I have a customer that needed to rebuild the Windows VM that was acting as their Proxy/Repository server at their secondary site. He is using all of the same IP information on the new server. The repository itself is on a NetApp E-Series access via iSCSI. Once the server is rebuilt, what is the process for getting it back into the B&R server. If the server is the same name, will the B&R server just need to push out the Veeam proxy/repository services to the rebuilt VM or will we need to delete and re-add the proxy and repository to Veeam? If we discovery the repository LUNs on the new VM via iSCSI will we need to re-scan the repository to discover the backups or will they just be there?
If there is a process document for doing this it would be appreciated.
Page 1 / 1
@dmarkham ..you ‘may’ be able to have the customer install components manually, then rescan the server; but the “cleanest” way would be to remove the server currently in VBR, then just re-add it as a Proxy & Repo. Reason is because services/installers needed for each VBR component will be done automatically (Installer Service, Transport Service, Mount Service, etc). The standalone installers are located in C:\Program Files\Veeam\Backup and Replication\Backup\Packages, but you may install a component not needed. It doesn’t take long to remove the component and re-add it. My recommendation (& what I’ve done in the past).
Cheers!
And at the definition of the repo check the option to import the existing backups.
Not much to add to this one as @coolsport00@JMeixner have pretty much covered it. The repo check and import is a good thing to note.
Thanks for the response. I was thinking the same thing. The other question I have is whether I would have to delete the repositories currently on that server and then recreate them and rescan to discover the backups on those repositories? It seems likely it will not let me delete the repository server from the B&R server when there are repositories defined on them.
Thanks. Sounds like my second question was just answered as well.
If the hostname, DNS name and the IP configurstion is the same, I would to try first to do a rescan on all the components. Beginning with the managed server and then going to proxy and repo...
NO! Do *not* remove your backups currently residing on your volumes! At the end of adding the Repo, you are presented with scanning it for existing backups as @JMeixner stated above.
Cheers!
Ok. I understand that part. I was talking more about the repository definition in the B&R server.
Trying this in my lab. The challenge I see is that I cannot remove the server/repository in the B&R server while it is in use by backup/backup copy jobs. I cannot remove it from Managed Servers either while it is defined as a repository/proxy server. So I would have to change all backup and backup copy jobs to not use that proxy or repository prior to delete and re-adding the server in Managed Server and as a Backup Proxy and Backup Repository in the infrastructure.
Am I missing something here?
I totally agree with what @coolsport00 and @JMeixner already mentioned. The automatic process (deleting and adding the component) gives a better check that everything regarding the components is correct installed and running.
Regarding your latest post @dmarkham , I would suggest to create or use a temporary repository. So disable all the jobs you have to change to the temporary repository. Afterwards delete the server which you haverebuilt, add it afterwards and change at last the disabled jobs towards the rebuilt server.
That is correct. You will need to temporarily assign a Repo to your backup/copy jobs, and recommended to also temp disable all jobs while doing this for, before you can remove it. If you use a single Proxy/Repo for several jobs...yes, it is a bit tedious to do. But again...better than attempting a manual process with potential for misconfiguration.
Cheers!
Got it thanks. Will give it a try and let you know.
Sooo good to see this community helping customers of Veeam on this way!
Thanks for everyone’s help. Final process was:
Disabled all jobs and backup copy jobs using the repositories on the rebuilt server.
Deleted the Proxy on that server from B&R server and then re-added which pushed the required proxy components to the rebuilt server.
Added a temp repository to C:\backups on the rebuilt server which pushed the required repository components to that server
Re-discovered the old repository server LUNs via the iSCSI initiator on the rebuilt server and mapped them to the appropriate drive letters.
Re-scanned the two repositories that were on the old server and now on the new server. This successfully discovered the repos.
Re-enabled backup and backup copy jobs and they appear to be working now.
Glad to hear you fixed it and also listed the steps to help others.
Fine that you managed to get your “new” repo server working
