Invalid amis create an erroneous credentials error
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Canonical Juju |
Fix Released
|
High
|
Ian Booth |
Bug Description
I ran into this while attempting to reproduce https:/
When I juju deploy a centos7 image on aws, my model (and controllers) gets paused, due to what juju things is a credential error. This wedges the controller.
Here are the commands I ran to get into a bad state (This is using jaas):
juju add-model bug1873490-centos7 aws/us-east-1
juju add-machine --series centos7
watch --color 'juju status --color' # wait for error to appear
Here's the full text of my error message that I see after things settle out:
```
failed to start machine 2 in zone "us-east-1c", retrying in 10s with new availability zone: cannot run instances:
The provided credentials could not be validated and
may not be authorized to carry out the request.
Ensure that your account is authorized to use the Amazon EC2 service and
that you are using the correct access keys.
These keys are obtained via the "Security Credentials"
page in the AWS console.
: Not authorized for images: [ami-6d1c2007] (AuthFailure)
```
"ami-6d1c2007" is not the id of the centos image for us-east-1, according to the table at https:/
Changed in juju: | |
milestone: | none → 2.8-rc1 |
status: | New → Triaged |
importance: | Undecided → High |
Changed in juju: | |
assignee: | nobody → Ian Booth (wallyworld) |
status: | Triaged → In Progress |
Changed in juju: | |
status: | In Progress → Fix Committed |
Changed in juju: | |
status: | Fix Committed → Fix Released |
When Juju calls a cloud api and gets back an AuthFailure error, it interprets that as the credential being invalid. Here though, an AuthFailure is returned trying to use a certain image. So Juju will need to try and introspect the error if possible to determine the difference.