With desktop builds on mock machines, we can move more of the ix boxes to windows builders. This bug will pass through many hands before done: - buildduty to identify & take out of production 10 ix slaves currently assigned to pool build-scl1 - relops to coordinate changes with dcops - hwine to put new builders into production
(In reply to Hal Wine [:hwine] from comment #0) > - buildduty to identify & take out of production 10 ix slaves currently > assigned to pool build-scl1 The following slaves were disabled and marked as "decommissioned" linux-ix-slave32 linux-ix-slave33 linux-ix-slave34 linux-ix-slave35 linux-ix-slave36 linux-ix-slave37 linux64-ix-slave18 linux64-ix-slave19 linux64-ix-slave20 linux64-ix-slave21
Moving to relops
Filed bug 798046 to disable nagios checks
Looks like a (In reply to Ed Morley [:edmorley UTC+1] from comment #4) > https://hg.mozilla.org/mozilla-central/rev/6e34c2118a92 Looks unrelated to this bug
Dustin: please change the switch ports for and populate DHCP for the following machines (and their management NICs) that are moving into the winbuild network in scl1: linux-ix-slave32 w64-ix-slave111 linux-ix-slave33 w64-ix-slave112 linux-ix-slave34 w64-ix-slave113 linux-ix-slave35 w64-ix-slave114 linux-ix-slave36 w64-ix-slave115 linux-ix-slave37 w64-ix-slave116 linux64-ix-slave18 w64-ix-slave117 linux64-ix-slave19 w64-ix-slave118 linux64-ix-slave20 w64-ix-slave119 linux64-ix-slave21 w64-ix-slave120
dns, inventory, and nagios updated. bug filed with dcops for physical labeling.
I have added these to DHCP on the Windows side
ports are changed. DHCP is updated. I dropped link on all the ports, and management ports are starting to come up.
Management ports are all up.
Matt's totally going to kick this out on Sunday.
These machines have been reimaged.
(In reply to Rail Aliiev [:rail] from comment #5) > Looks like a (In reply to Ed Morley [:edmorley UTC+1] from comment #4) > > https://hg.mozilla.org/mozilla-central/rev/6e34c2118a92 > > Looks unrelated to this bug I've just seen this by chance; this was for bug 795456. Bugs are marked using a script when merged from inbound to mozilla-central. If people have used the wrong bug number, then unfortunately the wrong bug will get closed. Please can you always ping me via IRC/email/CC me to the bug, if you see a miss-closure like this, since I have no other way of knowing (the script intentionally doesn't CC me since I'd then be CCed on hundreds of extra bugs per week). Thank you :-)
arr and I discussed that these machines will need to be re-imaged since passwords and _dumbwin32proc.py got deployed since October (maybe more things). I will set them up in bug 860890.
These are also finished reimaging again.