Open Morgimiel opened 1 week ago
Looks like you may have been bit by this recent change:
I think the resolution to should be a flag/envvar that kicks Hauler into the "legacy" fail-fast mode. Something like --failfast
with a default of false
with value of $HAULER_FAILFAST
parsed for truthiness as env override.
@Morgimiel can you provide a simple example of a script that you expect to fail but does not?
I agree with your proposal. A flag to optionnally fail on error seems OK to me. My script is pretty simple:
helm template myproject | grep image: | tr -d ' "' | sed -nr 's/^image:(.*)$/\1/p' | sort -u | xargs -I {} hauler store add image {}
I want it to fail if any of the add image command fails.
PS: It would be a great feature to have hauler store add chart
command have an option to find docker images within in and add them to the store :D. But thats purely bonus ;)
Is this RFE related to an Existing Problem? If so, please describe:
I have a script that load images to hauler store. If something went wrong, I have no error exit code to early exit my script.
-
Describe Proposed Solution(s):
We should have an exit code > 0 if something went wrong during a process. For exemple, my hauler store add image command fails because of a bad credentials but I still have exit code 0.
-
Describe Possible Alternatives:
-
Additional Context:
-