← Back to team overview

canonical-ci-engineering team mailing list archive

Re: Offlining a failed device via jenkins

 

On 10/16/2013 09:16 AM, Francis Ginther wrote:
Do we have a touch device health check already defined? If
not, what would be a good starting point?

We[1] had the concept of a health-check in LAVA that ran once a day on every device in our lab. We just took a known good image for a device and did a flash, boot, run simple test. This was really handy for:

 * spotting devices going bad (we had sd-cards and those suck)
 * validating code/infrastructure updates. After upgrading LAVA we
   ran a health-check on every device before allowing them to accept
   jobs.

I think this would be a good idea for us. Not sure if we could make it work as well since Jenkins doesn't seem to have built-in scheduling logic around this.

1: I think Paul invented it.


Follow ups

References