Mapping new v6 job to v5 replicas and v5 replica leftovers

[mwpreston]

Hi,

Just wondering if anyone knows if I am doing this correctly.

I've upgraded to v6 and I'm looking to upgrade my old v5 Legacy replication jobs to new v6 jobs. I've created a new job, mapped my target vms to my older replica vms. Everything is working great, it's using snapshots now on the replicas and things look awesome! Only issue I know see is with the older rollback points. They are not being deleted based on my retention policy. I'd really like to reclaim this space and I'm not sure what I need to do.... Do I need to delete them out of my replica's section? If so will that not attempt to delete the complete VM? Or do I need to go and manually delete the older vrb files and if so what all is safe to remove and what isn't?

Hopefully someone can help...

thx

[Gostev]

Hi Mike, congratulations on successful upgrade and glad you are liking the new replication. Please feel free to delete all remains of v5 replication manually with the vSphere Client's datastore browser. Since you have created the new job, those VRB will not be removed automatically according to retention. Thanks!

[mwpreston]

Great....I was hoping there I could get rid of them....thanks!

[reddrinker]

Hi,

I am in the process of doing the same thing and so far it is going relatively smoothly. When you say delete the files through the datastore browser, which files exactly do I delete?

Thanks

[mwpreston]

I deleted all of the .vrb files from the older v5 jobs.

Also, i went into Veeam and just Removed the old replicas in order to get them out of the database....

Mike

[Gostev]

That's right.

Also, replica.vbk is OK to delete.

Just don't delete any files with _working.vmdk in the end, if you have such. These are descriptor files pointing to the real VMDK file (they are just a few KB in size despite datastore browser shows them as having the size of VMDK they are pointing to). If you delete them, you will delete actual VMDK instead (along with this descriptor file).

[kingpinwop]

When you say "mapped my target VM's to my older replica VM's" how exactly are you doing this? Is this in Veeam or right from VMWare that you're doing the mapping? I also am looking to move my v5 replica job to v6 but seem to be stuck on this part. Thanks

[Gostev]

It is the new v6 feature called replica mapping. Essentially, new and easy way to seed the replica, when you already have one created in the DR site. Please refer to the User Guide (page 111) for more information.

[foggy]

And just in case the user guide description is not completely clear to you, here is a more detailed seeding procedure.

[kingpinwop]

Thanks guys, that explanation made it a little more clear. I agree that the documentation that you referred to is not as clear as it could be. In fact I think that the replica mapping should not be triggered by the 'slow connection' check box as it is misleading. My WAN connection is not slow, so intuitively I would not choose this box, however would like to use the seeding/mapping feature that it triggers. I suggest that it could be laid out a little more clearly as it's own check box for those of us that do not associate seeding/mapping with a slow connection.

Thanks for your replies, I am running my new replication job now and we'll see how it works vs the legacy job that has been working well for the last long while.

[Gostev]

Thanks for your feedback on the user interface. Please note that the checkbox does not just say "slow connection", but also explicitly says that it enables replica seeding functionality, so I do not see how it can be more clear than that.

Regarding "slow connection" part, this is intended for new users. We found that most people who are new to replication do not have a clue what "seeding" is. Most people do not know if they want to "enable seeding" or not, however everyone can answer the question "is your WAN connection slow"

Basically, we do not want to force first-time users doing test replication between two local hosts on learning various advanced terms and options - we want them to be able to go through the replica in 1 minute, and get replica created in a few minutes

[kingpinwop]

Your point is well taken on the new user aspect. I think that from a UI standpoint for a seasoned user, asking me to 'Describe My DR site:' I would not classify my DR site as slow. All I want to be able to do is map a production VM to a replica at my DR site that is already seeded. That is not an option, of course, but that's what I was looking for as a user. I don't think the 'Low connection bandwidth' option is as clear as you may think. My 2 cents....

Great job with the new v6 and all the work you guys are doing at answering all the questions flooding the forums the last week!