Comprehensive data protection for all workloads
Post Reply
1-0-1
Enthusiast
Posts: 54
Liked: never
Joined: Nov 30, 2010 1:38 pm
Full Name: Bernd
Contact:

Perform initial replication over this removable storage

Post by 1-0-1 » Feb 15, 2011 2:25 pm

I have tried this option only once but wanna test it again. However there are a ouple of questions regarding this option:

1. Once the initial replication is done and the replica files have been copied to the respective folder (as per README.TXT), do I have to introduce the replica into the inventory? If I do not vSphere will not see it or does the next replication job take care of this?

2. If I have a replica Job with mulitply virtual servers for which one needs to do a full re-sync - i would believe this would not be possible. Is it safe to assume then to batch replica jobs together based on how much the link to the destination can handle? Would it be then safe to assme that creation a single job for each of the 16 virtual servers (in my case) would be the way to go. Problem I see with this however is that if several jobs overlap it will most likly effect general throughput to the destination and would create a scheduling nightmare also.

Vitaliy S.
Product Manager
Posts: 22697
Liked: 1498 times
Joined: Mar 30, 2009 9:13 am
Full Name: Vitaliy Safarov
Contact:

Re: Perform initial replication over this removable storage

Post by Vitaliy S. » Feb 15, 2011 4:39 pm

1-0-1 wrote:1. Once the initial replication is done and the replica files have been copied to the respective folder (as per README.TXT), do I have to introduce the replica into the inventory? If I do not vSphere will not see it or does the next replication job take care of this?
A subsequent replication job will take care of it, just place all the files to Veeambackup folder located on the destination datastore as it is explained in Readme.txt file
1-0-1 wrote:2. If I have a replica Job with mulitply virtual servers for which one needs to do a full re-sync - i would believe this would not be possible. Is it safe to assume then to batch replica jobs together based on how much the link to the destination can handle? Would it be then safe to assme that creation a single job for each of the 16 virtual servers (in my case) would be the way to go. Problem I see with this however is that if several jobs overlap it will most likly effect general throughput to the destination and would create a scheduling nightmare also.
If you have multiple VMs added to the same job, then all VMs will be affected if you choose to perform a full run.

If I were you I would group VMs based on their production importance. In order not to have an overlap I would daisy chain replication jobs with the help of a Post-backup job action/PowerShell or Windows Scheduler.

Please have a look at this topic for further reading: Help! How can execute 7 job in sequence?

Thank you!

Post Reply

Who is online

Users browsing this forum: No registered users and 7 guests