error code 58 on linux client.

[Albatross_]

Hi,

I am facing error code 58 during the backups .

we have one master server ( 7.7.1 ) and 6 media servers 7.7.1. ( 5 linux and 1 windows )

the backups are scheduled. everytime the backup runs on linux client it fails with error code 58. ( other clients in the same policy succeed )

Client version is 7.5.0.4 ( linux RH 5.4 )

two media servers are located in DC N and and two media servers are located at DC C. 

we have stroage units groups configured and SLP as well.

everytime backup is scheduled, the client is talking to media server in DC "C" ( earlier working fine with no issues ), and the backup is failing.

I have checked the ports from DC N to client and vice versa. it seems to be fine.

So I tried to remove the servers list ( DC "C" ) from the bp.conf on the client. Restarted the job manually, for some reason its again trying to talk the media server in DC "C".

Can some body help me out in solving the issues. Tired all the soltuions from the forum but nothing worked.

awaiting for valuable replies.

Thanks

[sdo]

Status 58 is unable to reach the client.

One or more of the NetBackup Servers (i.e. master and/or media) is unable to reach the client.

First step is to check name resolution.  Can the master AND the media server both resolve the client name?

[Albatross_]

Hi Sdo,

Yes there is no issue with the  name resolution from master / media and client. ( from servers in DC "N" to client and Vice versa )

There seems to be issue with the media servers in DC "C" hence I have removed the servers in bp.conf ( client file ) and restarted manually, but still the client is trying to contact media serves in DC "C".

My intention is to take the backup from the media servers in DC "N"

thanks

[sdo]

Ok - try this on both the master server and on the media server:

bptestbpcd -verbose -client myclientname.mycompany.com

...does this command work?

If not, you may have a firewall issue, either on network hardware firewall device between the NetBackup Server(s) and the client, or in the OS software based firewall on the backup client.

[Albatross_]

Yes I have proper output when executed on master server and media servers ( in DC "N" ).

when executed bptestbpcd command from the medias server ( DC "C" ) the command runs for longer time and returns error. which means there is issue.

when  there is no issue with media servers in DC "N" ,  I would like to make  use of them to take the backup

[sdo]

Ok - you know which media servers work... so why not put the backup client in a new backup policy which has (in the policy and/or in the schedules) a storage unit, or storage unit group, or SLP... which reference the media server and/or media servers which are able to actually reach the backup client?

[Marianne]

I agree. When you use STUG, you need to ensure that all media servers in STUG can connect to all clients in the policy. If client can only connect to one media server, then move client to own policy with specific STU/media server.

[Albatross_]

Hi All,

The client was de'commed . The de'comm team haven't povided any information to the backup team, due to which the backup is running on the client.

Now we have confirmed with the de'comm team and removed the client from the backup rotation.

Thanks for your valuable comments

Cheers

[Marianne]

Interesting....

Curious to know how you had proper connectivity from certain media server on Monday......

" Yes I have proper output when executed on master server and media servers ( in DC "N" )."

[Albatross_]

Hi Marianne,

As a company's policy The decommission process takes a week time from the request. 

Monday late hours ( after my shift hours ), they have decommed the server and unplugged the cables.

Tuesday when I came back to my work, have checked that Client was not reachable and checked with the decommed team and they let me know.

Since I am newly joined into the project, my email is not included into several distribution list. I am expecting in a week everything gets into the track.

Thanks for your help

Cheers