[OpenAFS] Error when moving volumes
Jeffrey Hutzelman
jhutz@cmu.edu
Mon, 28 Jun 2004 16:48:59 -0400
On Monday, June 28, 2004 22:38:25 +0200 Frode Nilsen
<mailing-lists@cyberpunks.no> wrote:
> marvin is running RedHat 7.3, OpenAFS 1.2.7
> oliven is running Fedora Core 1, OpenAFS 1.2.11
>
>
> Here is the output from afsdump_scan:
>
># afsdump_scan -PHVv user.h100554.dump
> * DUMP HEADER [0 = 0x0000000000000000]
> Magic number: 0xb3a11322
> Version: 1
> Volume ID: 536871605
> Volume name: user.h100554
> Dump Range: 0 => 0
> * VOLUME HEADER [39 = 0x0000000000000027]
> Volume ID: 536871605
> Version: 1
> Volume name: user.h100554
> In service? true
> Blessed? true
> Uniquifier: 10802
> Type: 0
> Parent ID: 536871605
> Clone ID: 536872065
> Max quota: 102400
> Min quota: 0
> Disk used: 7183
> File count: 1216
> Account: 0
> Owner: 1663
> Created: Mon Aug 25 22:18:20 2003
> Accessed: Thu Jan 1 01:00:00 1970
> Updated: Tue Sep 23 16:37:43 2003
> Expires: Thu Jan 1 01:00:00 1970
> Backed up: Thu Jan 1 01:00:00 1970
> Offine Msg: A volume utility is running.
> MOTD:
> Weekuse: 0 0 0 0
> Weekuse: 0 0 0
> Dayuse Date: Fri Apr 23 00:00:00 2004
> Daily usage: 1
> * VNODE 1/1 [214 = 0x00000000000000d6]
> afsdump_scan: Unknown tag in AFS volume dump Unexpected tag 'D' at 260 =
> 0x0000000000000104 *** FAILED: Unknown tag in AFS volume dump
>
>
> Okey, that didn't look good :-(
No, not really.
I should have told you to use -PHVvi so we'd get more complete inode
information. It might still be useful to do that. However, given that the
problem happens so early in the dump, it may yet be possible to debug.
dd if=user.h100554.dump of=user.h100554.dump.BEGIN bs=256 count=2
This should construct a nice small file containing the first 512 bytes of
the dump. This will include the headers and the first couple of vnodes,
including whatever the problem spot is.
Put that somewhere on the web or in AFS, and send a pointer.
I want to take a look at the dump and see if I can figure out what's going
on.
-- Jeff