I have one query. I have one disk group with 5 volumes, which are more than 2 TB and are configured for replication. Now i want to add 2 more volumes in the RDS but it will resync all the volumes after adding these 2 volumes, so is it possible to make second RDS for only those 2 new volumes.
Also can we use same IP for replication.
SFW VVR actually does not syncronize the volumes automatically when adding them to an existing RDS. The VEA GUI pops up a big (almost full screen) warning that the data on the newly added volumes will not be synced automatically.
Now, this is a problem if there is data on the newly added volumes and would require a full resync of all volumes in the RDS. However, if there is no data on the newly added volumes, then all you need to do is quick format the volumes and start putting data on them. The writes done to do the quick fromat will be replicated to the new volumes at the secondary so it will be insyc after the quick format completes (at least from an NTFS perspective.)
Just to clarify - it sounds like you have not added the volumes yet and therefore what you need to do is:
Hi Mike and Wally thanks for reply,
But the new volumes are already created and contains some data, so is it possible to create new RDS for only those two volumes.
You could create a new RDS and it can share the same IPs, but the volumes in the new RDS will not maintain Write Order Fidelity with the existing RDS and so this will cause issues for some applications, like databases, but if ti is not important that volumes in each of the 2 RDS have EXACTLY the same point it time, then this is fine.
We are using Enterprise vault archieve data.
also, can explain "but if ti is not important that volumes in each of the 2 RDS have EXACTLY the same point it time" this.
Lets call your exising 5 volumes data1 and your 2 new volumes data2:
If you are just writing files and so you write
file1 at 13:02:03 in data1
file2 at 13:02:04 in data2
Now if data1 and data2 are in the same RDS, then at your DR, file1 will ALWAYS be written before file2, but if data1 and data2 are in a different RDS, then it could be that at your DR site, you have file2 and not file1 if the rlinks are async. For files this probably doesn't matter, but for databases it is different.
So for a database suppose you write:
13:02:03 - Start transaction
13:02:05 - Create link to data X
13:02:09 - End transaction
13:02:07 Write data X
Now if data1 and data2 are in the same RDS, DR will be in the same time order as primary site so suppose DR is a time point 13:06 when primary crashes, then "End transaction" has not be written so the database will rollback everything in transaction and so delete link to data X.
But if data1 and data2 are in a different RDS, then it could be that at your DR site, you have "End transaction" , but have NOT written data X and the database won't roll back and delete link to data X because as far as the database is concerned it has "End transaction" and therefore when database tries to link to X, it will fail which will probably cause the calling application to crash.
We are using Enterprise vault server and the two new volumes contain EV archive data new partitions and the other 5 volumes contain index and archieve data but all the partions are closed. Now one of the old 5 volumes only write indexing data, so is it possible to make 2nd RDS in EV. Also no database is installed in that server.
I don't know Enterrprise Vault, so I couldn't say, but as in my posts above, ft it matters if the existing 5 volumes are at a different point in time than the 2 new volumes, then they should be in the same RDS, but if it doesn't matter, then you can use separate RDS.