This message was deleted.
# rancher-desktop
a
This message was deleted.
f
There is no timeline for it right now. It has been a low priority because it "just works" and does not have any performance impacts; it is more a tidiness issue.
q
it does cost installing rosetta which might not otherwise be installed (in fact, i'm guessing we wouldn't need rosetta on these computers at all if not for this). but, i get it.
f
Dropping support for older k8s versions is probably the most sensible thing; unfortunately we don't know what people are using.
q
optional analytics to collect that info would be kinda nice </pipe-dream>
f
Rosetta isn't that big either, and you may want to use it with VZ to run amd64 container images, but that too is a rare use case
We have a checkbox to allow us to collect analytics, but there is no actual logic to do so ๐Ÿ˜„
q
yeah, i remember the checkbox
not particularly surprised to hear that it hasn't gained any additional functionality ๐Ÿ™‚
(that's fine)
f
The k3s team already thinks we are crazy to allow people to select ancient versions on Rancher Desktop.
q
offhand, i think they're right
i'd be inclined to try removing 2 ancient versions per release and see how much pushback you get ๐Ÿ™‚
(possibly have a hidden pref for "show-obsolete-k8s-versions")
f
I would probably prefer to say anything before 1.21 or 1.22 would be removed. Anything that has special logic in the codebase ๐Ÿ˜„
q
I'd take that
f
I don't want to keep the dead code; if we remove it, there won't be a hidden options to re-enable it
q
happy to go for the aggressive thing
f
The not-so-hidden option would be to go back to 1.14 or whatever the latest would be that still supports it.
โœ”๏ธ 1