Maybe my problem is in part linked to an issue seen by Jayme earlier, but then the resolution that worked for him did not succeed for me :

I first upgraded my Self Hosted Engine from 4.4.5 to 4.4.6 and then upgraded it to Centos-Stream and rebooted

Then I tried to upgrade the cluster (3 ovirt-nodes on 4.4.5) but it failed at the first host.

They are all ovir-node hosts, originally first installed in 4.4.5

In Host Event Logs I saw :

...
Update of host ps-inf-prd-kvm-fr-510.hostics.fr.
Upgrade packages
Update of host ps-inf-prd-kvm-fr-510.hostics.fr.
Check if image was updated.
Update of host ps-inf-prd-kvm-fr-510.hostics.fr.
Check if image was updated.
Update of host ps-inf-prd-kvm-fr-510.hostics.fr.
Check if image-updated file exists.
Failed to upgrade Host ps-inf-prd-kvm-fr-510.hostics.fr (User: gpav@hostics.fr).

ovirt-node-ng-image-update-4.4.6.3-1.el8.noarch was installed according to yum,

I tried reinstalling it but got errors: "Error in POSTIN scriptlet" :

Downloading Packages:
[SKIPPED] ovirt-node-ng-image-update-4.4.6.3-1.el8.noarch.rpm: Already downloaded
...
Running scriptlet: ovirt-node-ng-image-update-4.4.6.3-1.el8.noarch
Reinstalling : ovirt-node-ng-image-update-4.4.6.3-1.el8.noarch
Running scriptlet: ovirt-node-ng-image-update-4.4.6.3-1.el8.noarch
warning: %post(ovirt-node-ng-image-update-4.4.6.3-1.el8.noarch) scriptlet failed, exit status 1

Error in POSTIN scriptlet in rpm package ovirt-node-ng-image-update
---
Reinstalled:
ovirt-node-ng-image-update-4.4.6.3-1.el8.noarch

nodectl still showed it was on 4.4.5 :

[root@ps-inf-prd-kvm-fr-510 ~]# nodectl info
bootloader:
default: ovirt-node-ng-4.4.5.1-0.20210323.0 (4.18.0-240.15.1.el8_3.x86_64)
...
current_layer: ovirt-node-ng-4.4.5.1-0.20210323.0+1

I tried to upgrade the Host again from oVirt and this time there was no error, and the host rebooted.

However, it did not pass active after rebooting and nodectl still shows that it's 4.4.5 installed. Similar symptoms as OP

So I removed ovirt-node-ng-image-update, then reinstalled it and got no error this time.

nodectl info seemed to show that it was installed :

[root@ps-inf-prd-kvm-fr-510 yum.repos.d]# nodectl info
bootloader:
default: ovirt-node-ng-4.4.6.3-0.20210518.0 (4.18.0-301.1.el8.x86_64)
...
current_layer: ovirt-node-ng-4.4.5.1-0.20210323.0+1

However, after reboot the Host was still shown as "unresponsive"

After Marking it as "Manually rebooted", passing it in maintenance mode and trying to activate it, the Host was automatically fenced. And still unresponsive after this new reboot.

I passed it in maintenance mode again, And tried to reinstall it with "Deploy Hosted Engine" selected

However if failed : "Task Stop services failed to execute."

In /var/log/ovirt-engine/host-deploy/ovirt-host-deploy-ansible-20210602082519-ps-inf-prd-kvm-fr-510.hostics.fr-0565d681-9406-4fa7-a444-7ee34804579c.log :

"msg" : "Unable to stop service vdsmd.service: Job for vdsmd.service canceled.\n", "failed" : true,

"msg" : "Unable to stop service supervdsmd.service: Job for supervdsmd.service canceled.\n", failed" : true,

"stderr" : "Error: ServiceOperationError: _systemctlStop failed\nb'Job for vdsmd.service canceled.\\n' ",

"stderr_lines" : [ "Error: ServiceOperationError: _systemctlStop failed", "b'Job for vdsmd.service canceled.\\n' " ],

If I try on the Host I get :

[root@ps-inf-prd-kvm-fr-510 ~]# systemctl stop vdsmd
Job for vdsmd.service canceled.

[root@ps-inf-prd-kvm-fr-510 ~]# systemctl status vdsmd
● vdsmd.service - Virtual Desktop Server Manager
Loaded: loaded (/usr/lib/systemd/system/vdsmd.service; enabled; vendor preset: disabled)
Active: deactivating (stop-sigterm) since Wed 2021-06-02 08:49:21 CEST; 7s ago
Process: 54037 ExecStartPre=/usr/libexec/vdsm/vdsmd_init_common.sh --pre-start (code=exited, status=0/SUCCESS)
...

Jun 02 08:47:34 ps-inf-prd-kvm-fr-510.hostics.fr vdsm[54100]: WARN Failed to retrieve Hosted Engine HA info, is Hosted Engine setup finished?
...
Jun 02 08:48:31 ps-inf-prd-kvm-fr-510.hostics.fr vdsm[54100]: WARN Worker blocked: <Worker name=jsonrpc/4 running <Task <JsonRpcTask {'jsonrpc': '2.0', 'method': 'StoragePool.connectStorageServer', 'params': {'storage>
File: "/usr/lib64/python3.6/threading.py", line 884, in _bootstrap
self._bootstrap_inner()

Retrying to manually stop vdsmd a second time then seems to work...

I tried rebooting again, restarting the install always fail at the the same spot

What should I try to get this host back up?

Guillaume Pavese
Ingénieur Système et Réseau

Interactiv-Group

Ce message et toutes les pièces jointes (ci-après le “message”) sont établis à l’intention exclusive de ses destinataires et sont confidentiels. Si vous recevez ce message par erreur, merci de le détruire et d’en avertir immédiatement l’expéditeur. Toute utilisation de ce message non conforme a sa destination, toute diffusion ou toute publication, totale ou partielle, est interdite, sauf autorisation expresse. L’internet ne permettant pas d’assurer l’intégrité de ce message . Interactiv-group (et ses filiales) décline(nt) toute responsabilité au titre de ce message, dans l’hypothèse ou il aurait été modifié. IT, ES, UK.