portainer: Portainer 2.17.0 Business Edition no longer able to recreate containers - Error 500

Previously version 2.16 I could click on a container in Portainer, click on recreate, and then select to pull a new image - everything worked just fine.

Recently when I try this, I’ll get a spinning circle in the top right of the screen, and after a minute or two I’ll get an error message saying;

Failure
Request failed with status code 500

The container will then be restarted with the original non-updated image, although the image will show up as having been renamed from its original name to a hex string such as fdf04daf6952.

About this issue

  • Original URL
  • State: open
  • Created 4 years ago
  • Reactions: 4
  • Comments: 63 (2 by maintainers)

Most upvoted comments

Same here. Please fix this 😦

Not working for me again after updating to 2.19.0

I have just about give up on this as a solution, that a Business Edition application can have such huge problems for so many weeks without it being fixed or rolled back is ridiculous.

Update:

Upon further testing, this only occurs on containers that were created via docker run and not Portainer.

I have forwarded this information to the team for review.

Thanks!

I was the original reporter on Reddit. This happened on every container I’ve tried to update so far.

@Andrew-H2O Yes, I copied your post and pasted it on GitHub. Sorry, I wanted to report a case quickly (I didn’t want to describe the problem again on GitHub). You’re not angry, I owe you a beer… 😉

I was the original reporter on Reddit. This happened on every container I’ve tried to update so far.

I gave up on this and used https://github.com/mag37/dockcheck it was more reliable.

Since my last post back in 09/2023… (we are now 1.5 years since my first post 02/2023. that was on version 2.16)

Not sure when the issue went back to “across the board” from “only two remaining containers”. I’ve updated versions multiple times and the issue remains. My current situation…

ALL containers (local & via agent) whether the image is already downloaded (doesn’t matter if the image is the same or newer) require updating from “stacks editor” as opposed to “containers recreate”.

The “two remaining containers” from my last post (calibre & home assistant) require that I first manually download the images otherwise no matter how I attempt (via stacks editor or container recreate) the update fails.

Having the same issue since update to 2.19.4 BE Happening to all my containers regardless of network config. And also happening regardless of how the container was created. Getting a 500 error on recreate and repull, 2nd attempt works but wipes env

It wasn’t a stack, and the volume is only part of the equation. If the image depended on environment variables to properly function, then those are lost. I just lost an entire BookStack install due to this; and not realizing those env values needed to be set, it wiped out the volume with a fresh install.

We’ve just released 2.17.1 which should fix this issue - are you able to update and retest?