[OpenAFS] Error when moving volumes

[Jeffrey Hutzelman]

On Monday, June 28, 2004 07:56:02 -0400 Kris Van Hees 
<aedil-afs@alchar.org> wrote:

> Please reply to the list, or at least Cc the list.
>
> On Mon, Jun 28, 2004 at 01:04:38PM +0200, Frode Nilsen wrote:
>> The VolserLog on marvin says:
>>
>>     ...
>>     Mon Jun 28 04:58:34 2004 1 Volser: Clone: Cloning volume 536871605
>>     to new volume 536872061 Mon Jun 28 04:59:07 2004 1 Volser:
>>     DumpVolume: Rx call failed during dump, error 1492325122 Mon Jun 28
>>     04:59:42 2004 1 Volser: Delete: volume 536872061 deleted ...
>
> That error translates to:
>
> 1492325122 (vols).2 = badly formatted dump
>
> It would be interesting to see if it is 100% reproducible, and whether
> you get any additional information with adding -verbose to the vos move
> command.

That error is being reported to the source volserver by the target 
volserver; that is, it is the result of the AFSVolForward RPC.

Nearly all of the cases where this error can be returned are the result of 
a failure to read the next chunk of data from the dump, which would usually 
indicate a connection problem or a short dump.  One of those places 
corresponds to the error in the target volserver log.

There are a bunch of things I'd like to know...

- Is the problem reproducible?  Does it happen every time?
- What versions of OpenAFS are you running on each server?
- How big is the volume?
- What output do you get with -verbose ?
  (among other things, this will tell us if the failure is during
  the initial full dump or the later incremental)

-- Jeffrey T. Hutzelman (N3NHS) <jhutz+@cmu.edu>
   Sr. Research Systems Programmer
   School of Computer Science - Research Computing Facility
   Carnegie Mellon University - Pittsburgh, PA