Before using ansible or some other solution to automate specific kernel updates, I just wanted to see if I am missing something built-in or related to Azure Site Recovery to help keep Linux VMs running a supported kernel version.
Problem:
Understandably, updates to the ASR Mobility Agent installed on the VMs lags behind Linux kernel updates. However, replication for VMs break if something like unattended upgrades or someone running apt update/upgrade updates a linux kernel outside of the supported range. For example, the current kernel version for Ubuntu 22.04 LTS is 5.15.0-76-generic, while the highest supported kernel version by ASR is 5.15.0-70-generic.
On smaller deployments, the solution has been to add the kernel packages to the unattended upgrades blacklist and then upgrade the kernel manually based on the latest kernel provided in the support matrix. This is the first deployment I've worked on that has a substantial number of linux VMs where this isn't feasible.
Question:
Is there a preferred way to keep VMs within the supported kernels? Is there a feature or function of Azure Site Recovery that handles this?