Backup/Replication over MPLS Qs

[mcmcomput]

I have several sites with servers I have been playing with replicating back to our DC for backups. Below are my questions. The sites are inter-connected with T1/1.5M-3Mb lines. All servers are ESXi v5.1 and i have the latest Veeam Enterprise from February installed and running. Both DC and remote sites have backup proxies installed and running.

• 1Q. I can do a Replication with up to 28 restore points, but how do i go back to point 15 and boot the machine up? it appears I have to consolidate or delete some snapshots on ESXi? I think i got file level recorvrty down and comfortable, just not restoring a whole VM from the last 28 points.

• 2Q. I have started both a backup and replica from 2 different sites with approx 80-100Gb at both. The replica i have seeding turned on and appears it picks up where it left off. However the backup appears to start from scratch each time? Is this right? The first backup will take a long time, so each night i start it then stop it during the day to re-start it again.

• 3Q. Is there any gotch-yas i should be aware of using the backup feature and 90 points or more? Do i need to re-run full backups periodically or change the synthetic setting?

• 4Q. I have an LTO 2,3,4, & 5 i'd like to use, but am a bit sure how to toss them in the mix exactly. From what I can tell i need a backup or replica job to start with then tell it to Copy? and that job is what i send to tape? My tapes are on 2 separate physical servers so think i need to install Veeam's suite on the other two?

If you have any questions about the architecture or recommendations to make my life or job easier, let me know. Maybe i am setting some of these up incorrectly.

Thanks!

[veremin]

I can do a Replication with up to 28 restore points, but how do i go back to point 15 and boot the machine up? it appears I have to consolidate or delete some snapshots on ESXi? I think i got file level recorvrty down and comfortable, just not restoring a whole VM from the last 28 points.

Are talking here about testing or actual failback/failover operations? In the latter case, all you will need to do is to select required restore point and VB&R will do the rest.

I have started both a backup and replica from 2 different sites with approx 80-100Gb at both. The replica i have seeding turned on and appears it picks up where it left off. However the backup appears to start from scratch each time? Is this right? The first backup will take a long time, so each night i start it then stop it during the day to re-start it again.

You mean you were able to seed successfully the replication job, but not the backup one, right? Or that replication job works incrementally, in contrast to the backup one?

Is there any gotch-yas i should be aware of using the backup feature and 90 points or more? Do i need to re-run full backups periodically or change the synthetic setting?

Please, see the best practice regarding frequency of full backup.

I have an LTO 2,3,4, & 5 i'd like to use, but am a bit sure how to toss them in the mix exactly. From what I can tell i need a backup or replica job to start with then tell it to Copy? and that job is what i send to tape? My tapes are on 2 separate physical servers so think i need to install Veeam's suite on the other two?

Only backups jobs/repositories can be used as a source for tape jobs. For now the tape device should be directly connected to the backup management server. Though, some users utilize certain tools to bypass this limitation.

Thanks.

[mcmcomput]

Are talking here about testing or actual failback/failover operations? In the latter case, all you will need to do is to select required restore point and VB&R will do the rest.

I needed to do some testing on a previous copy, so not a failover, just clone a copy to our dev environment so I could boot it up and do a few things.

[ You mean you were able to seed successfully the replication job, but not the backup one, right? Or that replication job works incrementally, in contrast to the backup one?

I assume everytime I stop the backup job and restart it at night, its starting over? There is no way to pause it? I am just trying to avoid shipping drives all over the palce, even if it takes a week or two. The replication is moving fordward fine, i got one more VM to go and i'll have my replica's.

Only backups jobs/repositories can be used as a source for tape jobs. For now the tape device should be directly connected to the backup management server. Though, some users utilize certain tools to bypass this limitation.

Perfect, I will have to give that a try! I assume I can't have two machines running Veeam B&R sharing the same SQL DB and accomplish the same thing?

Thanks.

[Vitaliy S.]

I assume everytime I stop the backup job and restart it at night, its starting over? There is no way to pause it? I am just trying to avoid shipping drives all over the palce, even if it takes a week or two. The replication is moving fordward fine, i got one more VM to go and i'll have my replica's.

If you have at least one full backup, your backup job will start running incremental passes. If you want to pause your backup job, you can wait till full run is completed, then disable the job. Can you please tell me how do you see that your job starts from scratch?

Perfect, I will have to give that a try! I assume I can't have two machines running Veeam B&R sharing the same SQL DB and accomplish the same thing?

Yes, Veeam B&R servers cannot share databases.

[mcmcomput]

By starting jobs from scratch i mean clicking stop for the job and hitting retry after hours to re-start it. If i schedule a job and have it terminate if it exceeds the window and it starts the following night will it pickup a backup job where it left off if a full hasn't been completed?

[veremin]

In this case, the job would start backup cycle anew - either incremental or full one depending on corresponding schedule. Thanks.

[foggy]

I needed to do some testing on a previous copy, so not a failover, just clone a copy to our dev environment so I could boot it up and do a few things.

You can just start the corresponding restore point and perform the required tasks.