Video Screencast Help

BE 2012: deleted servers keep reappearing

Created: 14 Feb 2014 • Updated: 30 May 2014 | 9 comments
This issue has been solved. See solution.

BackupExec 2012 Rev. 1798 64 Bit with Service Pack 3 running on Microsoft Windows Server 2008 R2 Standard, backing up a Windows 2008 file server and Active Directory domain controller, an Exchange 2010 server, a Windows 2008 terminal server, a Linux server, a VMware ESXi host, and itself to disk deduplication store and copied from there to LTO-5.

In the Backup Exec console on the Backup and Restore tab, I expect and want to see the servers for which there are actual backup jobs, ie. four Windows servers, one Linux server, and one ESXi host. But in addition to that, there appear 7 out of the 54 VMs on the ESXi host and one Exchange DAG which a colleague created and deleted again half a year ago. One of the VMs (as well as the DAG, obviously) show as "never backed up", while the other six VMs show up as last backed up during the corresponding ESXi job, but when entering their details the Jobs and Job History panes are empty.

I have tried to delete these servers from the Servers pane several times. It seems to work at first but at some later time the deleted entries just reappear, typically next time I start the console but sometimes even when just making some other change.

Ideas?

aTdHvAaNnKcSe,
Tilman

Operating Systems:

Comments 9 CommentsJump to latest comment

sidd3009's picture
Hello Tilman,
Have you upgraded Backup Exec in the past? If there are some backups taken using IP Address/Netbios/FQDN then the server names would appear in the list. We pull resource containers from catalogs. 
Also, since you are backing up the virtual machines using the ESX host(I am assuming with GRT) the Virtual Machines would be added to the list automatically. With that being said and considering that you have the GRT option checked, the only option to restore the individual contents of Virtual Machines would be by using the those Automatically added Virtual Machines. 
Regarding the 'Never Backed Up' state, you could refer to the following link: http://www.symantec.com/business/support/index?page=content&id=TECH185820
A workaround to this is to create a small backup of the individual Virtual Machines and then, you'd be able to view the contents to restore by using those individual Virtual Machines. 
Edit: For more details, you could refer to the best practices for Agent for VMWare: http://www.symantec.com/business/support/index?page=content&id=HOWTO74443

Regards,

Siddhant Saini
Advanced Technical Support Engineer, Symantec Corporation 
www.symantec.com

Artegic's picture

Hello Siddhant,

thanks for your prompt reply.

Backup Exec has not been upgraded, we installed 2012 directly.

It is possible that some of the VMs in question have been backed up directly instead of via ESXi in the past.

Only some of the reappearing servers are backed up with GRT. There's at least one Linux server which because of the limitations of RALUS is deliberately backed up without GRT. OTOH of the nine VMs we do back up with GRT, only five appear in the servers list.

Would you recommend that I add the remaining GRT-enabled VMs to the server list? What about the non-GRT ones?

Thanks,
Tilman

joeyharper's picture

I'd read somewhere that the deleted servers reappear because they still have a backup exec agent installed. Ours come back each time the BE services are re-started. Bit of a pain - I'd prefer it if it I didn't have to fully remove the agent to get it to stop re-appearing in the console.

pkh's picture

Yes. This is the expected behavior since the remote agent will be communicating with the media server when it scan for servers to backup.

You can try stopping the BE remote agent service on these servers.

SOLUTION
pkh's picture

One of the reasons why the servers keep coming back could be that there are still backup sets that belong to these servers. If you delete all remaining backup sets, the servers may go away

Artegic's picture

Update:

After reducing the holding time for job logs from 365 to 180 days, the Exchange DAG now stays deleted, and the one VM reporting "never backed up" is also permanently gone. I guess there had been old job records for those.

Four VMs still show up which are backed up via ESXi with GRT. I didn't try deleting them again as I gather from the preceding discussion that those entries may be useful in case of a restore.

One remaining VM I couldn't identify yet, because its name "Test-PC" does not match any of my VMs' names. I assume it has been renamed sometime in the past. Its Backup Exec version is blank, but it reports daily successful backups, so I'm not too worried about it. I guess I'll identify it eventually.

Thanks, all!

Tilman

Artegic's picture

Another Update:

The Exchange Test DAG is back. Its Backup Exec Version column is blank, and its Status column shows  "Never backed up".

Of course I've changed plenty of things - installed SP4, removed the Linux server and re-added it under a different name, dropped a few VMs from the ESXi host, adapted the ESXi jobs' selections manually because Backup Exec cannot handle deleted VMs correctly on its own, recreated one of the ESXi jobs from scratch as requested by Symantec support on the vain hope that it might fix some unrelated GRT problem, recreated all of the monthly off-site tape duplication jobs - on top of the regular backup cycles. But nothing of that has any obvious relation to that former Test DAG.

Im baffled.

Artegic's picture

Yet another update after even more changes all over the place - backed-up servers as well as media server:

Now only servers that have an agent installed and running come back, which is somewhat understandable. I guess I can live with that. What's annoying is that some (not all) of them show up as "never backed up" even though they are backed up daily as VMs via AVVI. It seems that the "backed up/never backed up" cannot detect AVVI backups reliably. But that's a different topic.

Marking as solved.

 

Colin Weaver's picture

I's recomend upgradding to BE 2014 (next week)

1) The backup status update against virtual machines is more reliable

2) We have introduced a retired servers group to allow you to manage the history of decomissioned/obsolete servers that may still have backup data available but keep them out of the current "All Servers" list