uk>
To:        freebsd-fs@freebsd.org
Subject:   fsync: giving up on dirty & panic: snapacct_ufs2: bad block
Message-ID:  <ig25q7$rt2$1@dough.gmane.org>

next in thread | raw e-mail | index | archive | help
Hi,

(beware of line breaks)

# uname -a
FreeBSD ib-sri11.XXX 8.2-PRERELEASE FreeBSD 8.2-PRERELEASE #0: Tue Dec
7 03:42:10 GMT 2010     root@ib-sri11.XXX:/usr/obj/usr/src/sys/GENERIC
amd64


I got a "fsync: giving up on dirty" before xmas, during snapshot
creation on a ufs2 volume:

Dec 19 00:00:03 ib-sri11 kernel: fsync: giving up on dirty
Dec 19 00:00:03 ib-sri11 kernel: 0xffffff00019b8ce8: tag devfs, type VCHR
Dec 19 00:00:03 ib-sri11 kernel: usecount 1, writecount 0, refcount 313
mountedhere 0xffffff0001a5da00
Dec 19 00:00:03 ib-sri11 kernel: flags (VV_COPYONWRITE)
Dec 19 00:00:03 ib-sri11 kernel: v_object 0xffffff0001aa3bd0 ref 0 pages
1238
Dec 19 00:00:03 ib-sri11 kernel: lock type devfs: EXCL by thread
0xffffff0030433460 (pid 79859)
Dec 19 00:00:03 ib-sri11 kernel: dev ufs/root


Google suggests this is harmless and the machine kept running happily
for a while. There was enough space on ufs/root and I deleted all
snapshots by hand.

Last night the box started to panic with (transcribed from screenshot):

panic: snappacct_ufs2: bad block
cpuid = 1
KDB: stack backtrace:
#0 0xffffffff805bf48c at kdb_backtrace+0x5c
#1 0xffffffff80590200 at panic+0x1b4
#2 0xffffffff8079e88e at snapacct_ufs2+0xf2
#3  <omitted by me>   at indiracct_ufs+0x1c9
#4                    at indiracct_ufs+0x27c
#5                    at expunge_ufs2+0x2ee
#6                    at ffs_snapshot+0x1e9b
#7                    at ffs_mount+0x160c
#8                    at vfs_donmount+0xf8c
#9                    at nmount+0x6a
#10                    at syscallenter+0x2d9
#11                    at syscall+0x38
#12                    at Xfast_syscall+0xe2
Uptime: 1m45s
Physical memory: 3313 MB
Dumping 278 MB:


No dumps available, the machine hangs after printing "Dumping 278 MB:".
Looks like it crashes during background fsck (while creating the
snapshot for it)?

Don't know if these two are related...
Is there anything to help debug this or is it known to be fixed in newer
versions?
Otherwise I just newfs this and restore from backup.


Cheers,

Johannes




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?ig25q7$rt2$1>