NDMP backup error-ndmp_data_connect_v3 failed, status = 23 (NDMP_CONNECT_ERR)

[spenukon]

Hi

I am configuring NDMP backups on new NBU servers. The port 10000 is open both ways. The storage array is sun7410. I am setting up a 3 way backup. I have checked with the network team that there is no firewall between array and nbu servers. The credentials are all set. When I try to run the backup, I am getting below error. Any help would be greatly appreciated.

Error ndmpagent (pid=31196) ndmp_data_connect_v3 failed, status = 23 (NDMP_CONNECT_ERR)
 

[spenukon]

I have also configured the ndmp.cfg file on the both media servers under /usr/openv/netbackup/db/config as per the article in Symantec site.

[Will_Restore]

When you authorize NetBackup access to the NDMP host and configure robots

using the Administration Console, NetBackup automatically verifies NDMP

credentials and the robotic configuration. If you want, you can re-verify them.

tpautoconf -verify ndmp_host_name

[spenukon]

Yes tpautoconf is working fine. I have verified that.

[Symboy]

Hi 

  Looking at the error message , it definately seems to be a Network configuration/ Comunication issue.

Not sure which Symantec technote you have referred and what changes have you made in ndmp.cfg file .

But please have a look at the /etc/host file .

In case you have multiple NIC's on Filer , you may need to provide in ndmp.cfg file , which one to connect .

We have also seen issues in case IPV6 is used on the filer , which can be disabled using the same file .

Collect the required logs for more details... 

[spenukon]

The ip I added in the ndmp.cfg was wrong, so I changed it. I still have problems but the errors are different now. Here is the output.

09/21/2012 17:59:00 - estimated 0 kbytes needed
09/21/2012 17:59:00 - Info nbjm (pid=31058) started backup
09/21/2012 17:59:01 - Info bpbrm (pid=3966) reading file list from client
09/21/2012 17:59:01 - Info bpbrm (pid=3966) starting ndmpagent on client
09/21/2012 17:59:01 - Info ndmpagent (pid=3968) Backup started
09/21/2012 17:59:01 - Info bpbrm (pid=3966) bptm pid: 3969
09/21/2012 17:59:01 - Info bptm (pid=3969) start
09/21/2012 17:59:01 - Info bptm (pid=3969) using 30 data buffers
09/21/2012 17:59:01 - Info bptm (pid=3969) using 262144 data buffer size
09/21/2012 17:59:01 - started process bpbrm (pid=3966)
09/21/2012 17:59:01 - connecting
09/21/2012 17:59:01 - connected; connect time: 0:00:00
09/21/2012 17:59:02 - Info bptm (pid=3969) start backup
09/21/2012 17:59:03 - Error ndmpagent (pid=3968) ndmp_mover_listen failed, status = -1 (-1)
09/21/2012 17:59:03 - Error ndmpagent (pid=3968) NDMP backup failed, path = /export/messaging/prod_mc1emdb_backup
09/21/2012 17:59:03 - begin writing
09/21/2012 17:59:05 - Info bptm (pid=3969) EXITING with status 99 <----------
09/21/2012 17:59:05 - Info ndmpagent (pid=0) done. status: 99: NDMP backup failure
09/21/2012 17:59:05 - end writing; write time: 0:00:02
NDMP backup failure  (99)
 

[Will_Restore]

OK, now you have to verify that matches the path on the filer (not the share the host)

See this technote  http://www.symantec.com/business/support/index?page=content&id=TECH128997

[spenukon]

I checked the sun array and the path looks the same.

[spenukon]

I opened a case with symantec and the support person said that port 10000 is being used by another application. I checked the lsof and /etc/services looks good.  I am not sure what else to look. Any help would be really appreciated.

[Will_Restore]

I don't know about that, seems like he could have provided some proof?

[Will_Restore]

Oh my, Symantec Hardware Compatibility List does NOT show 3-way as supported option with 7x10 storage, only Direct attach or Remote. 

[spenukon]

Thanks a lot Wr for researching.  I am working on direct attach now.