Calculating digests -- v6 replication

[averylarry]

Well, it appears that I won't really be able to use my old replicas after all. The best that I can tell -- each entire replica at the destination site has to be read to calculate the digests. Oh well. At least my remote location is only 10 miles away.

[averylarry]

Correction -- I can use the replicas, I just can't take off an run with it as it is. I have to calculate the digests either by starting over and re-seeding, or by bringing the remote server local. Unless I'm missing something.

[averylarry]

Never mind. Looks like if you create a backup repository at the remote site for the "replica metadata", then the remote server takes care of it.

I just had it misconfigured. (It's not like I've read the user guide, yet.)

[foggy]

Yes, it's better to install a backup repository server close to the source as it stores and processes replica metadata.

[Gostev]

As per description right in the UI, backup repository located in the source site (close to the protected VM) must be used for replica metadata. Also, backup repository placement does not affect the performance of digest calculations (in other words, scanning content of target replica VM disk). What may affect digest calculation performance is target proxy selection (don't except good performance when source site's proxy is pulling replica VM data from the target site).

[crichardson]

So, correct me if I'm wrong.

When replicating a VM over the WAN to our DR site we should be storing the replica metadata at the DR site as well? I've been storing the replica meta data on one of our repositories at our primary site Come to think of it, that's probably not good...

[Gostev]

You are wrong with your statement, but you are doing it right today.

[cmgowcity]

I am storing my replica metadata at the source side. I have the seed at the target side in the repository but my digest is taking hours over a 10Mb line. Am I missing something? The seed is only 1 day out of date and there should be very little changed data.

Thanks

Colin

[foggy]

Colin, do you have proxies set up at the both sites?

[cmgowcity]

Sorry, I missed the reply to this post. I have proxies setup at the source and destination sites and the job is set to manually proxy selection. I am still having this issue if I use an existing VM as the seed at the target side.

Regards

Colin

[foggy]

Shouldn't be the case if everything is configured correctly. Worth opening a case for the support person to review your configuration.

[cmgowcity]

Will do, but the last time I used the European bases support, it was terrible.

[foggy]

You always have the ability to involve the support management if you don't feel your case is handled properly (please use the Talk to a Manager button available on the support portal).