hi
last night we had a scheduled site failover
1 dag, 2mbx servers in site a and 1mbx server in site b, dac configured
each site 1 cas/hub server
all exchange 2010 sp2 ru3
when i went to run the commands to activate the mbx server in site b i ran into a problem with powershell when running stop-databaseavailability uk-dag-1 -activedirectorysite a -configurationonly, also tried without configurationonly switch
it came up an error caused a change in the current set of domain controllers, i have only seen this reported when using emc not powershell
so i restarted the domain controllers and the cas/hub and mbx server but to not avail
i failed back to site a and dag came up
i attempted it again and this time the stop-databaseavailability command worked fine
i ran stop-service clussvc and fine too
but restore-databaseavailibilitygroup uk-dag-1 -activedirectorysite b failed.
i received this error
WARNING: The operation wasn't successful because an error was encountered. You may find more details in log file
"C:\ExchangeSetupLogs\DagTasks\dagtask_2012-10-19_22-26-16.635_restore-databaseavailabilitygroup.log".
File share witness resource '\\uk-cas02.domain.com\UK-DAG-1.domain.com' couldn't be brought online. The current state is
'Failed'.
+ CategoryInfo : InvalidArgument: (:) [Restore-DatabaseAvailabilityGroup], DagTaskFileShar...OnlineExcept
ion
+ FullyQualifiedErrorId : 431F9449,Microsoft.Exchange.Management.SystemConfigurationTasks.RestoreDatabaseAvailabil
ityGroup
i then tried setting up another FSW on our 2008r2 server and added in the exchange trusted subsystem group as local admin but i got error saying that this group was not part of local admin group (which was not true)
i tried it setting it up on another 2008 r2 server but same issue
after trying it about 5 or 6 times i noticed that email started to flow and the mailstores mounted. result i thought
but when i checked the cluster status it states:
the cluster network name is not online
quorum configuration: warning: cluster is running on forcequorum state. this could result....
the FSW folder exist on the file server but there is no data inside it?
i did a failover 1 month ago without any hiccups and now this
so i have 3 problems (well 4 if you count the original problem about an errror caused a change in the current set of domain controllers)
1) i need to failback this evening, can i use the same procedure as usual by organising downtime or will this forced quorum issue cause problems
2) how do i get rid of all the failed FSW setups? i tried various commands
3) i need to test this again next month but need to make sure the FSW is working okay, how do i get this working correctly on the original hub/cas server which seems the most obvious place
lastly what is the risk of running a forcedquorum state?
i had quite the night and it took me nearly 5 hours to get the DR site up and running which usually takes about 20 mins.
many thanks