SureBackup Temp VMs

[markfellows]

Just started to look at Backup and Replication v5 and so far I'm impressed.

My question is this :

When a SureBackup job has completed is there any way of instructing Veeam to stop the temporary VM the SureBackup creates and remove it?

Thanks,

Mark.

[Alexey D.]

Hello Mark,

Please go to SureBackup job settings to make sure that "Keep VMs running..." checkbox is unchecked (which is the default setting).
If it is checked, the behavior you described is expected by design: with VMs up and running one may perform any custom actions such as U-AIR restore.

In other cases, B&R automatically shuts down and removes temporary VMs after SureBackup job finishes.

[Gostev]

Hi Mark, actually this is exactly how SureBackup job works
Oops, looks like Alexey beat me by a few seconds.

[markfellows]

Thanks guys. Another quick question slightly off topic (and not really a Veeam issues), is thier anyway to automatically clean up the orphaned VM in VCenter?

[markfellows]

. . . Had a quick look I have 'Keep the applcation group running once the job completes' and this is unchecked but can't seem to locate 'Keep VMs running'. Could you tell me which section of the job settings it's in please.

Thanks,

Mark.

[Alexey D.]

Hello again Mark,

Yes, I meant 'Keep the applcation group running once the job completes', sorry I wasn't precise (didn't have B&R installation under my fingers at that time).
Ok, by the way, what do you treat as 'temporary VM' - appliance VM or VMs running from application group?

[markfellows]

Alexey,

Thanks for the quick reply, thought this was the option. The temporary VMs I'm referring to are those created by the SureBackup job (VM running from the application group), I presume the appliance VM should always be running.

The problem seems to be that the SureBack session has to be stopped manually. When I stop it manually the VM is stopped and removed (showing as orphaned in VCenter).

In the SureBackup demo video I watched the SureBackup verication job started and then the session automatically stopped once the job was complete.

Have I configure something incorrectly?

Thanks,

Mark.

[Alexey D.]

Mark,

More clarification needed again Did you start this surebackup job manually or it was started by U-AIR request?
And, maybe, did you use "Start" command of pop-up menu in session window, to restart failed VM, if any?

[markfellows]

Alexey,

I've started the SureBackup job manually and via a schedule and the same thing happens.

[Alexey D.]

That's an odd behavior, please send the logs to the support team, we will take a look and help you.

[Gostev]

The problem seems to be that the SureBack session has to be stopped manually. When I stop it manually the VM is stopped and removed (showing as orphaned in VCenter).

This typically happens when you delete VM directly from ESX(i) host added to vCenter, instead of doing this through vCenter.
Did you possibly add your ESX(i) host to Veeam Backup as standalone host, and that host is a part of vCenter?

[markfellows]

Anton, this sounds plausable, only remember adding the individual host. I'll try adding the vCenter server, is that straight forward?

Would this also correct the problem with the SureBackup job not stopping.

[markfellows]

. . . It's just a radio button, must have missed this when I orinally set up B&R v4 (unless it wasn;t in this verison).

[Vitaliy S.]

Yes, adding vCenter Server should resolve "orphaned" VMs issue. I would appreciate it, if you could update us on the SureBackup job results as well...

[markfellows]

Vitaliy/Anton

Since making the change I now get the following error when I run the SureBackup job.

11/11/2010 15:19:02 Sending email report
11/11/2010 15:19:03 SERVERNAME: Unpublishing
11/11/2010 15:19:06 Stopping virtual lab routing engine
11/11/2010 15:19:06 Fail
Unable to mount vPower NFS volume (<BACKUPSERVERNAME>:/VeeamBackup_BACKUPSERVERNAME). SERVERNAME.domain: Error during the configuration of the host: NFS Error: Unable to Mount filesystem: The NFS server does not support MOUNT version 3 over TCP

<IP Address> : Error during the configuration of the host: NFS Error: Unable to Mount filesystem: The NFS server does not support MOUNT version 3 over TCP

Not sure if it relates to my setup (http://www.veeam.com/forums/viewtopic.php?f=2&t=5553).

Mark.

[Gostev]

Hi Mark, for any technical issues please work with our support directly. Thanks!

[markfellows]

Will do Anton, I'll post the outcome here . .

As an aside I’ve just taken a call from a salesman trying to sell me a backup solution. Explained we used Veeam and how impressed we were with it and he implied that it doesn’t de-dupe.

I backup my Vms individually but when I ran some tests before we purchased last year I’m the combined file was smaller than the combined individual backup files.

