Please keep and show logs for every re-run, whether it passes or fails.
At present if I have a failed run, which might have valuable stuff in the logs, which I might want to refer to in an issue for an upstream fix or whatever, that gets lost when the job is re-run.
- example atsign-foundation/at_server_trunk #25 initially failed due to a missing dependency, but re-run was successful after implementing a local fix, but now there’s no way to get the logs showing the original issue.
Similarly if a successful run is re-run and then fails, there’s no evidence that there was ever a successful run.
- example atsign-company/labels #95 (can’t link due to new user limits) succeeded on its initial run but then failed on re-run.