Use SDK's InstanceReady waiter

[mwhooker]

Closes #5705

This replaces our home grown instance ready waiter with the one from the SDK. This brings build times back to where they were in the previous release, but without the errors.

I cannot for the life of me reproduce the error with this patch, but since the error isn't falsifiable, it's going to be hard to tell if this fixes it completely. I'll keep running builds in a loop and see if I see any problems, but I think we'll probably be good with this patch

[SwampDragons]

I think we should have a comment explaining why this convoluted extra code for c5. something like // c5 instances need us to use the time-sucking WaitUntilStatusOK feature instead of just waiting for SSH to become available

[lorengordon]

Pretty sure the new m5 instance type is also backed by the new hypervisor. Ought to be easy to test if the issue is limited to c5, or if a more general approach is needed.

[SwampDragons]

Apart from future-proofing the code with a small "why" comment, I think this is fine. I know you weren't a fan of special casing it, but it's probably the best way to go.

[rickard-von-essen]

I think it would be better to check if the virtualization type is the new KVM based type. I'm assuming all new instance types on that will be affected by this.

[rickard-von-essen]

Isn't this still needed for C5's?

[patrickf55places]

Why does the KVM hypervisor make the status check necessary? Why isn't SSH access enough?

[mwhooker]

@patrickf55places because the IP isn't available until some time after the InstanceReady check

@rickard-von-essen

I think it would be better to check if the virtualization type is the new KVM based type. I'm assuming all new instance types on that will be affected by this.

will investigate if that's feasible, but it sounds way better

[mwhooker]

I've taken this in an odd direction. This replaces our home grown instance ready waiter with the one from the SDK. This brings build times back to where they were in the previous release, but without the errors.

I cannot for the life of me reproduce the error with this patch, but since the error isn't falsifiable, it's going to be hard to tell if this fixes it completely. I'll keep running builds in a loop and see if I see any problems, but I think we'll probably be good with this patch

[mwhooker]

I very occasionally get ssh timeout errors, but I don't think these are related to discovering the public IP

~/dev/packertest ᐅ packer build -only 5627-ebs 5627.json
5627-ebs output will be in this color.

==> 5627-ebs: Prevalidating AMI Name: packer-qs-1515628931
    5627-ebs: Found Image ID: ami-cb9ec1b1
==> 5627-ebs: Creating temporary keypair: packer_5a56a983-6f58-dfe8-b7bf-4fdc80c15b14
==> 5627-ebs: Creating temporary security group for this instance: packer_5a56a98a-3c63-0cde-272a-4e72984db1fd
==> 5627-ebs: Authorizing access to port 22 from 0.0.0.0/0 in the temporary security group...
==> 5627-ebs: Launching a source AWS instance...
==> 5627-ebs: Adding tags to source instance
    5627-ebs: Adding tag: "Name": "Packer Builder"
    5627-ebs: Instance ID: i-0e105264a5b0995d1
==> 5627-ebs: Waiting for instance (i-0e105264a5b0995d1) to become ready...
    5627-ebs: Public DNS: ec2-54-147-134-244.compute-1.amazonaws.com
    5627-ebs: Public IP: 54.147.134.244
    5627-ebs: Private IP: 172.31.50.130
==> 5627-ebs: Waiting for SSH to become available...

==> 5627-ebs: Timeout waiting for SSH.
==> 5627-ebs: Terminating the source AWS instance...
==> 5627-ebs: Cleaning up any extra volumes...
==> 5627-ebs: Deleting temporary security group...
==> 5627-ebs: Deleting temporary keypair...
Build '5627-ebs' errored: Timeout waiting for SSH.

[SwampDragons]

Way better and simpler. Verified it works with both c5 instance types and older instance types.

[sparkcodeuk]

👍 had same issue with a t2.nano build, was missing "ec2:DescribeInstanceStatus" permission on packer IAM user

[ghost]

I'm going to lock this issue because it has been closed for 30 days ⏳. This helps our maintainers find and focus on the active issues.

If you have found a problem that seems similar to this, please open a new issue and complete the issue template so we can capture all the details necessary to investigate further.