cancel
Showing results for 
Search instead for 
Did you mean: 

Unable to connect to the selected NetBackup host "masterserver.xxx.com"

Vickie
Level 6

Hi All,

Windows 2000 (Service Pack 4)
X86 Family 6 Model 10 Sepping
NBU Master server version 6.5.6

NetBackup console is not opening, giving below error.

Unable to connect to the selected NetBackup host "masterserver.xxx.com"

1. Make sure the user "admin" has privileges on the host.

2. Make sure the local host is......

and some more points, can be find in attached screen shot.

Below is what I tried.....

1) Restart the services, Console opens...but after some time getting same error.
2) During bpdown, failed to stop some demon. Killed them manually.
3) Not able to stop bpdbm, getting permission issue.
4) Name resolution ok (nslookup, reverse nslookup)
5) Checked in registery, name is defined accurately.
6) Put the entry in HOSTS file.

Please assist if anyone has faced such issue earlier.

I know that 6.5.6 is EOL, but cant upgrade to 7.x as server is windows2000. But still the things are in progress to upgrade the hardware first then the NBU.

1 ACCEPTED SOLUTION

Accepted Solutions

VerJD
Level 4
Employee

@Vickie,

bpclntcmd -pn, is telling you that your registry SERVER entry is correct. (MASTERSERVER)
bpclntcmd -self, is telling you that your client information is different. (MASTERSERVER.XXX.COM)

  • NOTE 1: NetBackup is case sensitive, so the entries should match what's reported in 'nbemmcmd -listhosts'

Please check your registry Client_Name entry is using MASTERSERVER, and your hosts file entry matches the 'nbemmcmd -listhosts -verbose' output:

MASTERSERVER
        MachineName = "MASTERSERVER"
        FQName = "MASTERSERVER.xxx.com"

So the hosts file entry should look something like this:

10.250.13.8     MASTERSERVER     MASTERSERVER.xxx.com

After making changes, be sure to restart NetBackup services. ('bpdown -f -v' ⇒ 'bpps' ⇒ 'bpup -f -v')

  • NOTE 2: Make sure your client name in the policy also matches, since it looked like it was all lower case FQDN in the detailed status, if I remember correctly.

(Also Reference: Explanation of bpclntcmd options and recommended troubleshooting when the commands return errors. - TECH50198)

JD | Veritas NetBackup Support

View solution in original post

11 REPLIES 11

Will_Restore
Level 6

Been more than a few years ago, we had Dell server running Windows 2000 and same trouble ("Unable to connect to the selected NetBackup host ...") when NIC teaming was implemented.  It should have worked but was always trouble.  Anyway that was ancient history.  If you have NIC teaming, try disabling it and see if that helps. 

 

VerJD
Level 4
Employee

@ Vickie,

Wow, you're right, that system definitely needs a hardware refresh and an upgrade to NetBackup 7.x. wink

 

Anyway, when services are restarted... you definitely want all of them to stop before you restart them again, especially "Adaptive Server Anywhere - VERITAS_NB - ASA Database service" (NB_dbsrv) and "NetBackup Database Manager" (bpdbm).

- Do you have the exact error when you try to stop bpdbm, or when you run this command?

install_path\Veritas\NetBackup\bin\bpdbm -terminate

 

Otherwise, it sounds like the NetBackup services are running under a different user account than the account you logged in with, or your account doesn't have full admin privileges... 

- Are the NetBackup services running under Local System or a domain admin account?

(Use this technote as reference for what services should be running on a NB 6.5 master: TECH57455)

 

As an alternative, if you cannot get all of the services to restart...

- Can you reboot the server, log in under the NetBackup service account, to get all the services to start up properly together, and then open the admin console?

JD | Veritas NetBackup Support

Vickie
Level 6

@ JD

There is no error during bpdbm terminate. I restarted the ASA, and it worked from services window.

I logged in with admin rights on the server. The services are running on local system account.

I rebooted the server, when server came up I open the console....ran a backup but it gives error like below.

Lack of connectivity detected between media server and master server. 

3/27/2014 3:56:33 AM - requesting resource masterserver-storage-unit
3/27/2014 3:56:33 AM - requesting resource masterserver.NBU_CLIENT.MAXJOBS.masterserver.xxx.com
3/27/2014 3:56:33 AM - requesting resource masterserver.NBU_POLICY.MAXJOBS.nbupol
3/27/2014 3:56:33 AM - awaiting resource masterserver-storage-unit Reason: Media server is currently not connected to master server, Media Server: masterserver,
     Robot Number: 0, Robot Type: TLD, Media ID: N/A, Drive Name: N/A,
     Volume Pool: DAILY, Storage Unit: masterserver-storage-unit, Drive Scan Host: N/A
client process aborted(50)

 

Marianne
Level 6
Partner    VIP    Accredited Certified

It is always difficult to assist with troubleshooting when you cannot share real hostnames.

Something has changed at OS level as far as hostname is concerned. If you open cmd and issue 'hostname' command, what do you see?

Compare above output with 'nbemmcmd -listhosts -verbose' and 'nbemmcmd -getemmserver' output.

Also create bprd and bpdbm log folders and stop/start NBU with bpdown and bpup. Check these 2 logs to see how hostname is compared with NBU configured hostnames.

 

PS: You REALLY need to migrate to supported OS and NBU version. Even Microsoft has dropped support for W2000 some years ago.

