Consulter notre FAQ
{"searchBar":{"inputPlaceholder":"Rechercher par mot-clé ou poser une question","searchBtn":"Rechercher","error":"Veuillez saisir un mot-clé pour effectuer une recherche"}}
PowerChute Network Shutdown – Hosts fail to enter maintenance mode when vSphere Clustering Service (vCLS) is enabled
Issue:
PowerChute Network Shutdown – vSphere Clustering Service (vCLS) introduced in VMware vSphere 7.0 update 1 prevents hosts from entering maintenance mode since vCLS agent VMs cannot be shut down
Products:
PowerChute Network Shutdown v4.x
Environment:
VMware vSphere 7.0 update 1 and above
vCenter Server Appliance 7.0 update 1 and above
Cause:
vSphere 7.0.1 introduced a new feature called vSphere Cluster Services (vCLS) that allows DRS and HA to continue functioning even when vCenter Server is unavailable.
The issue occurs because there are 3 vCLS Agent VMs deployed in the cluster that cannot be powered off during the VM Shutdown step along with other User VMs. When vCenter Server detects that these VMs are powered off, it will automatically power them back on. Since these VMs remain powered on, the Hosts on which they reside cannot be placed into maintenance mode.
Solution:
Enable VM Prioritization and add all User VMs to a priority group except for the vCLS VMs. vCLS VMs are named vCLS (<number>).
Configure a 0 second duration for VM Shutdown and VM Startup for the Un-prioritized group. This ensures that the vCLS VMs will not be shut down or powered on by PowerChute.
Ensure that Delay Maintenance Mode is enabled.
Please follow the instructions in the attached PDF (vCLS_ScriptInstructions). Alternatively, if you are not using DRS, you can choose to disable vCLS permanently by following the instructions here - https://kb.vmware.com/s/article/80472
NOTE: If the vCLS VMs are added to a higher priority group than the vCenter Server Appliance VM, the scripts are not required to place the cluster into retreat mode. This is only applicable if vCenter is inside the cluster.
PowerChute Network Shutdown – vSphere Clustering Service (vCLS) introduced in VMware vSphere 7.0 update 1 prevents hosts from entering maintenance mode since vCLS agent VMs cannot be shut down
Products:
PowerChute Network Shutdown v4.x
Environment:
VMware vSphere 7.0 update 1 and above
vCenter Server Appliance 7.0 update 1 and above
Cause:
vSphere 7.0.1 introduced a new feature called vSphere Cluster Services (vCLS) that allows DRS and HA to continue functioning even when vCenter Server is unavailable.
The issue occurs because there are 3 vCLS Agent VMs deployed in the cluster that cannot be powered off during the VM Shutdown step along with other User VMs. When vCenter Server detects that these VMs are powered off, it will automatically power them back on. Since these VMs remain powered on, the Hosts on which they reside cannot be placed into maintenance mode.
Solution:
Enable VM Prioritization and add all User VMs to a priority group except for the vCLS VMs. vCLS VMs are named vCLS (<number>).
Configure a 0 second duration for VM Shutdown and VM Startup for the Un-prioritized group. This ensures that the vCLS VMs will not be shut down or powered on by PowerChute.
Ensure that Delay Maintenance Mode is enabled.
Please follow the instructions in the attached PDF (vCLS_ScriptInstructions). Alternatively, if you are not using DRS, you can choose to disable vCLS permanently by following the instructions here - https://kb.vmware.com/s/article/80472
NOTE: If the vCLS VMs are added to a higher priority group than the vCenter Server Appliance VM, the scripts are not required to place the cluster into retreat mode. This is only applicable if vCenter is inside the cluster.
Publié pour:APC France
Pièce(s) jointe(s)
Cela vous a-t-il été utile ?
En savoir plus
Gamme :
En savoir plus
Gamme :