Skip to content

Octopus Cloud Dynamic Worker AWS Toolset Updates May Cause AWS Related Deployment Errors #9951

@Clare-Octopus

Description

@Clare-Octopus

Severity

May cause deployment errors

Version

N/A - This is a Octopus Cloud hosted Dynamic Worker issue - it does not affect Octopus Server

Latest Version

Not applicable

What happened?

Recent commits to our Dynamic Worker toolset, specifically to a more recent version of the aws iam authenticator has resulted in some AWS interactions to fail. We are currently investigating this as a priority.

This can be displayed as warnings in your logs similar to the below:

level=info msg="failed to get region from IMDS for token generation, defaulting to us-east-1. imds error: operation error ec2imds: GetRegion, failed to get API token, operation error ec2imds: getToken, http response error StatusCode: 400, request to EC2 IMDS failed"

It can also cause some deployments to fail with errors in the logs similar to the below:

E0401 18:01:22.791272   63172 memcache.go:265] "Unhandled Error" err="couldn't get current server API group list: the server has asked for the client to provide credentials" 
error: You must be logged in to the server (the server has asked for the client to provide credentials) 
"/opt/homebrew/bin/kubectl" create namespace default --request-timeout=1m 
error: You must be logged in to the server (Unauthorized)

Or

aws: [ERROR]: Invalid endpoint: https://sts/..amazonaws.com

Reproduction

Have a deployment which involves AWS Resources and uses our Octopus Cloud Dynamic workers.

Error and Stacktrace

"/opt/homebrew/bin/kubectl" get namespace default --request-timeout=1m 
E0401 18:01:22.495277   63172 memcache.go:265] "Unhandled Error" err="couldn't get current server API group list: the server has asked for the client to provide credentials" 
E0401 18:01:22.570413   63172 memcache.go:265] "Unhandled Error" err="couldn't get current server API group list: the server has asked for the client to provide credentials" 
E0401 18:01:22.636382   63172 memcache.go:265] "Unhandled Error" err="couldn't get current server API group list: the server has asked for the client to provide credentials" 
E0401 18:01:22.709735   63172 memcache.go:265] "Unhandled Error" err="couldn't get current server API group list: the server has asked for the client to provide credentials" 
E0401 18:01:22.791272   63172 memcache.go:265] "Unhandled Error" err="couldn't get current server API group list: the server has asked for the client to provide credentials" 
error: You must be logged in to the server (the server has asked for the client to provide credentials) 
"/opt/homebrew/bin/kubectl" create namespace default --request-timeout=1m 
error: You must be logged in to the server (Unauthorized)

aws: [ERROR]: Invalid endpoint: https://sts..amazonaws.com

More Information

Initial ticket (internal) - https://octopuscd.zendesk.com/agent/tickets/196906
RnD thread (Internal) - https://octopusdeploy.slack.com/archives/CNHBHV2BX/p1774888154681829

Workaround

If you are affected, adding your AWS region as an environment variable in your script steps may resolve the issue:

export AWS_REGION=[REGION] for Bash
$env:AWS_REGION = "[REGION]" for Powershell

E.g

export AWS_REGION=ap-southeast-2
$env:AWS_REGION = "ap-southeast-2"

Metadata

Metadata

Labels

kind/bugThis issue represents a verified problem we are committed to solving

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions