Configuration for moving failed jobs into archive table

[klesgidis]

Hi! Thanks a lot for all of your work done on this project.

The archiveCompletedAfterSeconds archives not only completed jobs, but also failed.

Is there a way to move failed jobs into archive table with a different configuration option?

My use case is that we need more time to handle failed jobs than the completed. (e.g. deploy a fix in production and then retry the job)

[timgit]

Yes, that sounds like a good use case for a new config option

[klesgidis]

Ok, I'll open a PR soon

[klesgidis]

Any suggestions on where this should be implemented? Or should I take it on my own?

[timgit]

archiveFailedAfterSeconds seems like a good choice. If it's not set, it can default to archiveCompletedAfterSeconds

[klesgidis]

#364