VMware Cloud Community
einstein-a-go-g
Hot Shot
Hot Shot
Jump to solution

vSAN weirdness.....Policy ?

Some vSAN weirdness, all was well with the BETA, and all was well for a few weeks, on another test lab.

So rebuilt, servers, and setup vSAN from new....and all gone pair shaped, the Disk Group was missing from one of the servers from the GUI, and only started 2 eligible, so thought this was wrong, and removed all groups, removed all vSAN, and removed all Hosts from Cluster....

started again, a little better, all disk groups correct across hosts

3 x hosts

1 x SSD (real not fake) 1 x HDD (magnetic) as per minimums

(have many more hosts up to 64, I know max is 32), and more SSDs and HDDs to play with, but wanted to test scaling.....on the fly....

vSan2.jpg

vSan1.jpg

but whatever I deploy a VM, get an error....

vSan3.jpg

and I puzzled!!!

Reply
0 Kudos
1 Solution

Accepted Solutions
einstein-a-go-g
Hot Shot
Hot Shot
Jump to solution

Okay, that could be the issue, on esxdev001, using esxcli vsan storage list

Used by this host: true

   In CMMDS: false

for both SSD and HDD magnetic.

I assume just remove the disk group and re-create?

Message was edited by: einstein-a-go-go Okay, fixed.... This has been a weird day, because I've had three different Clusters with 5.5 U1, do the same thing.... this:- I also noticed that esxdev001, was in a weird state, it would not accept vMotion migrations, complained that the "host could not do this in this state", although it could power on VMs... It would also not go into maintenance mode, and move VMs off.. So a quick, reboot/restart and checking again Used by this host: true    In CMMDS: true I think it's going to be a while, before I get a nice warm feeling about vSAN, before Production! Thanks Guys Duncan & CHogan! PS I should have done, what I tell users, "just turn on and off" is it fixed!

View solution in original post

Reply
0 Kudos
6 Replies
depping
Leadership
Leadership
Jump to solution

Thanks, I would like to see a couple of other things:

  • The VM Storage Policy you create
  • Please run "RVC" command:
    • vsan.cluster_info
    • vsan.check_state
einstein-a-go-g
Hot Shot
Hot Shot
Jump to solution

policy....

just 1

vSan4.jpg

get the other info now...

Reply
0 Kudos
depping
Leadership
Leadership
Jump to solution

nothing strange there either... hmmm

Reply
0 Kudos
CHogan
VMware Employee
VMware Employee
Jump to solution

Can you run "esxcli vsan storage list" and check the "In CMMDS" for each disk? It should be set to true.

Alternatively, from RVC, run "vsan.disks_stats" and check that each host is indeed contributing an SSD & HDD.

Maybe the UI isn't reporting correctly. The error certainly suggests a lack of storage somewhere.

http://cormachogan.com
einstein-a-go-g
Hot Shot
Hot Shot
Jump to solution

Okay, that could be the issue, on esxdev001, using esxcli vsan storage list

Used by this host: true

   In CMMDS: false

for both SSD and HDD magnetic.

I assume just remove the disk group and re-create?

Message was edited by: einstein-a-go-go Okay, fixed.... This has been a weird day, because I've had three different Clusters with 5.5 U1, do the same thing.... this:- I also noticed that esxdev001, was in a weird state, it would not accept vMotion migrations, complained that the "host could not do this in this state", although it could power on VMs... It would also not go into maintenance mode, and move VMs off.. So a quick, reboot/restart and checking again Used by this host: true    In CMMDS: true I think it's going to be a while, before I get a nice warm feeling about vSAN, before Production! Thanks Guys Duncan & CHogan! PS I should have done, what I tell users, "just turn on and off" is it fixed!

Reply
0 Kudos
joergriether
Hot Shot
Hot Shot
Jump to solution

Hmmmm...i wonder how this could happened in the first. You have not per accident "cloned" some ESXi hosts?

Best regards,
Joerg

Reply
0 Kudos