Skip to content

Instantly share code, notes, and snippets.

@grenade
Last active September 15, 2015 10:13
Show Gist options
  • Select an option

  • Save grenade/7da25768362baff316e2 to your computer and use it in GitHub Desktop.

Select an option

Save grenade/7da25768362baff316e2 to your computer and use it in GitHub Desktop.
  • We can't use the 2008-spot names for manually created instances. It confuses cloud-tools (we spin up duplicate instances when cloud-tools finds the enabled slave name). We can use the on-demand instance names (buildbot: staging, production):
    • staging parallel builds, not reported in treeherder (ensure locked_masterid == 372 in slavealloc):
      • us-east-1:
        • build: b-2008-ec2-0990 - b-2008-ec2-0994
        • try: y-2008-ec2-0990 - y-2008-ec2-0994
      • us-west-2:
        • build: b-2008-ec2-0995 - b-2008-ec2-0999
        • try: y-2008-ec2-0995 - y-2008-ec2-0999
    • production primary builds, reported in treeherder (remove locked_masterid in slavealloc):
      • us-east-1:
        • build: b-2008-ec2-0001 - b-2008-ec2-0015
        • try: y-2008-ec2-0001 - y-2008-ec2-0015
      • us-west-2:
        • build: b-2008-ec2-0016 - b-2008-ec2-0030
        • try: y-2008-ec2-0016 - y-2008-ec2-0030
  • Anyone know if we have an orphaned volume cleanup process in ec2? Eg: delete volumes for instances which have been terminated but didn't have the delete-volume-on-termination flag set. We're currently showing 60 'available' volumes averaging 70 gig each. us-east-1, us-west-2
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment