NDMP RESTORATION FAILING WITH 2850

[Sagar_Kolhe]

Hello All,

I am facing very critical ndmp restoration issue in our environment. We have tried almost everything to go through it but no success.Previousely restoration has been completed successfully for the same ndmp and path. Please find the details below.

We are restorting the data on same ndmp.

NDMP details : 

[SF440]# tpautoconf -verify abcd
Connecting to host "abcd" as user "ndmp"...
Waiting for connect notification message...
Opening session--attempting with NDMP protocol version 4...
Opening session--successful with NDMP protocol version 4
  host supports TEXT authentication
  host supports MD5 authentication
Getting MD5 challenge from host...
Logging in using MD5 method...
Host info is:
  host name "server_3"
  os type "DartOS"
  os version "EMC File Server.T.7.1.76.4"
  host id "abc1997"
Login was successful
Host supports LOCAL backup/restore
Host supports 3-way backup/restore.

Activity Monitor details :

12/18/2015 00:04:25 - Info ndmpagent (pid=25415) INF - Restoring NDMP files from /root_vdm_2/voicelgrarch1srv_FS/voicelgrarch1srv_FS/Archived_Calls/20131109190054/COLL/KANJU/32111/10102013/PRATIKSHA_COLL_KANJU_32111_10102013_091547_000014_0/ to [See line below]
12/18/2015 00:04:25 - Info ndmpagent (pid=25415) INF - Restoring NDMP files from [See line above] to /root_vdm_2/voicelgrrestoresrv_FS/Restore
12/18/2015 00:05:51 - Error ndmpagent (pid=25415) ndmp_data_connect_v3 failed, status = -1 (-1)
12/18/2015 00:05:51 - Error ndmpagent (pid=25415) NDMP restore failed from path /root_vdm_2/voicelgrarch1srv_FS
12/18/2015 00:06:14 - Info ndmpagent (pid=25415) done. status: 25: cannot connect on socket
12/18/2015 00:06:14 - Info bpbrm (pid=25395) got ERROR 5 from media manager
12/18/2015 00:06:14 - Info bpbrm (pid=25395) terminating bpbrm child 25405 jobid=161575
12/18/2015 00:06:14 - restored from image abcd_1419923491; restore time: 0:02:15
12/18/2015 00:06:14 - end Restore; elapsed time 0:02:17
Failed to get status code information (2850)

10000 port is open bi-directionally.

[mnolan]

Hi, I was helping behind the scenes on your support call earlier today. (I recognize the names listed here.) I believe the issue is that a new NDMP host is now the last write host on the tape, but shouldn't be involved in this restore.

Confirm with

C:\Program Files\VERITAS\NetBackup\bin\admincmd>nbemmcmd -listmedia -mediaid <mediaid>

So for this instance we are setting up a 3-way restore and THAT ndmp host informed us which IP addr to give to your RESTORE host which it cannot seem to reach.

Regardles, if you set the last write host to the media server per the article the above come from, you should be able to do a remote restore.

https://www.veritas.com/support/en_US/article.TECH68059

Now further, the ip that the new NDMP host informed of us does not really matter ( as I believe you were unsure where that IP was anyways) as we are less concerned about this new host functionaly to be a host for other filers, I would rather you did the below.

Steps:
C:\Program Files\VERITAS\NetBackup\bin\admincmd>nbemmcmd -listmedia -mediaid <mediaid>

Is the new filer listed as last write host on the tape that you are trying to restore from?

Yes?

Move it to the media server:

bpmedia -movedb -m <MEDIA> -oldserver <last write host> -newserver <desired media server for restore>

The attempt restore once more. Ports for remote NDMP from filer to media server need to be open if any firewalls exits, but IIRC there are none here.

[Sagar_Kolhe]

Hello Nolan,

Last write host is the same media server on which backup was taken for this ndmp. There is no other media server involved. 

[Marianne]

Any chance you can show us nbemmcmd output as per mnolan's request? Do you have bprd log that covers failed restore attempt? Please upload as .txt file.