Replication Job to SAN

[sbohannan]

I am having an issue with a replication job that i setup to a new SAN for our DR site.

The Replication job backs up a 4TB server and would take a couple of weeks to replicate across our internet connection, so to help the job along i setup the SAN in our DATA center and ran the job to get the server on the SAN. Once the server had replicated several times i shut the SAN down and moved it to our DR site, added the storage to the vmware server and then added the replica into inventory.

Here is where my issues starts.

When i run the replication job the job is not seeing the seeding of the old server, so it is trying to replicate the server as a new job. I have checked seeding on the job and it is set to the correct server.

Any ideas thoughts?

Thank you
shane

[veremin]

Chances are, everything got screwed after SAN (along with VMs it hosts) had been added to new vCenter/ESXi server. Try to leverage replica mapping functionality; should be a way to go. Thanks.

[sbohannan]

that is what i was afraid of.

Thank you for the link, i will recreate the replica and ship it to the location.

shane

[veremin]

If replica VMs are already in DR site, you don't need to start everything from scratch - just set up the mapping rules. This way, only changes between production and target VMs will be transferred. Thanks.

[sbohannan]

well that is what i was thinking when i moved the SAN to the site, but when i try to run the job it is slow on "Calculating digests for Hard disk 2 (4TB). By slow i mean i have let the job run for 32 hours and only get to 75% on Calculating. Is that normal?

i put a call in to support yesterday and they check a couple of things and said it was not seeing the seeding and was trying to replicate a new server. If i let the job run and complete it holds up a job that i have running locally.

Shane

[veremin]

By slow i mean i have let the job run for 32 hours and only get to 75% on Calculating. Is that normal?

Digest calculation might take some time indeed, especially for big VMs. However, it's more or less one time operation, and subsequent cycles should run much faster. Thanks.