cancel
Showing results for 
Search instead for 
Did you mean: 

Netbackup, Sharepoint 2007 Cross farm restore - lab enviroment creator

NDinev
Level 3
Partner Accredited Certified

I have to make restore of one SharePoint 2007 farm to another farm which will be used for lab and developing purposes, so both farm have to work in same time but with different names.

GRT full and Full Non-GRT backup complete successfully without any warning. I’ve read checklist from article TECH148598, and clarify all settings.

http://www.symantec.com/business/support/index?page=content&id=TECH148598

Restore process doesn’t start when I try to restore of SharePoint Web application as is advised in article HOWTO47543. I can restore without problems files from GRT backup.

http://www.symantec.com/business/support/index?page=content&id=HOWTO47543

I’m just not sure how to configure properly Distributed Application Restore Mapping settings. In “NBU for SharePoint admin guide”  there are not any examples.

All backup goes to disk. NBU server is Netbackup 7.1 onto Windows Server 2008 R2.

Production and lab frontend servers are SharePoint 2007 running onto Windows Server 2008 SP2 / 64 bits. SQL server is cluster, SQL 2008.

I have following server configuration:  

Production SharePoint Farm:

Intranet – Virtual name of NLB frontend

SPSNode1 – Sharepoint frontend

SPSNode2  - Sharepoint frontend

ProdSQL – SQL cluster virtual name, backend in SharePoint config

SQLnode1 – first node in SQL cluster

SQLnode2 - second node in SQL cluster

Lab SharePoint Farm:

TESTintranet - Virtual name of NLB frontend

TESTnode1 – frontend

TESTnode2 – frontend

TESTsql – SQL server, backend in SharePoint config

Do you have any idea how to configure Distributed Application Restore Mapping?

Any other ideas are welcome.

I've just open a case let see who will be helpfull community or Tech Support :)

 

1 ACCEPTED SOLUTION

Accepted Solutions

Symontec
Level 5

I would map every SP frontend to the every SP+SQL backend server since the option (from my pour understanding) is to allow (in the process restore) the frontend from where you are recovering, to write its configuration to the SQL config DB on the backend member(s) of that farm.

I have done this using this way. Just try it and let ud know.

Good luck!

View solution in original post

15 REPLIES 15

Zahid_Haseeb
Moderator
Moderator
Partner    VIP    Accredited

I have a very very bad experience in Sharepoint restore. I too successfully Backed up the Sharepoint 

 but not able to restore even could not get any healthy help from Symantec forum and from Symantec support too.

watsons
Level 6

Basically it means the mapping between your SP frontend and backend server.

For your production, that will be "Intranet" -> "ProdSQL"

For your lab, it is "TESTintranet" -> "TESTsql"

As for farm restore, I don't know but I believe there are lots of issues pending to be rectified in NBU SP option, probably due to the complexity. I think the best is to push for enhancement request if it doesn't work as expected. Not just farm restore, but GRT restore can also be a headache sometimes especially when it comes to verifying the end result. Restore may work but some of the restored components are not tallied and requires workaround.

Roger_C
Level 4
Employee

Hi NDinev,

I have to admit I have yet to test out a fully cross farm recovery on NBU 7.1 as of yet.

However, you mentioned that the restore process does not start when comitting the restore.

OK, have you tried simply restoring to same existing farm as if you were doing a normal restore - does the restore job invoke correctly?

If not, then check in Task Manager if you have bpspserver.exe running in the process list. If this exists then please kill this off, sometimes it can remain after a failed restore attempt, this process if running does prevent other restore job hitting the queue,

Failing that, increase the logging levels on the client and check <install_path>\Veritas\Netbackup\logs\sps2vra & <install_path>\Veritas\Netbackup\logs\nbwin

Check for any errors at the time you invokved the restore,

RogerC

Symontec
Level 5

I would map every SP frontend to the every SP+SQL backend server since the option (from my pour understanding) is to allow (in the process restore) the frontend from where you are recovering, to write its configuration to the SQL config DB on the backend member(s) of that farm.

I have done this using this way. Just try it and let ud know.

Good luck!

NDinev
Level 3
Partner Accredited Certified

Hi guys,

Thank you for you advices!

Brief answers to your notes...

@Zahid.Haseeb  - I’m Sorry to hear that, I hope that I’ll have better experience.

@watsons – I try such type of mapping, but it didn’t help, tomorrow I’ll try mapping as Symontec adviced

@Roger C – I didn’t try restore to Production farm because I need special permission for that action, I make restores of files from GRT backup redirected to disk. Tomorrow I’ll check logs, is bpspserver.exe process running and will try restore to production farm. 

@Symontec– I’ll test this mapping tomorrow.

Tomorrow, I will configure Distributed Application Restore Mapping, that all SharePoint frontend servers to point to ProdSQL and TESTsql and then run some tests.

Funny thing is that the guy from Symantec support case tell me that SharePoint alternative restore is currently not supported this is contradiction of what is written here: https://www-secure.symantec.com/connect/blogs/netbackup-71-features-other-minor-enhancements-part-1-...

