On Thu, 26 Mar 2009, Fazil Saiyed wrote:

Hello,
Since  there is lot of data, it may run for a while, before restoring.
Questions:
When you rebuilt the new cellera, did you put the same cellera  OS code &
NDMP Options , Domain etc for the restore, were the NMDP versions for
backup and restore the same ? if you have multiple clients for Cellera is
your password userid set the same  on each instance of the cellera client
?
Have you tried to backup any thing from the newly built Cellera, it may
shed some light if in the process a config issue could be identified,
while ndmp debug is turned on.
You mentioned that you see the directories being created under root vol ?
is this where the original data was ? if not you may have path coded
incorrectly in restore.

Hi,
I didn't do any of the Cellera work, soI honestly don't know,

Rachel Polanskis <r.polanskis AT uws.edu DOT au>
Sent by: EMC NetWorker discussion <NETWORKER AT LISTSERV.TEMPLE DOT EDU>
03/25/2009 06:21 PM
Please respond to
r.polanskis AT uws.edu DOT au


To
NETWORKER AT LISTSERV.TEMPLE DOT EDU
cc

Subject
Re: NDMP restore results in no data






On Wed, 25 Mar 2009, Fazil Saiyed wrote:

Hello,
I gather the info from previous Cellera post, i would ask EMC where the
variable needs to be set, my notes say, it effects only the restore,

that

may explain why it's taking so long to restore ( how much data are you
talking about ?), was there no previous restores done for cellara that

you

can reference.
You could also, try to use direct dump method for celerra ? Check your
root vol make sure it's not being dumped there, typically, you should
atleast see
directories being created on the destination after the restore has run

for

few minutes if not you got an issue, check the networker restore log for
progress report.
Thanks

We're trying to restore a 3.2 TB filesystem that was done as dump level
NDMP over ethernet.   It used to take 3-5 days to backup full.

The system we are restoring to has been scratch built for the purpose.

We are using the "-v off" setting to try to get whatever data we can.

We looked at the root_vol - the restore creates the required directory,
but no data.

We did some restores for this client last year, but it worked OK.
Now it is an issue and to make matters worse, last night the tape
drive doing the recover decided to throw errors so I had to
mess about and dismount the volume etc.

We're just about to give up on this month and try the May 2007
data sets.   The problem was that the client in 2007 only backed up
successfully a couple of times in the year and some annoying
person wants data back from the very filesystem we had trouble with!


rachel

Rachel Polanskis <r.polanskis AT uws.edu DOT au>
Sent by: EMC NetWorker discussion <NETWORKER AT LISTSERV.TEMPLE DOT EDU>
03/24/2009 06:53 PM
Please respond to
r.polanskis AT uws.edu DOT au


To
NETWORKER AT LISTSERV.TEMPLE DOT EDU
cc

Subject
Re: NDMP restore results in no data






On Tue, 24 Mar 2009, Fazil Saiyed wrote:

Hello,

From what i gather, you are now trying to recover NDMP data to outside

of

Cellara, this is not possible, what you can do is, restore to NAS (
Cellara) then copy over to the original server.( NDMP Data has to be
recovered to like device)
Are the restores to NAS working fine ? Please check DSA_SAVE_NODE=IP
variable.

Hi,
thanks for your info.

I have no idea what DSA_SAVE_NODE=IP is or where it is supposed to be

set.

So far, the restore has taken 3 days and there is 1 tape left to go,
which should be completed by this afternoon.   But initial examination
shows no data present thus far in the restore filesystem.

The backups were done to a celerra that is now retired.
The restores are being done to a celerra that has been ressurected
for the purpose with a similar configuration.    The restore is
being done from the same nsrserverhost that performed the original

backup.

My question still stands - is there a way to "force" a restore no matter
what?
IE to tell legato to blart out whatever is in the SSID and dump it,
regardless of data quality or missing data?   We would be happy to
just see something come back....


I will report back if we have success/failure!    I finally got a login
into EMC PowerClunk
so I will have to log a case through there if we get no joy.....


Cheers....

--
Rachel Polanskis                Systems Admin, University of Western Sydney
ADD Werrington North Campus     (+61 2) 9678 7291  <r.polanskis AT uws.edu DOT 
au>
                The price of greatness is responsibility.

To sign off this list, send email to listserv AT listserv.temple DOT edu and type 
"signoff networker" in the body of the email. Please write to networker-request 
AT listserv.temple DOT edu if you have any problems with this list. You can access the 
archives at http://listserv.temple.edu/archives/networker.html or
via RSS at http://listserv.temple.edu/cgi-bin/wa?RSS&L=NETWORKER