Am I right in thinking Veeam B&R does de-dupe data?

[Alexey D.]

Yep, you're right: B&R does de-dupe data. And does compress. Everything is done to save your storage space which, as we know, is quite expensive
Just combine VMs in one job to get advantage of these features, grouping similar VMs gives even greater effect.

[Gostev]

As an aside I’ve just taken a call from a salesman trying to sell me a backup solution. Explained we used Veeam and how impressed we were with it and he implied that it doesn’t de-dupe.

Wow, some of these competitors are either clueless, or simply desperate Veeam was probably the first VMware backup vendor to have deduplication, we had built-in dedupe in the product since version 1.0 released over 3 years ago. Every other competitor I know of added dedupe years later, after looking at us.

By the way, we have fairly fresh technical whitepaper with some metrics around our dedupe: Deduplication and VMware Backup Sprawl

If you have further questions about our dedupe, I have whole section dedicated to this topic in sticky FAQ thread.

[markfellows]

Anton/Alexey,

I thought so, think he was just desperate because everything he said his product could do I explained Veeam already does it!! (. . and better, especially in v5).

Keep up the good work (with the product and forum support).

[markfellows]

Ok, I've been dealing with the Veeam support team for the last few days (I was pleased to discover that there service was excellent like the forum team).

I added my vCenter server to Veeam B&R rather than just adding my ESXi host. Unfortunately this completely stopped Instant Recover and SureBackup from working as described previously in this post.

After some routing around the error (NFS Error: Unable to Mount filesystem: The NFS server does not support MOUNT version 3 over TCP) was due to the fact that the VeeamNFSService was not running on port 111. Instead this port was being utilised by Windows Services for Unix. Once this was uninstalled and I restarted the Veeam NFS Service everything worked fine.

If you’re experiencing issues when using Instant Recovery or SureBackup regarding the mounting of the NFS share then ensure the VeeamNFSService is running on ports 111,1058 and 2059.

The SureBackup job is still not stopping after it is complete (Dev team are looking into this) but all VMs and being stopped and cleaned up correctly now (if the job is stopped manually).

[Alexey D.]

Hello Mark, thanks for updating the topic with resolution.

[markfellows]

Support managed to figure it out and it was a user related error!! I had configured the server in my application group but also added it as a Linked Backup during the SureBackup job setup.

Removed it as a linked backup and the SureBackup job now stops once it has finished.

[Gostev]

OK, thanks again for updating the topic.

[chi-ltd]

[merged into existing discussion]

Hi All

I have recently installed our vcentre server on our network (mainly for the converter snap in option). I have now noticed that the surebackups are leaving aa numebr of orphaned servers after each scheduled job.
The vsphere client doesn show these at all, only the vcentre client connection.

Is there any reason why this is?

Thanks

[Vitaliy S.]

Hello Gaz,

Could you please clarify if you've added your ESX(i) host to Veeam Backup as standalone host, and that host is a part of vCenter?

If this is the case, then this is expected and usually happens when you delete VM directly from ESX(i) host added to vCenter, instead of doing this through vCenter. To resolve your issue you need to configure your SureBackup job through vCenter Server connection.

Thanks.

[ThomasMc]

Ah right, seeing this post make some sense, here is my setup

Backup Job 01
DC01,DC02,SQL01,TS01,CA01,SP01,FS01

App group DC
DC01,DC02

SB Job 01
App Group = DC
Linked Job = Backup Job 01

Job Start -> Power-up App Group -> Remaining VMs in linked job are powered up and tested individually and powered down -> email sent and App Group stays online

Because I like this behavior I'm happy to have the App Group still running as its just a case of stopping the job in the morning but don't get why if your App Group VMs are part of the linked backup then they stay online

[chi-ltd]

I have added the esxi host initially into veeam and was sworkign fine.

added vcenter server onto the network and noticed the orphans.

now after some testing ihave issues with the virtual labs.... it thinks one of my offline servers (cloned and testing purposes) is the host for the v_lab...

will try and add the vcentre server and configure the backups and surebackups...

[chi-ltd]

im now getting (when connecting the the v_lab - either through esxi or vcentre server):
'Virtual Lab 'CHI V_Lab' already exists in database.'

Now on a surebackup im getting:
'The object has already been deleted or has been completely created'

[Gostev]

I would recommend to delete and re-create virtual lab, as populating Veeam Backup with vCenter server after you had everything configured and working through standalone hosts may "confuse" the product.