Date: Mon, 2 May 2005 03:38:49 +0100 From: "Steven Hartland" <killing@multiplay.co.uk> To: <freebsd-hackers@freebsd.org> Subject: 5.4-STABLE corrupting fs on shutdown -p now? Message-ID: <007901c54ec0$14bb3e30$b3db87d4@multiplay.co.uk>
index | next in thread | raw e-mail
Seems like doing a shutdown -p now on the current
5.4-STABLE branch is resulting in FS corruption.
I've done this 4 times now and all but 1 time the
machine has failed to boot with a panic:
mod=0100600, inum=1271817, fs=/var
panic: ffs_valloc: dup alloc
cpuid = 0
boot() called on cpu#0
The only way to fix is single user mode and manually
run fsck. Even the one time it didn't cause a panic I
still got a warning about / not being clean which it
should have been as the machine shutdown cleanly
no outstanding buffers.
I notice in UPDATING:
20050228:
The responsibility of recomputing the file system summary of
a SoftUpdates-enabled dirty volume has been transferred to the
background fsck. A rebuild of fsck(8) utility is recommended
if you have updated the kernel.
To get the old behavior (recompute file system summary at mount
time), you can set vfs.ffs.compute_summary_at_mount=1 before
mounting the new volume.
Could this change have something to do with the panic
happening instead of just recovering?
Why would FS's be being corrupted by "shutdown -p now" where
as "reboot" doesnt seem to?
Steve
================================================
This e.mail is private and confidential between Multiplay (UK) Ltd. and the person or entity to whom it is addressed. In the event of misdirection, the recipient is prohibited from using, copying, printing or otherwise disseminating it or any information contained in it.
In the event of misdirection, illegible or incomplete transmission please telephone (023) 8024 3137
or return the E.mail to postmaster@multiplay.co.uk.
help
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?007901c54ec0$14bb3e30$b3db87d4>