search

docker-archive / for-aws

92 stars 26 forks source link

ELB listeners not recreated when instance is added to empty auto-scaling group #120

Closed Mobe91 closed 6 years ago

Mobe91 commented 6 years ago

Steps to reproduce the behavior

  1. Create a swarm with 1 manager node and 1 worker node
  2. Start a replicated service using docker deploy ... with a published port (e.g. via --publish 80:80) on the ELB
  3. Alter the worker auto scaling group to force the shutdown of all instances, i.e. set desired=0 and min=0
  4. Wait until instances have been shut down and the ELB listener for port 80 has been removed
  5. Restore the previous configuration of the worker auto-scaling group (e.g. desired=1 and min=0)
  6. Wait until a new instance is started

Expected behavior

After step 6 I expect the ELB to contain a listener for port 80.

Actual behavior

There is no such listener present step 6.

Information

OK hostname=ip-172-31-4-42-eu-central-1-compute-internal session=1510758358-yDq0TzVZO0BTIwIeBJs5v9rga8Pelfyt
OK hostname=ip-172-31-0-232-eu-central-1-compute-internal session=1510758358-yDq0TzVZO0BTIwIeBJs5v9rga8Pelfyt
Done requesting diagnostics.
Your diagnostics session ID is 1510758358-yDq0TzVZO0BTIwIeBJs5v9rga8Pelfyt
Please provide this session ID to the maintainer debugging your issue.
Mobe91 commented 6 years ago

Hm let me just test this again before you look into it, not sure if I did something wrong...

Mobe91 commented 6 years ago

My mistake, this works as intended!!