PSP_RR act as Fixed

[LondonD]

Hi

I have a cluster with 10 hosts.All the hosts datastere PSP is set to Round robin.

The storage guys complained the paths are not balanced. What I noticed on 2 hosts, is that all the datastores are set to (VMware)Roundrobin,

but if I look at "Manage paths" I see 3 Active Paths and 1 Active (I/O) path.

If I change 1 datastore to Fixed and back to RR, all 4 paths show up as Active(I/O)

There are 80 datastores, what is the quickest way to fix this?

Please help

[PduPreez · Accepted solution]

Hi

Not sure why this happens, but I have seen this behavior before. In my case it was after a complete power outage.

If you reboot these two host the paths should correct themselfs. worked for me

Other Option is to run a script to change all Datastores to Fixed, and then back to round robin, to save the manual proccess.

But changing the policy to fixed, you need to specify the prefered path manually, so i was not able to get a script to accomplish this.

Hope it helps :smileycool:

[LondonD]

Hi

The reboot has done the trick. All paths showing as Active(I/O) on those 2 hosts now

Thanks!