Closed madhavajay closed 2 years ago
@madhavajay thanks for creating this issue! I'm not sure if we can implement a retry mechanism there as it seems to be a very unspecific error and retrying does not always solve the problem. With the new pipelines feature you could actually implement this yourself:
functions:
build_images:
while ! __build_images "$@"
do
echo "Building failed, will try again in 5 seconds..."
sleep 5
done
@FabianKramm oh thats quite interesting, im trying to understand that syntax, so you can hook the existing steps and call them with some kind of __
syntax or something?
I am totally fine with a solution like that and love the flexibility, when is a good time to upgrade, it seems like maybe its still too soon?
I solved this on my end using glorious one liner ala stackoverflow: https://stackoverflow.com/questions/33353847/retry-a-sequence-of-commands-in-bash-for-a-limited-number-of-times
(r=3;while ! some_cmd ; do ((--r))||exit;sleep 60;done)
What happened?
Sometimes deployment fails during the image push phase.
What did you expect to happen instead?
It should push images and retry if there is an issue.
How can we reproduce the bug? (as minimally and precisely as possible)
I am using k3ds and a local registry. I dont know why this happens but it seems like you can just try and it works fine.
Local Environment:
Anything else we need to know?
This could be the fault of k3d or docker or my machine resources but it would be nice if there was a way to "retry" or extend the timeout on the devspace side.
/kind bug