It would be helpful if we exposed more information about what the provisioner is up to. I think the following things are easy to add to the provisioner and would give a good idea. 1. pulse message when submitting spot request 2. pulse message when a spot request is rejected or fulfilled 3. pulse message when an instance is killed 4. pulse message on each provisioner iteration 5. a view in tools.tc.net which shows information on these values With this information, it should be possible to figure out the health of provisioning.
Summary: Provide more insight into what the provisioner is doing → Publish pulse messages from aws-provisioner
We solved this by keeping data in the db for ec2-manager
Assignee: nobody → jhford
Status: NEW → RESOLVED
Last Resolved: a year ago
Resolution: --- → FIXED
Component: AWS-Provisioner → Services
Product: Taskcluster → Taskcluster
You need to log in before you can comment on or make changes to this bug.