I have encountered an issue with some of our Exchange servers. Without warning and without anything occurring that I am aware of, the logs and database drives go offline. There are no events in the event logs that refer to a volume going offline, only Exchange complaining that the DB and logs are unavailable.
They are drives going offline are Raw Device Mappings to Windows 2008R2 virtual machines running on ESXi 5.5 update 1.
I have looked this up and many articles relate to the Diskpart SAN profile. This is all well and good however, the disks are not offline when the server reboots, they just go offline at random times. You can right click on them and bring them online and all is well and even then reboot and they stay online. And then unexpectedly users will complain that they cannot access their emails and then we notice that the drives have gone offline. Its baffling us!
We have many VMs with Windows 2008R2 and with RDMs and we have never seen this activity before.
Could this be the DAG cluster?? Could it be Exchange??
Any help or advice would be most welcome.
Software/Hardware used: Windows Server 2008 R2,Exchange 2010, cluster,DAG, ESXi 5.5