Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 05 Jun 2010 13:11:18 +0200
From:      "Peter Cornelius" <pcc@gmx.net>
To:        FreeBSD-gnats-submit@FreeBSD.org
Cc:        freebsd-bugs@FreeBSD.org
Subject:   ufs_dirbad, nullfs, jail (was: kern/147420: [nfs] kldload nfs modules causes nfs-aware kernel panic after a while)
Message-ID:  <20100605111118.166660@gmx.net>
In-Reply-To: <20100604074500.211900@gmx.net>
References:  <201006030800.o53802PX037019@freefall.freebsd.org> <20100604074500.211900@gmx.net>

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

After two more failures, I now found the offending inode with find /usr -inum 46895494 and removed that directory with rm -rf <dir>.

I re-enabled INVARIANTS, and I hope that the file system now at least is clean and that the server is stable now.

As one point, I found the inode in a directory which usually is mounted for an (ez-) jail via nullfs. The jail is used to build packages and hence somewhat busy on file system level. I don't know whether this adds to the pointers of the issue.

If the box is stable now, I would not pursue this matter further but am more than willing to answer questions.

Thanks to all,

Peter.

---

P.S.:, for completeness sake, here the variations observered but without the expectation that that would add much more information.

>   panic: ufs_dirbad: /usr: bad dir ino 46895494 at offset 114368: mangled
> entry
>   cpuid=1
>   KDB: enter: panic
>   [thread pid 42063 tid 100153]

This is, as one may expect, the only line that changes:

   [thread pid 66833 tid 100168]

or, upon ls of the directory which used to have the inode no. 46895494:

   [thread pid 3244 tid 100154]

>   Stopped at      kdb_enter+0x3a movl    $0,kdb_why
-- 
GRATIS für alle GMX-Mitglieder: Die maxdome Movie-FLAT!
Jetzt freischalten unter http://portal.gmx.net/de/go/maxdome01



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20100605111118.166660>