02-01-2013 06:05 AM
Hi,
I have 2* 5220 MSs and 4*5020 Storage pool. I take backup over my existing MediaServer ( 7.5.0.3 ) ( MSDP ) to the storage pool. But while i want to make restore, data is coming from 5220. Why does data take this way? Will I not be able to make restore over my bakcup media server?
Regards
02-04-2013 01:53 AM
I guess that is all right as you mighe be duplicating the data from the local appliance to remote appliance.
As per the SLP the backup will be done first on the local appliance(remote location) or local MSDP and then get dupliated to Data Center location that will be attached to your data center.
You have primary copy on the local appliance(remote site) and 2nd copy at the data center location. Once the primary copy is expired and if you try the restore it will restore the data from the data center location MSDP. In these kind of setup you would be duplicating the data from remote MSDP to your data center MSDP and while restoring it will definately pickup the media server that has access to data center MSDP.
Let us know if this is not the case and do provide us the details how data is being written?
02-04-2013 02:16 AM
Hi sazz,
thank you for response.
this is not case.
I have one Datacenter. I take data to firstly media server that deduping and after write to Storage pool to 5020. and then I want to make restore data for test but on activity monitor, restore job opened onto media server column as 5220's hostname but My client that is taken backup does not know appliance's ip and hostname so restore failes ( adding 5020's ip to client's host file , case is resolved ) but I wanna understand why My data is coming from 5220 Appliance Media server not my backup Media server that is deduping and write to directly 5020.
Regards
02-04-2013 04:12 AM
Please post output of
nbemmcmd -listhosts -verbose
Also where are you running the restore from ?
Make sure under the host properties -> Master server properties -servers - all the media server listed. If not add them.
On the client make sure you have the master server and media server entries.
02-04-2013 04:27 AM
Output is below.
cluster istnbp01
media istnbpapp12 --> 5220 --> on java monitor gui , Openning Restore job's media server is that server. Why is not that istnbp02 ?
media istnbpapp11 --> 5220
media istnbp03
media istnbp02 --> I take backup over this Media server ( 7.5.0.3. )
ndmp istnbpdup11
ndmp istnbpapp21
ndmp istnbpapp22
ndmp istnbpapp12 --> Appliance
ndmp istnbpdup21 à5020
ndmp istnbpapp11
ndmp istnbpvcs01
ndmp istnbp01
ndmp istnbp03
api istnbpsun05
server istnbp01
master istnbpvcs01
master istnbpvcs00
I running restore on Java console from Master Server. All ip and host entires are correct because I can make restore correctly over 5020 ( istnbpapp12 ) adding its ip and host.
Regards
02-06-2013 02:09 AM
Can you please explain it with screenshot or may be paste the output of detailed status of backup and restore.
02-06-2013 06:19 AM
It is possible that the data flow does go directly from the 5020 to the client
The same is the case if you wanted to use client side de-dupe when client data would go directly from the the client to the 5020 - to use that your clients also need to know the name and IP address of the 5020's
If you ever want to use client side de-dupe or any related features you would need this adding to your clients anyway so it may be for the best that it is done now
02-07-2013 12:04 AM
Hi sazz,
I have attached picture. you can see at picture, restore job opens over appliance media server not backup media server taking the backup.
02-10-2013 11:25 AM
Was FORCE_RESTORE... entry added to bp.conf on master server at any point in time?
or Media Host Override in Host Properties on master server?
Please ensure that bprd log folder exists on the master server.
If not, create folder and restart NBU.
Post bprd log as File attachment after next restore attempt.