Chunk layout in archive
Neil Truby had a June ontape archive on 11.50FC6WGE but no record of its chunk layout and wanted to restore it elsewhere. Everett Mills said ontape creates the chunks on restore if the files exist. Neil found the chunk names with ontape piped to strings. Art Kagel suggested extracting reserved pages with dd and od -c; Mark Scranton noted reserved pages may not be contiguous.
Auto-generated by Claude from the posts below — may be imperfect; read the full thread.
I really feel I should remember this ...
11.50FC6WGE on x86.
I have an ontape archive of a Benchmark db I took last June, and I want to
restore it onto a potential customer's server to do some benchmarking.
Trouble is, I seem to have failed to record the chunk layout present in the
archive, so not sure how to set up the new server's chunks!
Back in the day I seem to recall that ontape showed you the layout when
restoring from TAPEDEV but it doesn't seem to now: it shows the dbspace names
but not the chunks.
Any ideas?
Thx
Neil
Ontape creates the dbspaces and chunks as it does the restore. All you need to
remember are the devices/files they attach to and make sure they exist and
have enough space.
--EEM
> -----Original Message-----
> From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of
> NEIL TRUBY
> Sent: Thursday, April 09, 2015 20:28 PM
> To: ids@iiug.org
> Subject: Chunk layout in archive [34952]
>
> I really feel I should remember this ...
>
> 11.50FC6WGE on x86.
>
> I have an ontape archive of a Benchmark db I took last June, and I want
> to restore it onto a potential customer's server to do some
> benchmarking.
>
> Trouble is, I seem to have failed to record the chunk layout present in
> the archive, so not sure how to set up the new server's chunks!
>
> Back in the day I seem to recall that ontape showed you the layout when
> restoring from TAPEDEV but it doesn't seem to now: it shows the dbspace
> names but not the chunks.
>
> Any ideas?
>
> Thx
> Neil
>
>
> ***********************************************************************
> ********
> Forum Note: Use "Reply" to post a response in the discussion forum.
>> Ontape creates the dbspaces and chunks as it does the restore. All you need
to remember are the devices/files they attach to and make sure they exist and
have enough space.
Yes, thanks, obviously that's true if the chunks are on cooked file systems.
But in order to set up the relevant directories and links I still need the
chunk names!
However ontape | strings | more does tell me this and, as you say, the rest is
workable.
You can use dd to extract the reserved pages which includes the chunk
table. Just pipe the dd output through od -c and into less to read it.
Art
On Apr 9, 2015 9:28 PM, "NEIL TRUBY" <neil.truby@ardenta.com> wrote:
> I really feel I should remember this ...
>
> 11.50FC6WGE on x86.
>
> I have an ontape archive of a Benchmark db I took last June, and I want to
> restore it onto a potential customer's server to do some benchmarking.
>
> Trouble is, I seem to have failed to record the chunk layout present in the
> archive, so not sure how to set up the new server's chunks!
>
> Back in the day I seem to recall that ontape showed you the layout when
> restoring from TAPEDEV but it doesn't seem to now: it shows the dbspace
> names
> but not the chunks.
>
> Any ideas?
>
> Thx
> Neil
>
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
--001a113ea43ed0dd2005137ee1b2
And Neil - remember that the reserved pages can be extended so the chunk pgs may/may not be "together" if you use Art's recommendation to extract them from the front of the archive. Thanks - Mark Mark Scranton The Mark Scranton Group www.markscranton.com