Question about expectations when running play kube and using volumes
Hello, Yesterday I opened https://github.com/containers/podman/issues/16348 - the heart of the issue is that when starting a pod with volumes, when the pod comes down, it can't come up again without removing the volumes. I'm open to the idea that I can improve how I'm doing the deployment; but it seems that having to remove volumes before restarting the pod does not seem like something we want to require. If we were to have another stateful service running under podman (postgres? strimzi?), having the pod fail to restart after a reboot seems like it might be a big issue. What do you think? -- Martin Jackson Sr. Principal Software Engineer, Ecosystem Engineering Red Hat <https://www.redhat.com/> mhjacks@redhat.com @RedHat <https://twitter.com/redhat> Red Hat <https://www.linkedin.com/company/red-hat> Red Hat <https://www.facebook.com/RedHatInc> <https://www.redhat.com/> <https://redhat.com/summit>
I'm not an expert on the play-kube stuff, but it sounds like a bug based on your description. The description in the issue is quite complex. It might be helpful to see if you can recreate the problem w/o play-kube. Just using basic/simple podman commands. If it's reproducible, maybe add a `--log-level=debug` into the final `podman pod restart` so we might see what's causing the block. HTH Chris Evich (he/him), RHCA III Senior Quality Assurance Engineer If it ain't broke, your hammer isn't wide 'nough. On 10/31/22 11:57, Martin Jackson wrote:
Hello,
Yesterday I opened https://github.com/containers/podman/issues/16348 <https://github.com/containers/podman/issues/16348> - the heart of the issue is that when starting a pod with volumes, when the pod comes down, it can't come up again without removing the volumes. I'm open to the idea that I can improve how I'm doing the deployment; but it seems that having to remove volumes before restarting the pod does not seem like something we want to require. If we were to have another stateful service running under podman (postgres? strimzi?), having the pod fail to restart after a reboot seems like it might be a big issue.
What do you think?
--
Martin Jackson
Sr. Principal Software Engineer, Ecosystem Engineering
Red Hat<https://www.redhat.com/>
mhjacks@redhat.com <mailto:mhjacks@redhat.com>
@RedHat <https://twitter.com/redhat> Red Hat <https://www.linkedin.com/company/red-hat> Red Hat <https://www.facebook.com/RedHatInc> <https://www.redhat.com/> <https://redhat.com/summit>
_______________________________________________ Podman mailing list -- podman@lists.podman.io To unsubscribe send an email to podman-leave@lists.podman.io
On 10/31/22 14:24, Chris Evich wrote:
I'm not an expert on the play-kube stuff, but it sounds like a bug based on your description. The description in the issue is quite complex. It might be helpful to see if you can recreate the problem w/o play-kube. Just using basic/simple podman commands. If it's reproducible, maybe add a `--log-level=debug` into the final `podman pod restart` so we might see what's causing the block.
Could you check this against main. I recall hearing this, and I thought it was fixed.
HTH
Chris Evich (he/him), RHCA III Senior Quality Assurance Engineer If it ain't broke, your hammer isn't wide 'nough.
On 10/31/22 11:57, Martin Jackson wrote:
Hello,
Yesterday I opened https://github.com/containers/podman/issues/16348 <https://github.com/containers/podman/issues/16348> - the heart of the issue is that when starting a pod with volumes, when the pod comes down, it can't come up again without removing the volumes. I'm open to the idea that I can improve how I'm doing the deployment; but it seems that having to remove volumes before restarting the pod does not seem like something we want to require. If we were to have another stateful service running under podman (postgres? strimzi?), having the pod fail to restart after a reboot seems like it might be a big issue.
What do you think?
--
Martin Jackson
Sr. Principal Software Engineer, Ecosystem Engineering
Red Hat<https://www.redhat.com/>
mhjacks@redhat.com <mailto:mhjacks@redhat.com>
@RedHat <https://twitter.com/redhat> Red Hat <https://www.linkedin.com/company/red-hat> Red Hat <https://www.facebook.com/RedHatInc> <https://www.redhat.com/> <https://redhat.com/summit>
_______________________________________________ Podman mailing list -- podman@lists.podman.io To unsubscribe send an email to podman-leave@lists.podman.io
Podman mailing list -- podman@lists.podman.io To unsubscribe send an email to podman-leave@lists.podman.io
Another thing I just noticed in the issue...it's against F37. Where we had a major systemd related SNAFU with podman. Might be good to confirm that's not in play (and/or try updating systemd to 251.7-611 or later). This fixed a LOT of issues in F37 podman CI recently (yesterday). Chris Evich (he/him), RHCA III Senior Quality Assurance Engineer If it ain't broke, your hammer isn't wide 'nough. On 11/1/22 08:33, Daniel Walsh wrote:
On 10/31/22 14:24, Chris Evich wrote:
I'm not an expert on the play-kube stuff, but it sounds like a bug based on your description. The description in the issue is quite complex. It might be helpful to see if you can recreate the problem w/o play-kube. Just using basic/simple podman commands. If it's reproducible, maybe add a `--log-level=debug` into the final `podman pod restart` so we might see what's causing the block.
Could you check this against main. I recall hearing this, and I thought it was fixed.
HTH
Chris Evich (he/him), RHCA III Senior Quality Assurance Engineer If it ain't broke, your hammer isn't wide 'nough.
On 10/31/22 11:57, Martin Jackson wrote:
Hello,
Yesterday I opened https://github.com/containers/podman/issues/16348 <https://github.com/containers/podman/issues/16348> - the heart of the issue is that when starting a pod with volumes, when the pod comes down, it can't come up again without removing the volumes. I'm open to the idea that I can improve how I'm doing the deployment; but it seems that having to remove volumes before restarting the pod does not seem like something we want to require. If we were to have another stateful service running under podman (postgres? strimzi?), having the pod fail to restart after a reboot seems like it might be a big issue.
What do you think?
--
Martin Jackson
Sr. Principal Software Engineer, Ecosystem Engineering
Red Hat<https://www.redhat.com/>
mhjacks@redhat.com <mailto:mhjacks@redhat.com>
@RedHat <https://twitter.com/redhat> Red Hat <https://www.linkedin.com/company/red-hat> Red Hat <https://www.facebook.com/RedHatInc> <https://www.redhat.com/> <https://redhat.com/summit>
_______________________________________________ Podman mailing list -- podman@lists.podman.io To unsubscribe send an email to podman-leave@lists.podman.io
Podman mailing list -- podman@lists.podman.io To unsubscribe send an email to podman-leave@lists.podman.io
_______________________________________________ Podman mailing list -- podman@lists.podman.io To unsubscribe send an email to podman-leave@lists.podman.io
participants (3)
-
Chris Evich -
Daniel Walsh -
Martin Jackson