Skip to content
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

Re-run initContainers in a Deployment when containers exit on error #871

majgis opened this Issue Mar 4, 2019 · 1 comment


None yet
2 participants
Copy link

majgis commented Mar 4, 2019

Enhancement Description

  • One-line enhancement description (can be used as a release note):
    Re-run initContainers in a Deployment when containers exit on error
  • Kubernetes Enhancement Proposal: (link to kubernetes/enhancements file, if none yet, link to PR)
  • Primary contact (assignee):
  • Responsible SIGs:
    /sig apps
    /sig node
  • Enhancement target (which target equals to which milestone):
    • Alpha release target (x.y)
    • Beta release target (x.y)
    • Stable release target (x.y)

Please to keep this description up to date. This will help the Enhancement Team track efficiently the evolution of the enhancement


This comment has been minimized.

Copy link

majgis commented Mar 4, 2019

A PR still needs to be created for this proposal. If someone else is passionate about this issue, I'm happy to hand off, but I'll continue to pursue it and am currently working on a PR for the enhancements file.

This proposal follows up on an issue that was closed because it was expected to be solved under a different KEP.
The original issue was kubernetes/kubernetes#52345.
It was anticipated the solution would be included in .
Why it wasn't included is discussed here: #691

I see two potential solutions to this problem.
a. Overload existing configuration to allow restarting pod when a container exits, ensuring that initContainers are rerun prior to containers restarting
b. Add new configuration/functionality to associate unique initContainers to specific containers.

Drawbacks and concerns in regard to option a. were discussed in the issue linked to above.
Option b is my own addition, which seems logical and straightforward, but no doubt much more involved to implement.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
You can’t perform that action at this time.