Symontec
Level 5

Great, we will be waiting for your feedback.

Regards,

SAF

NDinev
Level 3
Partner Accredited Certified

Hi guys,

I map all SP frontend to production and test  SQL, but again restore job didn't start.

I've killed process bpspserver.exe and then, I try redirected restores of one web application on custom port from production farm to test farm,  in <install_path>\Veritas\Netbackup\logs\sps2vra log i find only this error message:

18:42:49.129 [2740.1652] <2> bprd_start_spss_recovery_asst: Connect to bprd successfully!
18:42:49.129 [2740.1652] <2> bprd_start_spss_recovery_asst: Sent bprdreq successfully!
18:42:49.176 [2740.1652] <2> bprd_start_spss_recovery_asst: Sending 0:Windows SharePoint Services Web Application/SharePoint - 51918/ to master
18:42:49.176 [2740.1652] <2> bprd_start_spss_recovery_asst: get_short failed: 0 14

after that i try GRT restores which works like charm.

I do restore in following orders, am I miss some step?

1) I got Full backup with no errors and fresh installed test  SP farm

2) Configure Distributed REstore Mapping

3) Create Web application on Test farm, I don't create any site collection or SSP 

4) Logon to frontend server from test farm and begin restore process; which didn't run :(

Actully from this how to http://www.symantec.com/business/support/index?page=content&id=HOWTO47543

I didn'tcomplete step 3 Remove the source Web application

Any suggestions?

Symontec
Level 5

Just a couple of tips more that I recall from what Symantec Support did:

When you run the BAR Console on the client or the NBConsole, use Right Click on the icon -> Run as administrator.

Another tip (it was not tested) they said is that when you select for restore, do not select from "All Folders" section but from "Contents" pane of the BAR Console.

Symontec
Level 5

Additionally, I would check the usage of the No.Restrictions file by taking a look of this TN http://www.symantec.com/business/support/index?page=content&id=TECH45378

NDinev
Level 3
Partner Accredited Certified

Hi Symontec, No.Restriction file is already set onto master server. 

Roger_C
Level 4
Employee

Hi,

I noticed that you (step 3) created the Web Application on the Test farm with same structure, I don't think is a necessary step as I thought we created the Web Application on the fly - unless someone else can clarify here.

As I wonder if this created the same original IIS physical path ie "C:\inetpub\wwwroot\wss\*" etc,

I'm also confused as to whether you stated whether the restore actaully initiated this time round and the Job appeared in the Activity Monitor, could you clarify?

To check if the restore had errors - check the <install_path>\Veritas\Netbackup\logs\beds\SPSWrapper00.log on the Frontend commiting the restore.

This will help capture the 'stsadm' commands - which is called in the creating the Web Applciation on the new farm.

NDinev
Level 3
Partner Accredited Certified

Hi Roger C,

I can clarify that restore job is not initiated and is not seen in "Activity Monitor". The beds directory is empty, no logs.

SharePoint on both farms uses default location C:\inetpub\wwwroot\wss\*".

I also delete Web application from test farm and try again restore but same result restore job doesn't start.

Symontec
Level 5

Hi NDineV, any updates in this case?

NDinev
Level 3
Partner Accredited Certified

Hi Guys,

I’ve got success :) . I recheck everything in Sharepoint backup check list and also find this article TECH158316 as usefull  (Currently is not accessible , Google cache can help ).  

As additional steps I add service account for SharePoint backup to Netbackup Legacy Client Service”and recreate web application on destination farm to have exact match of IIS port and location of Web Application C:\inetpub\wwwroot\wss\80\

Now I have some troubles with some webparts that are missing after restore, but I think to push these troubles to SharePoint team.

But, (always have BUT J), now I have problem with GRT backup and restores I’m hitting known bug: http://www.symantec.com/business/support/index?page=content&id=TECH87800

GRT Backup completes without any error or warnings, but when I try to restore I cannot select file to restore files and folders appear to be missing.  If I try to restore file which is seen in recovery console, files are restored but their size is 0 KB.

I find following errors in Windows log about NFS and think that they might be the problem:

Source:        NfsRdr

Event ID:      16384

Level:         Warning

Description:  Client for NFS requested a mount with file locking enabled, and the remote Network File System (NFS) server does not support file locking.   Client for NFS mounted the shared resource anyway. File locking may not be available.   To avoid this warning in the future, enable file-locking suppport on the remote NFS server.

-------------------------------- 

Log Name:      Application

Source:        NfsClnt

Event ID:      16397

Level:         Warning

Description: Windows(R) Lightweight Directory Access Protocol (LDAP) failed a request to connect to Active Directory Domain Services(R) for Windows user < \netbackup_user>.   Without the corresponding UNIX identity of the Windows user, the user cannot access Network File System (NFS) shared resources.    Verify that the Windows user is in Active Directory Domain Services and has access permissions.

-------------------------

Do you have any ideas?

Thank you Guys, without your help i couldn’t make it

WHITE_EAGLE
Level 3
Partner

See planning a cross domain replication disaster recovery domain on page 49  PDF, and always perform a full backup prior to and after  any changes to your environment.