1168641 - Figure out what broke four Windows Try build slaves on May 22nd and one more on May 25th, and fix them

Status: RESOLVED FIXED

(Infrastructure & Operations Graveyard :: CIDuty, task)

Description

[Phil Ringnalda (:philor)]

Certainly possible that it's a coincidence that the first batch to die were

b-2008-ix-0051
b-2008-ix-0052
b-2008-ix-0053
b-2008-ix-0054

and even more possible that it's unrelated that a few days later 0050 died, but, more likely they're racked together and that explains their problem.

Comment 1

[Kim Moir [:kmoir] ET]

It looks like these are being used for testing of using runner on windows yet they are still enabled in production

https://bugzilla.mozilla.org/show_bug.cgi?id=1055794#c74

:grenade, mrrrgn is this correct? Should these be disabled in slavealloc or attached to a different master so the testing can proceed without flagging these machines as down in slave health?

Comment 2

[Morgan Phillips [:mrrrgn]]

Longer term testing should happen here it seems. Runner had been tested on a single machine, but more problems pop up each time it rolls out. I'll set up a try master and disable the next group.

Comment 3

[Kim Moir [:kmoir] ET]

:mrrgn so are you going to continue to use these machines in testing?  If so, please attach them to a dev-master. If you don't want to continue to use these machines, could you get them back into a state where they can be used in production.  Currently, they don't start buildbot when rebooted :-)

b-2008-ix-0051
b-2008-ix-0052
b-2008-ix-0053
b-2008-ix-0054

Comment 4

[Morgan Phillips [:mrrrgn]]

(In reply to Kim Moir [:kmoir] from comment #3)
> :mrrgn so are you going to continue to use these machines in testing?  If
> so, please attach them to a dev-master. If you don't want to continue to use
> these machines, could you get them back into a state where they can be used
> in production.  Currently, they don't start buildbot when rebooted :-)
> 
> b-2008-ix-0051
> b-2008-ix-0052
> b-2008-ix-0053
> b-2008-ix-0054

I'll stick them on my master.

Comment 5

[Kim Moir [:kmoir] ET]

thanks!

updated slavealloc to indicate they are loaned to mrrgn see bug 1169789 for details