aws: check for all possible Terminated states

[tomwans]

What type of PR is this?
/kind bug

What this PR does / why we need it:
It's possible that an EC2 instance is in the Terminated state for multiple autoscaler loops, so to avoid trying to terminate it more than once and causing the bug seen for this initial fix, we include this check as well.

Also, make the log more verbose and write out what state we found the instance in.

This builds upon this change: #5411

[k8s-ci-robot]

Welcome @tomwans!

It looks like this is your first PR to kubernetes/autoscaler 🎉. Please refer to our pull request process documentation to help your PR have a smooth ride to approval.

You will be prompted by a bot to use commands during the review process. Do not be afraid to follow the prompts! It is okay to experiment. Here is the bot commands documentation.

You can also check if kubernetes/autoscaler has its own contribution guidelines.

You may want to refer to our testing guide if you run into trouble with your tests not passing.

If you are having difficulty getting your pull request seen, please follow the recommended escalation practices. Also, for tips and tricks in the contribution process you may want to read the Kubernetes contributor cheat sheet. We want to make sure your contribution gets all the attention it needs!

Thank you, and welcome to Kubernetes. 😃

[drmorr0]

@tomwans Hi! Thanks for the PR, and also for adding a test in. I'm a bit confused on the test; it seems like there are two instances, and one of them was already in the terminating state? So we expect the instance count to go from 2 -> 1? But at the end we are asserting that the target size is still 2. Maybe I'm missing something, though.

[tomwans]

@drmorr0 thanks for the review. I cleaned up the test a bit, please let me know how it looks now. It just seems like the TerminatingWait test was copy pasted from something else, with light adjustments. The expectedInstanceCount variable is not really relevant, it was just a distraction. I reworded some of that from the Terminated test I added.

If you'd like, I can copy paste this and do the same for the TerminatingWait test, so they look the same. Or, I can just bring them all in to one test and do a table test, which might be fine since we expect the same result out of all the states anyway. I have no preference. What do you think?

[drmorr0]

@drmorr0 thanks for the review. I cleaned up the test a bit, please let me know how it looks now. It just seems like the TerminatingWait test was copy pasted from something else, with light adjustments. The expectedInstanceCount variable is not really relevant, it was just a distraction. I reworded some of that from the Terminated test I added.

If you'd like, I can copy paste this and do the same for the TerminatingWait test, so they look the same. Or, I can just bring them all in to one test and do a table test, which might be fine since we expect the same result out of all the states anyway. I have no preference. What do you think?

@tomwans If you have some time and it's not terribly difficult, putting into a table test would be nice, I guess? But I don't have particularly strong feelings on the matter, so if you don't have time/want to just ship it, I'll stamp it.

[tomwans]

Let's just stamp it out. Theres no other case of table tests in this test file, so it wouldn't be matching conventions anyway.

[drmorr0]

/lgtm
/approve

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: drmorr0, tomwans

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• cluster-autoscaler/cloudprovider/aws/OWNERS [drmorr0]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment