-
Notifications
You must be signed in to change notification settings - Fork 53
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Bump toolchain/luet to 0.32.0 #1268
Conversation
Signed-off-by: cOS-cibot [bot] <cOScibot@gmail.com>
Signed-off-by: cOS-cibot [bot] <cOScibot@gmail.com>
Signed-off-by: cOS-cibot [bot] <cOScibot@gmail.com>
Signed-off-by: cOS-cibot [bot] <cOScibot@gmail.com>
@davidcassany seems now during luet bumps it tries to upgrade from the local repo but fails as we have an hard dep on the toolchain versions in the repo:
On one side it is very handy that it will build everything with the new version, so maybe we can be less restrict on the version being pinpointed here? |
@mudler I am not sure I fully get it, but to me it feels there might be something wrong in our build process, if we build a new luet all packages requiring a new luet should be rebuilt too. Looks like we have little issues in toolchain package definition since those dependencies while they seam to be fine they are fake. I'd consider it a small bug in our package definitions. As an example lets consider the
The runtime toolchain constraint is based under the assumption that we keep luet backward compatible. If we assume that luet is not backward compatible on major release we could write it as:
|
yep, the problem at hand is that https://github.com/rancher-sandbox/cOS-toolkit/blob/master/packages/toolchain/collection.yaml#L34 uses an anchor to luet, so it hardly binds luet-makeiso <-> luet version, similarly for the other toolchain pieces, I think we had an issue around it because it bubbled up also in another occasion: #944 Indeed the dependency versions can be optimized, at the end For the build time, I wouldn't pin any version as the go.mod will pin by itself internally, so for us it's just required to be specific on the |
what is true is also that we don't have to pull anymore luet-makeiso during make deps as well :) |
Yes, just keeping this single constraint to limit the backward compatibility to some reasonable range sounds like a great option and forget about buildtime dependencies of go modules. |
And make it bound to luet <1.0.0 Signed-off-by: Ettore Di Giacinto <edigiacinto@suse.com>
Signed-off-by: Ettore Di Giacinto <edigiacinto@suse.com>
@davidcassany I've pushed two commits here on this branch:
If looks good, I'd go with auto-merge afterwards |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good here
btw this version brings in |
Signed-off-by: cOS-cibot [bot] cOScibot@gmail.com
Answer: Because Oct 31 = Dec 25