xfs
[Top] [All Lists]

Re: xfsdump segfault

To: Dave Chinner <david@xxxxxxxxxxxxx>
Subject: Re: xfsdump segfault
From: Daniel Browning <db@xxxxxxxxx>
Date: Sat, 16 Feb 2013 16:06:02 -0800
Cc: xfs@xxxxxxxxxxx
Delivered-to: xfs@xxxxxxxxxxx
In-reply-to: <20130216065111.GV26694@dastard>
References: <201302152220.46756.db@xxxxxxxxx> <20130216065111.GV26694@dastard>
User-agent: KMail/1.12.4 (Linux/2.6.32-279.22.1.el6.x86_64; KDE/4.3.4; x86_64; ; )
Thanks for the quick response, Dave.

On Friday 15 February 2013 10:51:11 pm Dave Chinner wrote:
> On Fri, Feb 15, 2013 at 10:20:46PM -0800, Daniel Browning wrote:
> > Hello folks,
> >
> > I ran into a segfault while running xfsdump, what should I do
> > to find the cause?
> 
> xfsrestore segfaulted, not xfsdump.

Ah, yes, that's right.

> Can you run xfsdump to dump to
> a file, then xfsrestore to restore from the file, rather than piping
> one into the other? 

Done. It took a little while since it's a 2.4TiB filesystem. Here is
the last line of the "-v debug" restore (path names scrubbed for
anonymity):

xfsrestore: restoring UNIX domain socket ino 128004852 
backup/secondary/backups/firefly/root/backups/[...]/etc/socket
Segmentation fault (core dumped)

[root@betelgeuse ~]# dmesg | tail -n 1
xfsrestore[18838]: segfault at 10 ip 00000033fd8478de sp 00007fcad10a0e00 error 
4 in libc-2.12.so[33fd800000+189000]

So it appears that a socket is causing the problem. What should I
try next?

--
DB

<Prev in Thread] Current Thread [Next in Thread>