To help me debug the gaia integration tests which run downstream from b2g-desktop builds.

Hi Gareth, We have two sets of machines running Ubuntu tests, so I want to make sure I set you up with the correct ones. talos-linux[32|64] - physical hardware, suitable for graphics and performance testing (talos) tst-linux[32|64]-ec2 - AWS instance, suitable for unit/reftests If this is an existing test (or set of tests) you're working on, TBPL will show you the slave used in the lower-left pane when you click on the job: "using slave: xxx"

Gareth is interested in EC2 ubuntu slaves.

I'll try to get these set-up before I head out today.

E-mail sent on Friday with access information. Gareth, please unassign this bug from yourself when you're done with these machines.

Gareth, are you still using these machines? tst-linux32-ec2-gaye tst-linux64-ec2-gaye

Gareth is on PTO this week but to my knowledge he is still interested in using them when he returns.

Fine if we stop (not kill) them and start them next week?

Yes, that should be fine though I think he is returning tomorrow so it might be a short-lived stoppage. But he can ping you when he's ready to re-engage here.

Ben and Armen - Thanks for your help! I'm trying to get access now but first have to find a vpn client.

Best recommendation I have is Viscosity http://www.sparklabs.com/viscosity/download/ -- free to try, and you can request a license from the helpdesk on ServiceNow. the mana page we linked in the e-mail has setup instructions

Are you still using these, Gareth?

I'm not using them continuously, but when issues come up it's nice to have (since there's no gaia try).

OK. I'm going to shut them down for now - you can ping in #releng the next time you need them.

Shut down again (last login was on Jan 17),ping in #releng when they are needed

FYI, this instance has no public IP, so if you try to use any of the newest scripts (e.g that connect to ftp-ssl, or hg-ssl) you'll likely fail, because they try to route over the public internet but cannot. We are willing to recreate your instance from scratch if you wish, but will only do so at your request. Please reassign to "nobody@" and ask us to, if you wish. (if said timing is fairly important you can also ping in #releng) ==== Also it looks like it has been almost a month since you last needed these, is it ok to terminate them and recreate them again when you need them again? (doing so will also make sure you have the latest machine state as is in our production). Or do you need them now?

Don't need now! Feel free to terminate!

Done