-
Notifications
You must be signed in to change notification settings - Fork 23
Description
We want to leverage the Automatic OS Upgrade feature within VMSS, however, this seems to have to come at the expense of the HA nature of VMSS.
We currently have instances running within a scaling set with application updates triggered via a custom script extension and Azure DevOps pipelines. The upgrade mode of the model is set to rolling updates meaning we only ever have 1 instance down at a time whilst the upgrade takes place.
On trying to enable automatic OS upgrades, we experience a failure as the upgrade mode on the scaling set needs to be automatic. Making this change causes all instances to upgrade at once, or in a random fashion, hence the application goes offline.
As the orchestrator for automatic OS upgrades carries this process out in the rolling nature, as detailed here https://docs.microsoft.com/en-us/azure/virtual-machine-scale-sets/virtual-machine-scale-sets-automatic-upgrade , it seems odd that the upgrade mode being set to 'rolling' would prevent the use of this feature?
Any pointers would be very useful.
Thanks