For every sub release 2024.12.x I am reading trough all the listed modules/integrations to determine if there is anything relevant for my system in this specific version. At the same time my HA system knows what it is using.
Proposal: in HA list all integrations (or a number of) that have an update in this subversion. If no-relevant upgrades are existing there should be no reason for me to spend time (and downtime of HA) on upgrading.
That is the opposite of the way I look at it. If there are no updates that affect my system I feel free to update without worry, Downtime is 40 seconds. So not a factor for me.
It’s in the release notes, so it is known somewhere.
The WTH is how to determine which updates are not affecting my system and filter out these, so that I only see what is important for my system.
It is not just the 40 seconds, it is looking at the release notes, read all the tens of update announcements, skipping lines with name that I never heard of, try to see if there is anything important giving me any reason to update, or see if there is something that needs more attention as it might cause or solve issues on my system.
I love HA, but I don’t have time to keep track of everything and solve everything. My HA system is working quite good, no major issues (most of the times). But some integrations, mostly cloud based ones (that I try to avoid as much as possible), do change and break some functionality. And if there is an update on one of those I will install, otherwise I wait for the next release.
Some background. In my system I have automations that depend on electricity prices, and with hourly changing prices, a broken automation can be costly. Running the warm water boiler, or charging my EV at the wrong time in winter when there is no wind and almost no sun and thus very high electricity prices (like today), can cost literally a multiple euros/dollars per hour. So for me it is important that I have Tibber, Nordpool, Shelly and Tesla (and a few others) running reliable and with limited downtime.