Error while validating API key: https://api.datadoghq.eu/api/v1/validate: Not Found

[AxelJoly]

Component(s)

exporter/datadog

What happened?

Description

I'm running the opentelemetry-collector-contrib:0.82.0 on my EKS cluster. I tried as well on previous version (0.75) and got the same result.
With the config below, I get the following log:

2023-11-22T08:25:43.051Z warn clientutil/api.go:48 Error while validating API key {"kind": "exporter", "data_type": "metrics", "name": "datadog", "error": "Get \"https://api.datadoghq.eu/api/v1/validate\": Not Found"}

When I run the following command on my terminal, I have no issues:

curl -X GET "https://api.datadoghq.eu/api/v1/validate" -H "Accept: application/json" -H "DD-API-KEY: myapikey" {"valid":true}
I tried it in a sidecar container and it's working fine as well with the curl command.

Steps to Reproduce

Run opentelemetry-collector-contrib:0.82.0 image
add the following config with a valid key

Expected Result

The /api/v1/validate should be validating my key

Actual Result

The /api/v1/validate doesn't validate my key

Collector version

0.82

Environment information

Environment

EKS 1.25, running on Amazon Linux 2 AMIs

OpenTelemetry Collector configuration

exporters:
  datadog:
    api:
      key: myapikey
      site: datadoghq.eu
    tls:
      insecure_skip_verify: true

Log output

`2023-11-22T08:25:43.051Z    warn    clientutil/api.go:48    Error while validating API key    {"kind": "exporter", "data_type": "metrics", "name": "datadog", "error": "Get \"https://api.datadoghq.eu/api/v1/validate\": Not Found"}`

Additional context

No response

[github-actions]

Pinging code owners:

• exporter/datadog: @mx-psi @gbbr @dineshg13 @liustanley @songy23 @mackjmr

See Adding Labels via Comments if you do not have permissions to add labels yourself.

[mx-psi]

@AxelJoly Is it possible you are using some kind of proxy? To validate this, you can try running the Collector locally with a minimal configuration and see if the API key is correctly validated.

[AxelJoly]

Hi @mx-psi , yes we are behind a proxy, we added a bunch of proxy settings in the deployment file.

      - env:
        - name: HTTP_PROXY
          value: http://myproxy:80
        - name: HTTPS_PROXY
          value:  http://myproxy:80
        - name: NO_PROXY
          value: 127.0.0.1,localhost,datadog-agent.datadog.svc.cluster.local:8126,172.25.0.0/16,169.254.0.0/16,10.147.0.0/16

For the sidecar test, I added the proxy option to my curl command and it works. Could it be the datadog ignoring the proxy context?
Thank you for your fast answer.

[mx-psi]

Could it be the datadog ignoring the proxy context?

The HTTP client used for validation takes the proxy from the environment variables

opentelemetry-collector-contrib/exporter/datadogexporter/internal/clientutil/http.go

Line 35 in d79f679

Proxy: http.ProxyFromEnvironment,

Could it be that you are not setting the environment variables correctly in the Collector's environment?

[github-actions]

This issue has been inactive for 60 days. It will be closed in 60 days if there is no activity. To ping code owners by adding a component label, see Adding Labels via Comments, or if you are unsure of which component this issue relates to, please ping @open-telemetry/collector-contrib-triagers. If this issue is still relevant, please ping the code owners or leave a comment explaining why it is still relevant. Otherwise, please close it.

Pinging code owners:

• exporter/datadog: @mx-psi @dineshg13 @liustanley @songy23 @mackjmr

See Adding Labels via Comments if you do not have permissions to add labels yourself.

[mx-psi]

Closing this as we haven't gotten a reply since November, @AxelJoly feel free to comment and we will reopen