EVA array's remote replication

[Pearl Park]

I have a primary disk in EVA array 1 replicated to a secondary disk in EVA array 2.
After replication, both primary and secondary disks have same name and same WWN(World Wide Name). I am wondering whether this is by design or a bug.
Same name for both primary and secondary disks do not bother me. However, I don't understand why same WWN is used for both primary and secondary disks.
Can anyone comment?
Thanks.

[Ivan Ferreira · Accepted solution]

I assume that this is by design. Some operating systems (like tru64) match the devices names against the WWN of the LUN. If you should perform a failover, and the WWN change, device files names will change and this will lead in umountable volumes.

[Steven Clementi]

Exactly. In a windows cluster environment, it does not matter too much as long as the disk signatures are the same, but on various unix/linux systems.. different UID's could mean your cluster does not work.


I am almost sure this is by design since, well... think about it.... You are "replicating" your Virtual Disk. Everything should be the same so that when there is a failure, the host has no idea that anything is different.



Steven

[SAKET_5]

Pearl,

Agreed with the above two responses - just like to add that you can change the replicated Vdisk UUID to a valid one that you fancy - by default the replica Vdisk on your secondary EVA is not presented to any host and is in read only mode and the Vdisk UUID field can be edited at this change to change the value.