Vickie
Level 6

Hi Marianne,

D:\VERITAS\NetBackup\bin\admincmd>nbemmcmd -getemmserver
NBEMMCMD, Version:6.5.6
Unable to get the local EMM server for the following reason: "cannot connect on socket"

C:\Documents and Settings\admin_rsingh18>hostname
MASTERSERVER


D:\VERITAS\NetBackup\bin\admincmd>nbemmcmd -listhosts -verbose
NBEMMCMD, Version:6.5.6
The following hosts were found:
MASTERSERVER
        MachineName = "MASTERSERVER"
        FQName = "MASTERSERVER.xxx.com"
        MachineDescription = ""
        MachineNbuType = server (6)
MASTERSERVER
        ClusterName = ""
        MachineName = "MASTERSERVER"
        FQName = "MASTERSERVER.xxx.com"
        GlobalDriveSeed = "VEND:#.:PROD:#.:IDX"
        LocalDriveSeed = ""
        MachineDescription = ""
        MachineFlags = 0x4
        MachineNbuType = master (3)
        MachineState = not active (6)
        NetBackupVersion = 6.5.6.0 (656000)
        OperatingSystem = windows (11)
        ScanAbility = 0
Command completed successfully.

Some of the error I found from old BPRD and BPDBM logs. And latest one are attached.

Error from BPRD logs.

00:39:20.649 [71900.76740] <16> bkarfiles: Unable to write progress log </D/VERITAS/NetBackup/Logs/user_ops/dbext/logs/vxbsa.1395905957.76716.6708.prog.pcb_std> on client MASTERSERVER.XXX.COM. Policy=Catalog-Pol Sched=Weekly

BPDBM logs error I found,

01:03:35.587 [10636.71648] <16> do_online_nbdb_backup: Error backing up files in D:\VERITAS\NetBackupDB\staging
01:03:35.884 [10636.71648] <16> exec_catalog_backup: exec_asa_database_backup() failed (2)
02:34:40.431 [69784.71088] <16> process_request: requestor MASTERSERVER is not a valid server for IMAGE request 9101:03:35.884 [10636.71648] 
04:11:45.790 [73944.73396] <16> delete_expired_backups: OVsystem(D:\VERITAS\NetBackup\bin\admincmd\nbdelete.exe -allvolumes -jobid 53974) failed (128)
04:15:41.634 [70444.71076] <16> close_handle: Terminating child process before client closed connection because parent bpdbm process died

Vickie
Level 6

Now after running backup again, getting below error in detailed logs.

 

cannot send mail because BPCD on MASTERSERVER.XXX.COM exited with status 23: socket read failed

Vickie
Level 6

I found one major difference, I believe that could be root cause.

bpclntcmd -pn  (gives output as MASTERSERVER)
bpclntcmd -self (gives output fqdn MASTERSERVER.XXX.COM)

D:\VERITAS\NetBackup\bin\admincmd>bpclntcmd -self
gethostname() returned: MASTERSERVER.XXX.COM
host MASTERSERVER.XXX.COM: MASTERSERVER.XXX.COM at 10.250.13.8 (0x12a1100a)
aliases:

D:\VERITAS\NetBackup\bin\admincmd>bpclntcmd -pn
expecting response from server MASTERSERVER 
MASTERSERVER MASTERSERVER 10.250.13.8 58197

Please assist if this can cause the issue, then how can I resolve this ?

Marianne
Level 6
Partner    VIP    Accredited Certified

You need to find out what has changed at OS level to make hostname show up as fqdn.

In the meantime, add IP address with both shortname and fqdn in master's hosts file, the stop and restart NBU with bpdown and bpup. That should resolve the hostname lookup issue.

VerJD
Level 4
Employee

@Vickie,

bpclntcmd -pn, is telling you that your registry SERVER entry is correct. (MASTERSERVER)
bpclntcmd -self, is telling you that your client information is different. (MASTERSERVER.XXX.COM)

  • NOTE 1: NetBackup is case sensitive, so the entries should match what's reported in 'nbemmcmd -listhosts'

Please check your registry Client_Name entry is using MASTERSERVER, and your hosts file entry matches the 'nbemmcmd -listhosts -verbose' output:

MASTERSERVER
        MachineName = "MASTERSERVER"
        FQName = "MASTERSERVER.xxx.com"

So the hosts file entry should look something like this:

10.250.13.8     MASTERSERVER     MASTERSERVER.xxx.com

After making changes, be sure to restart NetBackup services. ('bpdown -f -v' ⇒ 'bpps' ⇒ 'bpup -f -v')

  • NOTE 2: Make sure your client name in the policy also matches, since it looked like it was all lower case FQDN in the detailed status, if I remember correctly.

(Also Reference: Explanation of bpclntcmd options and recommended troubleshooting when the commands return errors. - TECH50198)

JD | Veritas NetBackup Support

Vickie
Level 6

Thanks all for your support.

Actually the issue was with name, I just modify and make all name entries same to same wherever I found the server name I just changed it identically.

By doing this the issue resolved and now the backup running fine.

Thanks once again. :)

VerJD
Level 4
Employee

You're most welcome Vickie, glad to assist. Thanks for bringing your question to us and marking the solution.

Have a great week! (seriously, no April Fools about it) wink

JD | Veritas NetBackup Support