Add an error message to controld_sched_timeout() so the fatal error i…

[thomas-jones-ibm]

…sn't a mystery

[knet-jenkins]

Can one of the project admins check and authorise this run please: https://haci.fast.eng.rdu2.dc.redhat.com/job/pacemaker/job/pacemaker-pipeline/job/PR-4116/1/input

[nrwahl2]

retest this please

[nrwahl2]

@thomas-jones-ibm Thanks for the patch!

Can you please rebase your branch on the current main branch, and then push again? The CI test failures are from calls to deprecated libxml2 functions -- but we removed those functions a while back.

[knet-jenkins]

Can one of the project admins check and authorise this run please: https://haci.fast.eng.rdu2.dc.redhat.com/job/pacemaker/job/pacemaker-pipeline/job/PR-4116/3/input

[thomas-jones-ibm]

@nrwahl2 done

[nrwahl2]

retest this please

[nrwahl2]

(previous comment is a directive that gets picked up by our CI system)

[nrwahl2]

Hi, Thomas. We actually deprecated crm_err a while back and replaced it with pcmk__err. I didn't notice that your patch used crm_err at first; that's the cause of the most recent CI failure. Can you update one more time to change it to pcmk__err?

I think we should be good at that point. This is just a one-line logging patch, after all ;)

[knet-jenkins]

Can one of the project admins check and authorise this run please: https://haci.fast.eng.rdu2.dc.redhat.com/job/pacemaker/job/pacemaker-pipeline/job/PR-4116/5/input

[thomas-jones-ibm]

@nrwahl2 Done!

I was originally looking at Pacemaker 2.1.6 while tracking a problem down, and just rebased the change so that's why it was crm_err() This seemed to be the only place in all of pacemaker that could have ended up with controld exiting with status 100 without a prior message that would explain why.