I can't reboot it.

Host is powered on, dep bug seems to indicate that its back up properly. but its still pinned to your dev master, do the jobs look good?

It's actually not ready. I checked on it and it was ping down and using the ipmitool did not cycle it.

I've rebooted it into production. This machine has never taken a job so let's see how it turns out: https://secure.pub.build.mozilla.org/buildapi/recent/t-w732-ix-100

Incredibly poorly - like all the other win7-ix slaves that have been disabled all summer due to graphics/resolution troubles, it's unable to run webgl tests. Disabled in slavealloc.

Rebooted into production.

8 attempts to run xperf since it came back, 8 failures. Disabled in slavealloc.

Joel: I'm unclear about the [xperf_q3_disabled] whiteboard tag: has xperf been turned off, i.e. can this slave be re-enabled?

FTR I've started working on re-imaging another machine like this to put it on staging. I'm doing this with 016 in bug 872915. I don't yet know the way to victory.

I am not sure what others use the xperf_q3_disabled flag for, I used it for all machines (w7) which have have history of failing xperf and we shouldn't be running on unless it is fixed and verified in testing. I would be happy to help where I can. IIRC I had a loaner (002?) which didn't run successfully in automation, but ran just fine when I had it as a loaner. This difference between running in automation and on a loaner machine needs to get resolved, so people can help solve these issues.

(In reply to Joel Maher (:jmaher) from comment #9) > I am not sure what others use the xperf_q3_disabled flag for, I used it for > all machines (w7) which have have history of failing xperf and we shouldn't > be running on unless it is fixed and verified in testing. > > I would be happy to help where I can. IIRC I had a loaner (002?) which > didn't run successfully in automation, but ran just fine when I had it as a > loaner. This difference between running in automation and on a loaner > machine needs to get resolved, so people can help solve these issues. Which bug is tracking xperf failures/fixing?

I don't know.

I don't have a specific bug for it- do we need one?

(In reply to Joel Maher (:jmaher) from comment #12) > I don't have a specific bug for it- do we need one? Well, it would be good to track it somewhere. Then we could block bugs like this on it at least. Right now it's hard to be sure where this slave stands without deep reading. Or, if we're not going to do anything about it we can just junk these slaves...

Oh sorry... I had been assigning them to myself. I will be releasing them as I fix them.

I'm putting it back into production as it has not failed any jobs on staging (including xperf). FTR I didn't even re-format the machine, I'm surprised it can take xperf jobs.

I've seen it fail a try debug test mochitest-2 job but not an xperf job. 09:30:48 INFO - 429 ERROR TEST-UNEXPECTED-FAIL | /tests/docshell/test/test_bug668513.html | /tests/docshell/test/test_bug384014.html finished in a non-clean fashion, probably because it didn't call SimpleTest.finish()

It's looking good.

Needs a reimage according to bug 1104225, disabled in slavealloc.

Re-imaged and returned to production.

Re-imaged and enabled in slavealloc

Attempting SSH reboot...Failed. Attempting IPMI reboot...Failed. Filed IT bug for reboot (bug 1364344)