`grafana_synthetic_monitoring_probe` could not get labels

[anhdle14]

Terraform Version

• Terraform: 1.2.0
• Terraform Grafana Provider: 1.24.0
• Grafana: 8.0+

Affected Resource(s)

Please list the resources as a list, for example:

• grafana_synthetic_monitoring_probe
• data.grafana_synthetic_monitoring_probes

Terraform Configuration Files

data "grafana_synthetic_monitoring_probes" "this" {}

Debug Output

Please provider a link to a GitHub Gist containing the complete debug output: https://www.terraform.io/docs/internals/debugging.html. Please do NOT paste the debug output in the issue; just paste a link to the Gist.

• Running from CI so if you need I can show the output later.

Panic Output

If Terraform produced a panic, please provide a link to a GitHub Gist containing the output of the crash.log.

Expected Behavior

What should have happened?

• No issues

Actual Behavior

What actually happened?

│ Error: probe list request: status="500 Internal Server Error", msg="status: Internal Server Error, error: loading probe 2432: extracting probe labels: extracting label "0": expected name=value", err="loading probe 2432: extracting probe labels: extracting label "0": expected name=value"

Steps to Reproduce

Please list the steps required to reproduce the issue, for example:

1. terraform apply

Important Factoids

Are there anything atypical about your accounts that we should know? For example: Running in EC2 Classic? Custom version of OpenStack? Tight ACLs? N/A

References

Are there any other GitHub issues (open or closed) or Pull Requests that should be linked here? For example: N/A

[mem]

This one looks like something we need to fix in the terraform provider.

It seems like it's passing invalid labels to the API.

[julienduchesne]

I tried a test here: https://github.com/grafana/terraform-provider-grafana/pull/573 and I couldn't reproduce it. Do you have a custom probe? And if so can you post its configuration here?

[anhdle14]

@julienduchesne yes I am using private probes right now. It is a little bit complex but can be summarized with this one.

# terraform.tfvars
probes = [
  {
    name      = probe.name
    latitude  = probe.latitude
    longitude = probe.longitude
    region    = probe.region

    uuid       = probe.uuid
    alt_region = lookup( # convert AZ to Region via probe's name.
      {
        "aws" = substr(regex("(.*)_(.*)_(.*)", probe.name)[2], 0, length(regex("(.*)_(.*)_(.*)", probe.name)[2]) - 1)
        "gcp" = substr(regex("(.*)_(.*)_(.*)", probe.name)[2], 0, length(regex("(.*)_(.*)_(.*)", probe.name)[2]) - 2)
      },
      regex("(.*)_(.*)_(.*)", probe.name)[1], probe.region
    )
  },
  ...
]

# main.tf
resource "grafana_synthetic_monitoring_probe" "private" {
  for_each = { for probe in var.probes : probe.name => probe }

  name      = lower(each.value.name)
  latitude  = each.value.latitude
  longitude = each.value.longitude
  region    = each.value.region
  public    = false # Private probe is not public by default
  labels = merge(each.value.labels, {
    "uuid"       = each.value.uuid
    "alt_region" = each.value.alt_region
  })

  lifecycle {
    create_before_destroy = true
  }
}

[julienduchesne]

Hey @anhdle14. This is probably caused by a comma in your label's value. The API has a bit of trouble with it. It will be fixed, but in the meantime, if you sanitize your label value, it should work. Let me know if that helps!

[anhdle14]

Hey @julienduchesne, we do not have comma or custom characters anywhere I think, except for _ and -. So I am not sure how to fix.

[julienduchesne]

Do you still have the issue? I believe a fix was implemented on the API side

[mem]

Right, we recently deployed a new version of the API that addresses this.

[julienduchesne]

I'll close this until there is confirmation that this is still happening. Feel free to reopen!

[anhdle14]

Sorry, it is fixed now. Thank you for the support ^^!

[julienduchesne]

No worries. Thanks for reporting!