SOLVED: HA Cluster shuts down VM's rather than migrating them when rebooting node?

[jlficken]

I figured it out in case anyone ever stumbles across this....
Datacenter --> Options --> HA Settings
Change to shutdown_policy=migrate





I'm at a loss as I have another cluster set up the same way that works correctly although they're both Intel systems. I can live migrate VM's between nodes as well on the cluster that isn't working as expected.

I'm using x86-64-v3 as the CPU type for all VM's since one system is AMD and other is Intel.

All VM's have Start at boot set to Yes and all are in the Datacenter --> HA settings as being set to a state of "started" when I configured them.

ETA: Quorum remains "OK" while the other node is rebooting.


HA status:
quorum - OK
master - S4PVE1
lrm - S4PVE1 (active, current time)
lrm - S4PVE2 (active, current time)

root@S4PVE1:~# ha-manager status
quorum OK
master S4PVE1 (active, Wed Apr 10 13:53:24 2024)
lrm S4PVE1 (active, Wed Apr 10 13:53:17 2024)
lrm S4PVE2 (active, Wed Apr 10 13:53:15 2024)
service ct:117 (S4PVE1, started)
service vm:100 (S4PVE1, started)
service vm:101 (S4PVE2, started)
service vm:102 (S4PVE1, started)
service vm:103 (S4PVE2, started)
service vm:106 (S4PVE1, started)
service vm:110 (S4PVE1, started)
service vm:111 (S4PVE1, started)
service vm:112 (S4PVE1, started)
service vm:113 (S4PVE1, started)
service vm:114 (S4PVE1, started)
service vm:118 (S4PVE1, started)

root@S4PVE1:~# pvecm status
Cluster information
-------------------
Name:             S4PVE
Config Version:   2
Transport:        knet
Secure auth:      on

Quorum information
------------------
Date:             Wed Apr 10 13:54:11 2024
Quorum provider:  corosync_votequorum
Nodes:            2
Node ID:          0x00000001
Ring ID:          1.99
Quorate:          Yes

Votequorum information
----------------------
Expected votes:   2
Highest expected: 2
Total votes:      2
Quorum:           1
Flags:            2Node Quorate WaitForAll

Membership information
----------------------
    Nodeid      Votes Name
0x00000001          1 192.168.0.253 (local)
0x00000002          1 192.168.0.252

Here's the corosync.conf:

logging {
  debug: off
  to_syslog: yes
}

nodelist {
  node {
    name: S4PVE1
    nodeid: 1
    quorum_votes: 1
    ring0_addr: 192.168.0.253
  }
  node {
    name: S4PVE2
    nodeid: 2
    quorum_votes: 1
    ring0_addr: 192.168.0.252
  }
}

quorum {
  provider: corosync_votequorum
  two_node: 1
}

totem {
  cluster_name: S4PVE
  config_version: 2
  interface {
    linknumber: 0
  }
  ip_version: ipv4-6
  link_mode: passive
  secauth: on
  version: 2
}

 

[Kingneutron]

Thanks for that

 

[jlficken]

Thanks for that

No problem.

I like figuring things out for myself but I was stumped when I posted this.

I figure I might as well post the solution as well to hopefully help someone else in the future.