Reuse Builder output for Post-processor debugging

Howdy. Is there a convienent way to only re-run post-processors, not the builders? I have artifacts saved from the builders (with “keep_input_artifact”), but I’m debugging my post-processor code, and rather not re-run the builders just to troubleshoot the post-processor code. Thanks! :slight_smile:

I don’t know if this will work, but you could combine a source null

with the only parameter.

So after your first run you only use a null builder which should do nothing.

Sounds logical in my head. But if it is designed for that - and in this handling - I have no idea. :smiley:

1 Like

yeah you’d want to combine a “null” builder with a postprocessor “sequence” definiton that starts with the “artifice” post-processor to override the “none” artifact from the “null” builder with the artifact files produced by your real build. Here’s a guide that gives an example in code: https://www.packer.io/guides/packer-on-cicd/pipelineing-builds#more-efficiencies

2 Likes

@SwampDragons That worked great! Thank you!