xfs
[Top] [All Lists]

Re: xfsrestore assertion failure

To: Ajay Shekhawat <ajay@xxxxxxxxxxxxxxxxx>
Subject: Re: xfsrestore assertion failure
From: Steve Lord <lord@xxxxxxx>
Date: Fri, 19 Oct 2001 10:57:56 -0500
Cc: linux-xfs@xxxxxxxxxxx
In-reply-to: Message from Ajay Shekhawat <ajay@xxxxxxxxxxxxxxxxx> of "Thu, 18 Oct 2001 19:39:10 EDT." <20011018193910.A9199@xxxxxxxxxxxxxxxxxxxxxxxxx>
Sender: owner-linux-xfs@xxxxxxxxxxx
> On Thu, Aug 02, 2001 at 01:16:26PM -0400, Steve Roseman wrote:
> > >   xfsrestore: seeking past media file directory dump
> > >   xfsrestore: drive_scsitape.c:1461: do_next_mark: Assertion
> > >    `rechdrp->first_mark_offset - rechdrp->file_offset <= ( off64_t )
> > >   ( contextp->dc_recsz )' failed.
> > >   Aborted (core dumped)
> > 
> > The following fix seems to resolve the problem.  xfsrestore -t works, at
> > least.  (I'll try a real restore later.) I suspect the problem is in
> > endian-converting first_mark_offset, maybe when it contains the value
> > -1.  
> 
> We just encountered this assertion failure, in xfsdump v1.1.6.
> Steve's fix does fix the problem and the restore proceeds, but
> I'm wondering if the patch breaks other things. If not, should
> the patch be in the official xfsdump distribution?

I attempted to work out if this fix is in the latest code, but it has 
changed radically, someone with some more experience with the dump code
would have to say if this fix is now present or not. I suspect not since
we are only at 1.1.7:

xfsdump-1.1.7 (18 October 2001)
        - xfsrestore -t will no longer fail if the current working
          directory is not xfs
        - xfsrestore will no longer issue (harmless) warnings related
          to space pre-allocation if it is writing to a non-xfs
          filesystem


Steve

> 
> Ajay



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