Hi,
I have a customer with a massive EV environment. we are migrating the whole environment to new servers for upgrade purposes of both EV & VCS.
At the moment, the environment is installed on veritas cluster using the geo-cluster feature between the production site and the DR site.
Due to some internal issues, the whole geo-cluster thing is not working. and we need to think of another solution.
We thought of the following:
Creating 2 separate clusters, not using the geo-cluster feature at all, one will be in production site, one will be in DR site.
Users have access to both sites.
So, the plan is - on the production site, do the whole DR procedure of the EV so it will be on the new servers.
Now - data is being replicated to the DR all the time, so I have Indexes, Storage location & SQL DB's on the DR site as well.
The thought was to take down the production site, and re-configure EV on DR site as well, and then bring it back to production site.
The difference in the DR site is that the storage name is different.
So, if in the production the partition entry is \\prod\ev partition1 on the DR it will be \\DR\ev partition 1
This can be handles by a simple SQL UPDATE command that will change the PartitionEntry table before DR site is going up.
Index will be mapped on DR with same drive letter, if not - we can update SQL as well.
Directory name will stay the same on both production and DR (being taken care of by SQL people)
So - I'm trying to think if there is any reason for the data not being accessible to users if I'll go with all that was mentioned above.
If the users are contacting the directory, and the directory name is not changed at all between sites - this should work right?
We are trying not to have to use the whole update DNS issue.
Anything else I might be missing here?
Any ideas and questions will be highly appreciated.
Sarah
