Netbackup, Sharepoint, GRT with restore error status 37

[u500302]

Hello there,

Sharepoint Backup with GRT enabled completes successfully.

restore failed with error 37, there is no other details about the failure!

we have Symantec Netbackup 7.1.0.3

Please advise where should i invistigate?

[RamNagalla]

hi,

could you give us more detail?

what is the storage unit that you are using for Backups?

form where did you trigger the restore job?

please check the T/N , which give more info about the EC 37

http://www.symantec.com/business/support/index?page=content&id=TECH46050

[u500302]

regarding the link you have posted, it is not valid for my case

i do not see the error

Operation requested by an invalid server

[u500302]

could you give us more detail?
 

we have 5 servers for the sharepoint 2010:

front servers: sr1001spw001, sr1001spw002

application server: sr1001spa001

SQL Cluster (virtual server sr1001csql2010): nodes sr1001sql011, sr1001sql012
 

what is the storage unit that you are using for Backups?

Symantec Deduplication

 
form where did you trigger the restore job?

from the Master server
 

[RamNagalla]

could you provide the restore log from the location /usr/openv/netbackup/user_ops/<userid>/logs

and also bpcd and bprd logs of master server at the time of restore job as attachment.

[u500302]

Thanks Nagalla for your support, the master and media servers are installed on windows based servers.

can you specify what logs to check?

I believe my issue is in configuring servers list somewhere.

[RamNagalla]

hi ,

make sure that you have the bpcd and bprd log directories created in master server. I am sure about your master server is Window or Unix

Keep Verobse = 5 

for Unix:-

/usr/openv/netbackup/logs/bpcd 

/usr/openv/netbackup/logs/bprd

for windows:-

Installpath/veritas/netbackup/logs/bpcd

Installpath/veritas/netbackup/logs/bprd

then open up your java conosle of Master server and trigger the restore job., while click on the restore log, you will have option to specify the Progress log.. 

by default it will be 

/usr/openv/netbackup/user_ops/<userid>/logs --->for unix

Installpath/veritas/netbackup/logs/<userid>/logs  --> for windows

please provide all these 3 logs 

[u500302]

please find attached logs, please let me know if you have any question.

[Mark_Solutions]

Have you completed the Distributed Application Mappings section on the Master Servers Host Properties - it says it is for Exchange but it is for SharePoint as well

Hope this helps - will take a look at your logs

[Mark_Solutions]

Do you have the logs from yesterday when you tried the restore (010213.log) or have you tried it today - if so at what time?

I do see:

14:48:48.374 [7144.5124] <16> is_redirected_restore_allowed: One or more input parameters are invalid.

So maybe it is the Distributed Application Mappings that are needed, the No.Restrictions file on the Master can also help

[RamNagalla]

hi

From bprd log

01:03:19.077 [6340.2024] <16> free_allocated_resources: Free resource allocations failed.

01:03:19.077 [6340.2024] <2> mail_msg_and_set_exit_status: Attempting to send mail to root on se1103dfil1001.services.musanada.net

01:03:19.077 [6340.2024] <2> vnet_cached_getaddrinfo_and_update: ../../libvlibs/vnet_addrinfo.c.1478: 0: in failed cache: 11001 0x00002af9

01:03:19.077 [6340.2024] <2> vnet_cached_getaddrinfo_and_update: ../../libvlibs/vnet_addrinfo.c.1479: 0: in failed cache name: se1103dfil1001.services.musanada.net

01:03:19.077 [6340.2024] <2> vnet_cached_getaddrinfo_and_update: ../../libvlibs/vnet_addrinfo.c.1480: 0: in failed cache service: NULL

From bpcd log:-

15:47:38.190 [916.8188] <2> bpcd main: VERBOSE = 0

15:47:38.190 [916.8188] <2> logparams: D:\Program Files\Veritas\NetBackup\bin\bpcd.exe -standalone 

15:47:38.190 [916.8188] <2> process_requests: offset to GMT -14400

15:47:38.190 [916.8188] <2> logconnections: BPCD ACCEPT FROM 10.0.31.1.1722 TO 10.0.31.1.1556 fd = 480

15:47:38.190 [916.8188] <2> process_requests: setup_sockopts complete

15:47:38.206 [916.8188] <2> bpcd peer_hostname: Connection from host sr1001bck001.services.musanada.net (10.0.31.1) port 1722

15:47:38.206 [916.8188] <2> bpcd valid_server: comparing sr1001bck001 and sr1001bck001.services.musanada.net

15:47:38.206 [916.8188] <4> bpcd valid_server: hostname comparison succeeded

15:47:38.206 [916.8188] <16> process_requests: read failed: The operation completed successfully.

