We now have try-linux64-slave0{1,5} that have been cloned from the ref image. Steps followed to get them into the right state: * rm -rf .ssh * mkdir .ssh * scp cltbld@try-linux-slave01:/home/cltbld/.ssh/* .ssh * fixed hostnames in /etc/sysconfig/network TODO: * add them to puppet * make sure they sync * put them on a staging try master

Comment on attachment 444424 [details] [diff] [review] add linux64 try slaves to puppet manifests http://hg.mozilla.org/build/puppet-manifests/rev/11d0ca805f03

The try linux 64 bit slaves can't upload to staging-stage since they were trying to use ffxbld which is not on the slaves (as it has to be). It seems that staging-stage was never set up to be used with the trybld key (lsblakk let me know if this is incorrect). I assume I will have to create a trybld account on staging-stage and accept the trybld key.

Comment on attachment 447589 [details] [diff] [review] staging configuration changes sync up with production I'm not sure what you're trying to do here - you can't take out the TRY_$PLATFORM SLAVES from the production code because that is what keeps them in a separate pool. As for staging - I did use ffxbld and staging-stage when testing the rollout of try-as-branch. Now that it's live, we should get a trybld account set up on staging-stage but that's not related to this bug so how about doing that in another bug? I don't mind doing that now, in fact.

I have moved the work for this to bug 563479. I will wrap this bug later to make sure I am not leaving anything loose.

All done in here.