One service is conflicting, but not running

[coreos]

One service is conflicting, but not running

[AndrewVos]

Also useful information, if I stop portiere.3.service then portiere.2.service will launch. From then on portiere.3.service will not launch with any amount of destroy/submit/start. Pretty sure this means the job offer is just not able to be accepted.

[AndrewVos]

curl 'http://127.0.0.1:4001/v2/keys/_coreos.com/fleet/?recursive=true'

https://gist.github.com/AndrewVos/42b984b11c63b524d4c9

[bcwaldon]

I bet one of your fleet agents thinks it has an outstanding bid for portiere.1.service and will not submit conflicting bids. Restarting fleet on the EC2 instance that refuses to run anything should fix it. I've added a feature to fleet that allows you to send that SIGUSR1 signal to a fleet daemon and cause it to dump the current AgentState to the logs. That will help us debug this kind of stuff in the future. Is there anything else we can do wit this specific issue.