"failed to get shared config profile" - Grafana 12.4 issue

[Mmefko]

What happened:
grafana-amazonprometheus-datasource error after 12.4 update.

I use "Credential file" as Authentication Provider with "default" Credentials profile Name. After updating from Grafana v12.3 to v12.4 this Auth flow stopped working with the following error:

Post "https://aps-workspaces.eu-central-1.amazonaws.com/workspaces/w...21b/api/v1/query": failed to get shared config profile, default - There was an error returned querying the Prometheus API.

Here some additional log lines:

err="failed to get buildinfo: error querying resource: Get \"https://aps-workspaces.eu-central-1.amazonaws.com/workspaces/w...21b/api/v1/status/buildinfo\": failed to get shared config profile, default"

msg="Failed to get prometheus heuristics" dsUid=deusd15u0ao00b err="failed to get buildinfo: error querying resource: Get \"https://aps-workspaces.eu-central-1.amazonaws.com/workspaces/w...21b/api/v1/status/buildinfo\": failed to get shared config profile, default" logger=tsdb.amazon-prometheus pluginId=grafana-amazonprometheus-datasource dsName=AWS-Prometheus-DEV endpoint=checkHealth pluginVersion=2.3.2

What you expected to happen:

I expected a successful connection to Managed Prometheus in Grafana v12.4.

How to reproduce it (as minimally and precisely as possible):

Install Grafana v12.4.0
Install Amazon Managed Service for Prometheus v2.3.2

• Configure datasource with Amazon managed prometheus
• Authentication Provider: Credentials file (placed to ~/.aws/credentials)
• Credentials Profile Name: default (set the profile name, do not leave it blank)
• Assume Role ARN: set the role

Screenshots

Anything else we need to know?:

I use an AWS Cloudwatch data source, which is based on the same credentials file, and everything works fine, so I decided that my credentials are fine. And after rolling back to Grafana 12.3, everything works fine.

Environment:

• Grafana version: 12.4
• Plugin version: 2.3.2 and 2.2.1
• OS Grafana is installed on: K8S Cluster
• Image: "grafana/grafana:12.4.0@sha256:b0ae311af06228bcfd4a620504b653db80f5b91e94dc3dc2a5b7dab202bcde20"
• User OS & Browser: MacOS, Chrome 145.0.7632.117
• Others:

[njvrzm]

Hey @Mmefko - I think this is probably due to this security PR added to Grafana 12.4.0, it's caused a few similar issues with AWS plugins. We have a fix in development but in the mean time if this is the cause you should be able to work around it by adding the following to your grafana INI file:

[plugins]
forward_host_env_vars = grafana-amazonprometheus-datasource

Please let us know if that doesn't help!

[Mmefko]

Thank you @njvrzm
I tested the workaround, and it works, so I will use this solution til the fix in main.