Solved!
Go to SolutionRecovering data from Client Direct disabled devices
Hello People, Im trying to restore a block based clone backup.
My case is the following, I have a costumer who has a block based backup configured to a data domain (1 month retention), and then he clones the data to tape (1 year retention).
He wants to make a test to restoring that backup. When I perform the recover on Data Domain, it run well.
Now when I try the same case, but from tape, the process doesnt recovering nothing, and stay in a loop until I press ctrl + c.
Im using this instructions: (Netwoker Adm Guide 8.2 SP1 Pag 549)
Using the CLI to perform the clone recovery
Run one of the following commands to recover the data from the Client Direct disabled
devices:
l recover -S save_set_ID/clone_ID -l pool_name
for file level recoveries
The pool that you select must have a Client Direct enabled device. The pool must also
be a backup clone pool.
And this is our results...
E:\Program Files\Legato\nsr\logs>recover -S 3713110580/1498532438 -d g:\restoreco03afp -l RetornoBBB
Did not find any existing clone on the same pool 'RetornoBBB' cloning now
154960:nsrclone: The following volumes are needed for cloning: 80471:
nsrclone: L05501L5 (Regular)
and the process dont stop, still in that point ultil I press ctrl + C
Note: Im running this command on backup server, I could not do it on the client because the server was not available to do it today.
I created a new pool to try this restore, I tried with the same clone pool that is used to pass it to tape without success.
Any Idea?
2707
0
Posted June 30th, 2017 10:00
Because the tape drive is not a 'client direct enabled device', the recover task you want to start in fact consist of these 2 steps:
- cloning the tape backup clone SS to a DD volume which has been labeled to a Backup Clone pool
- recovering the backup from here.
You may obviously achieve this by using only one command where you must specifiy the 'aux clone pool' like in your command:
recover -S 3713110580/1498532438 -d g:\restoreco03afp -l RetornoBBB
Of course you must create this 'aux clone pool' but of course you must also create a DDboost device and label the media to that pool.
Obviously the first command (nsrclone) has been started in the background but NW still waits for the appropriate media in this pool.
Most likely you will also see a NW media alert due to that reason.
So create the device and label the media appropriately.
##################################
Message to Dell/EMC Tech Docs:
The command option [-l disk clone pool] is still missing even in the latest (NW 9.1.1 Command Line Reference - recover command).
2707
0
Posted June 30th, 2017 13:00
Thanks for your prompt response.
I have performed the procedure to the letter, but at no time does the server mount the tape to start the work
No Alerts, no messages, no warning, nothing....
I have several storage node, I had created the new device using the same storage node where the tape is.
(the tape is on the library)
2707
0
Posted July 3rd, 2017 02:00
I could be wrong, but if I remember correctly for BBB restores you need disk pool and tape was not supported... was that changed? Since my understanding is that you are doing restore from tape.
2707
0
Posted July 3rd, 2017 06:00
Good Morning Hrvoje
As you can see on my first post, there is a procedure on EN Admin.. Guide pag 549, to recover ssid from disabled device. I understand that you told me, What I can see and understand is that the process is to clone the data from the tape to the data domain so that one can then retrieve the information from there.
The customer do a BBB to Data Domain with retention time of 3 months, then clone that to tape with retention time of one year. Now he wants to certify that can retrieve information from a tape.
Regards,
Oscar
2707
0
Posted July 3rd, 2017 08:00
Could you please run the following command:
mminfo -q "pool=CloneBBB" -r "pool,volume,volflags"
2707
0
Posted July 3rd, 2017 09:00
C:\>mminfo -s co03backup -q "pool=CloneBBB" -r pool,volume,volflags
6095:mminfo: no matches found for the query
This Pool is new.... also, the device labeled with this pool is new too.