I dont understand why its using port 1722 to communicate...

do you have firewall in place.. does it configured properly?

[Marianne]

Nagalla
Perfectly normal - source ports are random. The destination port is the only important one: 1556.

[RamNagalla]

Marianne,

Thank you very much..

[u500302]

Distributed Application Mapping has been configured,

SharePoint front servers are mapped with sql backend and cluster virtual server, and with the indexing server.

[u500302]

I have tried the restore one hour or less before i upload the logs.

Distributed application mapping has been configured.

I am not sure what do you mean by the No.Restrictions file.. i have not read about it in sharepoint admin guide, could you please explain?

[u500302]

There is no firewalls between the servers in our case.

[u500302]

I will add more details, this may help

Master server is: sr1001bck001

SharePoint Front servers: sr1001spw001 and sr1001spw002

SharePoint application server: sr1001spa001

SQL servers: sr1001sql011 and sr1001sql012, Cluster virtual server is: sr1001csql2010

[Mark_Solutions]

On your master server create a file named No.Restrictions (this is case sensitive) under the path:

\program files\veritas\netbackup\db\altnames\

This allows redirected restores when requested from clients - which is in effect how SharePoint works as it redirects the restores to the components of the Farm

If you are not happy to leave it in place just rename it after you have re-tried your restore.

If you could setup a folder named tar under the logs directory on the front end farm server and then re-try the restore telling us the time it started and finished and then re-upload the logs again so that we can pin point the restore and what is happening.

Also note that you should only do restores from the Master Server or the Front End Farm Server

[u500302]

the No.Restrictions file was created long time ago for SQL backup policies.

I am aware about the source limitations, restore should be run from Master or Front end server.

I am not sure about the tar folder, could you explain more please?

[Marianne · Accepted solution]

We see reference to hostname centraladmin.services.musanada.net in bprd log:

14:48:48.374 [7144.5124] <16> is_redirected_restore_allowed: One or more input parameters are invalid.

14:48:48.374 [7144.5124] <2> process_request: client sr1001spw002 peername centraladmin.services.musanada.net is invalid for restore request

Where does this hostname fit into the picture?

Browsing was successful - we see many fileslist entries in bprd log.

Restore request is processed:

15:47:01.267 [1744.7232] <2> process_request:   browse_clnt: sr1001spw00215:47:01.267 [1744.7232] <2> process_request:   requesting_clnt: sr1001spw00215:47:01.267 [1744.7232] <2> process_request:   destination_clnt: SR1001CSQL201015:47:01.267 [1744.7232] <2> process_request:   clnt_bp_conf_name: sr1001spw00215:47:01.267 [1744.7232] <2> process_request:   peername: centraladmin.services.musanada.net

The final result is that 'centraladmin' is seen as actual destination name. Restore ends with status 48:

15:47:01.298 [1744.7232] <2> local_bpcr_connect: Can't connect to client centraladmin.services.musanada.net
15:47:01.298 [1744.7232] <2> ConnectToBPCD: bpcd_connect_and_verify(sr1001spw002, centraladmin.services.musanada.net) failed: 48
15:47:01.298 [1744.7232] <2> log_to_progress_file: BPCD on centraladmin.services.musanada.net exited with status 48: client hostname could not be found

If 'centraladmin' is your Windows Admin Console, please check if 'configured client name' was at any point changed. Important that it must ALWAYS be the local hostname, unless you really want to redirect the restore to this machine. 

Use 'All Programs', Symantec NetBackup -> Backup Archive and Restore window on 'centraladmin', 
Open File -> NetBackup Client Properties.
Check Client Name.
Ensure that it is indeed 'centraladmin'.