CloudAMQP experiences ops nightmare today

So I just read this, the CEO added the below

    insts = [
      @ec2.instances.with_tag('Name', "CloudAMQP-#@name0*").to_a,
      @ec2.instances.with_tag('Name', "CloudAMQP-#@name-*").to_a,
      @ec2.instances.with_tag('Service', 'CloudAMQP').with_tag('Name', "#@name-*").to_a,
    ].flatten

unless they have a variable named "@name0" that first line would've evaluated to "CloudAMQP-*", which would then have the fun happen.

CloudAMQP's CEO/CTO from the support chat room:

  Yes, it seems to have been a bug in our instance deletion code,
  that deleted all instances in that region, instead of one.
  This of course is a disaster