I have good news for you, podman 2.0.4 got out last Friday, and I've uploaded it to unstable earlier today. Let me know how this version works with your CI.
Best, -rt On Fri, Jul 31, 2020 at 7:01 AM Martin Pitt <[email protected]> wrote: > Hallo Reinhard, > > Reinhard Tartler [2020-07-31 6:42 -0400]: > > Thank you for your bugreport. I looked into backporting the change and > > checked in with my contact upstream. It turns out that this is not the > only > > issue with the REST interface, and neither of them are easy to backport > to > > 2.0.3. A new release will come out soon that should be suitable for > > cockpit-podman. > > Right, I prodded the podman team about a new release this morning. So far > 2.0.2 > was the only upstream release where the REST interface works. > > > Cockpit looks like a really cool project. I do need to find some time to > > play with it, in particular with the upcoming cockpit-podman component. > > Thanks for packaging it! Are you working full time on "cockpit"? > > Yes, I am :) I'm currently working on getting the integration tests to > succeed > in upstream CI, and once they do and Debian has a podman.deb that actually > works, I'll package cockpit-podman for Debian. > > > Given that the REST interface is not available in testing (the respective > > systemd units were only introduced in 2.0.3+dfsg1-1) > > Right, I currently hack around that in our CI: > > > https://github.com/cockpit-project/cockpit-podman/blob/master/test/vm.install#L13 > > >and podman provides a lot of value even without them, I'd rather have the > >current package migrate to testing at this point. Having said this, I do > >really appreciate your bug report and will see to a remedy as soon as it > >becomes available in the form of a new upstream release. > > Understandable. > > > For the time being, I'd recommend disabling the cockpit-podman test for > the > > time being and re-enable it when this bug gets closed properly. > > I'll probably grab 2.0.2 from > http://snapshot.debian.org/package/libpod/2.0.2%2Bdfsg1-3/ > in the meantime then, to unblock the upstream CI work. That's not a big > deal. > > Thanks, > > Martin > -- regards, Reinhard