Im running now the command:
recover -vvv -D3 -s -S -l
and this is the output:
C:\oscar>recover -vvv -D3 -s co03backup -S 3713110580/1498532438 -l CloneBBB
07/03/17 11:48:16.928540 Save set ID to recover: 3713110580/1498532438
07/03/17 11:48:16.975341 Setting default RPC socket send buffer size to 262144
07/03/17 11:48:16.975341 Setting default RPC socket receive buffer size to 262144
07/03/17 11:48:16.975341 Auth methods not set, using default
07/03/17 11:48:16.990941 Auth methods not set, using default
07/03/17 11:48:16.990941 Auth methods not set, using default
07/03/17 11:48:16.990941 Auth methods not set, using default
07/03/17 11:48:17.178143 Setting DNS cache TTLs to 1800 secs for positive lookups and 1800 secs for negative lookups
07/03/17 11:48:17.396546 RPC Authentication: error in LookupAccountSid: No mapping between account names and security IDs was done. (Win32 error 0x534)
07/03/17 11:48:17.412146 RPC Authentication: error in LookupAccountSid: No mapping between account names and security IDs was done. (Win32 error 0x534)
07/03/17 11:48:17.443347 lgto_auth: redirected to co03backup prog 390103 vers 2
07/03/17 11:48:17.474547 lgto_auth for `nsrmmdbd' succeeded
07/03/17 11:48:17.568148 ssid is a valid BBB ssid and recoverable by BBB
07/03/17 11:48:17.583748 SS volume is of family disk
07/03/17 11:48:17.583748 SS volume is of type Data Domain
07/03/17 11:48:17.583748 SS volume is of family tape
07/03/17 11:48:17.583748 SS volume is of type LTO Ultrium-5
07/03/17 11:48:17.583748 Saveset '3713110580' found on a non disk device. It has to be staged onto a disk
Did not find any existing clone on the same pool 'CloneBBB' cloning now
07/03/17 11:48:17.646149 Executing clone command nsrclone -s co03backup -y "3 days" -b CloneBBB -S 3713110580/149853243807/03/17 11:48:17.646149 No callback registered on session channel 2 for user-defined message: (message type: 170, request type: 0)
07/03/17 11:48:17.646149 Recover Clone Job Created
07/03/17 11:48:47.848136 Recover Clone Job Started
154960:nsrclone: The following volumes are needed for cloning:
80471:nsrclone: L05501L5 (Regular)
07/03/17 11:49:18.970535 No callback registered on session channel 2 for user-defined message: (message type: 215, request type: 0)
07/03/17 11:50:19.826916 Recover Clone Job Still Running
07/03/17 11:51:20.667696 Recover Clone Job Still Running
07/03/17 11:52:21.508476 Recover Clone Job Still Running
07/03/17 11:53:22.349256 Recover Clone Job Still Running
07/03/17 11:54:23.190036 Recover Clone Job Still Running
07/03/17 11:55:24.030816 Recover Clone Job Still Running
07/03/17 11:56:24.871596 Recover Clone Job Still Running
07/03/17 11:57:25.712376 Recover Clone Job Still Running
07/03/17 11:58:26.553156 Recover Clone Job Still Running
07/03/17 11:59:19.593836 No callback registered on session channel 2 for user-defined message: (message type: 215, request type: 0)
07/03/17 12:00:20.434616 Recover Clone Job Still Running
07/03/17 12:01:21.275396 Recover Clone Job Still Running
07/03/17 12:02:22.116176 Recover Clone Job Still Running
the command is still running with the same message: Recover Clone Job Still Running.......
2707
1
Posted July 3rd, 2017 10:00
Sorry but your output ...
C:\>mminfo -s co03backup -q "pool=CloneBBB" -r pool,volume,volflags
6095:mminfo: no matches found for the query
... obviously shows that no volume at all has been labeled for this pool. What have you done?
BTW - the media alert will not show with the recover command - you will see it in the Admin GUI.
2707
0
Posted July 3rd, 2017 11:00
Do you see anything in clone tab sessions? The way I read it, NW in background is running nsrclone -s co03backup -y "3 days" -b CloneBBB -S 3713110580/1498532438 and for that it will need L05501L5 to read from.... was that volume loaded? Does it show reading data from it? (you can use nsrwatch to get most accurate information)
OscarMl
1 Rookie
•
25 Posts
2707
0
Posted July 3rd, 2017 12:00
bingo/Hrvoje Crvelin
When I ran the command
C:\>mminfo -s co03backup -q "pool=CloneBBB" -r pool,volume,volflags
6095:mminfo: no matches found for the query
I had labelled the pool with Default Clone Label to test......
But I relabelled again to CloneBBB, sorry for my mistake, I was trying too many things.
I had help from a technician EMC support.
We send Manually the save set to clone, and nothing.
It seems that the library had problems and gave constant errors when trying to mount the tape manually, then proceeded to restart the library and once this was done, we were able to clone the data from the tape to the Data Domain and once there we proceeded to perform with Success a restore from a direct client.
Thank you all for your help and cooperation.