XOSOFT Replication: 15.0, 16.0
Last Modified Date: 01/09/2012
You are either trying to create a new exchange scenario or trying to run an existing exchange scenario and you are seeing an error that there is a mismatch between the master and replica and it also lists that 1 of the 2 servers has no stores and or no storage groups.
If your exchange servers are 64 bit this is generally an issue with our helper service not properly reading your exchange configuration for the server in question with no stores/storage groups listing. It can also be caused by an ldap query failure.
If the helper service (called CA Xosoft helper or Arcserve RHA helper) on the server in question is running, try stopping it, if it is stopped, try starting it.
Then try hitting back and do autodiscovery again, then hit next and do the replica configuration page again. If you are trying to a run an already created scenario and you are seeing this mismatch issue try running it again.
If this does not resolve the problem the issue may be related to the ldap query failing.
If you use a different name to refer to your domain name (known as a disjointed domain) you may have to edit our config file and include the alternate domain name so we can properly query ldap. To verify if you have a disjoint domain name check the FQDN of the master and replica in the scenario. Go to the high availability properties tab and expand 'hosts', make sure what is listed after the master and replica computer name matches the name of the forward lookup zone in DNS. If it does not this is a disjointed domain and you will need to specify the alternate name of the domain or dns zone in our config file.
To do this do the following:
- Open up the ws_rep.cfg file located in the engine directory in the install path of our software on the master and replica (open in notepad)
- Find the 'AssuredDomainName' parameter and set it to equal the alternate name of your domain (e.g. AssuredDomainName = alternatename.com)
- Remove any # symbols from that line, save the files, and restart the engines and try auto discovery again
We may also fail to query exchange on the server in question if permissions are not correct. Make sure both the CA ARCserve RHA engine account and the helper service account on both servers are not only the same but are logging on with a domain admin level account.
For an exchange HA scenario in addition to our service account(s) being a domain admin the RHA account should also have the following permissions:
Exchange 2003: Exchange full admin
Exchange 2007: Exchange org admin
Exchange 2010: Organization management group
If this is exchange 2010 and you are having mismatch issues it could be from an issue we identified trying to correct mismatches related to the exchange prefixes, we have a patch for this so please contact technical support for the latest engine which will correct this. The exchange prefixes must match between the master and replica. If the master's first store is using prefix e01 instead of e00 we may have an issue fixing this type of mismatch because exchange is going to always by default want to assign e00 to the first store it creates on the replica, if it assigns e00 to the first store on the replica while the master's first store is e01 we will not be able to fix this type of mismatch. Generally this can be resolved by removing all stores from the replica then manually creating a temp store so exchange assigns e00 to it, then simply run the scenario creation wizard again and then the first store we create on the replica will get assigned e01. After wards you can then delete that temp store on the replica from the exchange manager.