I have strange problem with nmsql restore. I created a ticket for that but EMC does not provide any solution yet. So i decieded to ask comminity.
I am backing up sql databases with nmsql 5.2 sp2 . Yesterday i need to restore one the backed up databases to new server. But when i started to restore the gui says
nsrsqlrc.c(7447): Could not get saveset record from mmdb; error: 0x00013884
savetime 1363030213 not in media db
Error getting backup endtime.
nsrsqlrc.c(4050): Error getting backup endtime from media Database.
and then starting to wait.
EMC support says its a known issue and it can recover succesfully but it waits in browsing state forever and recover never start. I need to recover this database as soon as possible and its size 2.2 TB. Any idea ?
Win 2003 R2 Sp2 x64
NW Server 7.6.4
Nw Client 7.6.4
Nmsql 5.2 Sp2
If EMC said this is known issue, it means you need patch. I guess it would be interesting to hear which patch it is. From it says in error, it looks as if nsrsqlrc did call to indexdb for lookup of index space for given client and nsavetime 1363030213 and failed to get any response for nmsql namespace. Most likely, you can verify that from backup server by doing:
nsrinfo -t 1363030213 -n mssql <client name>
...and see if you get anything. From there there are two possibilities:
a) nsavetime value passed is wrong (eg. there is nothing returned by above command)
b) there is some bug in communication (eg. if you do get response)
But i am getting this message for alot of save set id. After them it says
2 %s 1 0 511 RESTORE database [ARP] FROM virtual_device='Legato#3815e3b9-2e97-4366-af5a-f85d47f0183e' WITH move 'ARPDATA1' to 'G:\ARP1\ARP_ARPDATA1_22431.mdf', move 'ARPDATA10' to 'G:\ARP1\ARP_ARPDATA10_22431.ndf', move 'ARPDATA11' to 'G:\ARP1\ARP_ARPDATA11_22431.ndf', move 'ARPDATA12' to 'H:\ARP2\ARP_ARPDATA12_22431.ndf', move 'ARPDATA13' to 'H:\ARP2\ARP_ARPDATA13_22431.ndf', move 'ARPDATA14' to 'H:\ARP2\ARP_ARPDATA14_22431.ndf', move 'ARPDATA15' to 'H:\ARP2\ARP_ARPDATA15_22431.ndf', move 'ARPDATA16' to 'H:\ARP2\ARP_
and start waiting.
Yeah ok, those are not save sets, but rather datafiles for same database (ARP) which you are moving (not sure why though - did you request that?). Do you see anything in event log?
As per NMSQL release notes:
Error message displayed when recovering old backup data (NW121054)
During point-in-time recovery of an old backup, the following error is displayed in
51281:(pid 5580):nsrsqlrc.c(7399): Could not get saveset
record from mmdb; error: 0x00013884
savetime -1 not in media db
80282:(pid 5580):nsrsqlrc.c(4002): Error getting backup
endtime from media Database.
However, the restore completes successfully.
So my understanding here is that this is a cosmetic error however the recovery should work fine.
This is a 2,2 TB DB, so as you understand this is going to take time.
Did yo monitor the nsrsqlrc process? what about NW server, is keeping the connections and restore sessions open/active?
Does it fail or just keeps on waiting? If just keeps on waiting, check on the NW server side if there is any activity, like reading from device, etc.
Problem resolved by Emc Support team . They send me a patched sql module and fix some name resolution issues. Now recover is working .