Copy Backup Objects between Jobs

[frowiers]

Hi

I am doing some redesigning of our backup jobs, and want to merge multiple jobs to larger jobs.
Being an IT-pro, doing this manually is just something I don't do, so I turned to Powershell to automate this process.

Using Get-VBRJob, Get-VBRJobObject and Add-VBRHvJobObject I have tried automating this process, but it seems like the objects I get when doing "Get-VBRJobObject" isn't usable as Entity in the "Add-VBRHvJobObject" cmdlet.

What I have done to make this happen using powershell is doing a Find-VBRHvEntity first, then using the Add-VBRHvJobObject after, but this is enumerating the cluster every time it is doing the find (it takes between 60-120 seconds for every find). This means adding 60 VMs takes me between one and two hours - automated still, but very slow!

Any help in writing the proper script/cmdlet to add all objects from one job to another (which goes fast) would be highly appreciated!

Thanks!

[veremin]

Have you though about creating a hash table (as described here) to run the said query only once? Thanks.

[frowiers]

Hi, and thanks for quick reply

The problem I am having, is that for each "Find-VBRHvEntity" it is enumerating the cluster, which is causing each line to take "forever" (in this case forever equals to 1-2 minutes). So it have just taken me 90 minutes to loop between a table of 50 objects.

Is there no way I can do this faster?

Please give me some examples if possible - thanks!

[veremin]

Kindly, check the link I've provided - it contains an example.

The idea here is to run Find-VBRHvEntity only once, assign its result to a hash table ( $vms=@{} ) and, starting from that moment, work with the hash table, instead of querying infrastructure directly.

Thanks.

[frowiers]

Sorry if I am being a bit slow on this one, but how do I run the Find-VBRHvEntity against a specific (already exsisting) backup job?

"Get-VBRJob -name "Job Name" | Get-VBRJobObject | Find-VBRHvEntity" don't work, as Find-VBRHvEntity can't take pipeline input.

Any idea?

[veremin]

So, the process would like the following:

1) Get an old backup job
2) Get its source objects
3) Find VMs in virtual infrastructure which correspond to the gotten objects (object references comparison)
4) Get a new backup job
5) Add objects returned on step 3 to the new backup job

Let me illustrate it with simple example that doesn't contain cycle (necessary to get all source objects) and hash table (required to speed up the process):

Code: Select all

asnp VeeamPSSnapin
$OldJob = Get-VBRJob -Name "Name of an old backup Job"
$JobObject = Get-VBRJobObject -Job $OldJob  -name "Name of source object"
$Server = Get-VBRServer -Name "Name of cluster"
$Entity = Find-VBRHvEntity -Server $Server | where {$_.Reference -eq $JobObject.GetObject().ObjectRef}
$NewJob = Get-VBRJob -Name "Name of new Job"
Add-VBRHvJobObject -Job $NewJob -Entities $Entity

Thanks.

[frowiers]

Hi again
Thanks, that got me going!

This is what I ended up with:

Code: Select all

asnp VeeamPSSnapin
$clusterName = "Cluster Name"
$NewJobInput = "New Job"
$OldJob = "Old Job"

$vms = Find-VBRHvEntity -server $clusterName | Where-Object {$_.Type -eq "Vm"}

$OldJob = Get-VBRJob -Name $OldJob
$JobObject = Get-VBRJobObject -Job $OldJob
$Server = Get-VBRServer -Name $clusterName
$Entity = $vms | where {$_.Reference -in $JobObject.GetObject().ObjectRef}
$NewJob = Get-VBRJob -Name $NewJobInput
Add-VBRHvJobObject -Job $NewJob -Entities $Entity

Thanks for helping me out

[veremin]

Since you're passing cluster name directly to Find-VBRHvEntity cmdlet, this line is not needed:

$Server = Get-VBRServer -Name $clusterName

Thanks.