Hello,
Somewhat new to the forum. Thanks in advance for any thoughts on this...
Situation:
Windows 2012 R2 failover cluster, 2 node, VMWare vm's, on ESX 5.5, 8gb memory each server. This is a file server cluster, aka, only the File Services role is used. More specifically only 3 File Server roles have been created and because of the issue only 1 of those roles is being used. And even that role has very limited users putting files on that share. So a pretty basic low use cluster right now.
What is happening is over the course of time (~2-4 weeks) the node with the active roles eventually runs out of memory. I hard reboot via the VMTools, roles failover to the other node and works fine for ~2-4 weeks, then repeat process.
What I am seeing in Task Manager is 22 instances of Failover Cluster Resource Host Subsystem, each using a various percentage of memory. I thought maybe every time an administrator on my team opened Failover Cluster Manager that maybe it wasn't releasing that "instance" when the exited. But in testing that does not seem to hold water. I open FCM and memory goes up but I don't see another instance of Failover Cluster Resource Host Subsytem get created...so that leads me to believe it is something else.
The OTHER thing that I see is when I open Failover Cluster Manager, there is an hour glass over the top of the clustered server icon of the cluster name...tiny image here:
This hour glass never goes away, but I cannot figure out what it is doing but it must be related IMO.
I have check the list of cluster hotfixes, nothing seemed relevant.
Any thoughts? Other info I could provide to help you help me?
Thanks,
Jeff