Update prometheus params passed to run monitored tmpnet cmd

[aaronbuchwald]

This PR updates the Prometheus params passed to run monitored tmpnet command.

Follow up to: #4149

[Copilot]

Pull Request Overview

Updates Prometheus configuration parameters in C-chain reexecution benchmark workflows to align with changes introduced in PR #4149.

• Adds prometheus_url parameter alongside existing push URL parameters
• Updates parameter naming from kebab-case to snake_case for consistency
• Defines new prometheus-url input in the reusable action

Reviewed Changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated no comments.

File	Description
.github/workflows/c-chain-reexecution-benchmark-gh-native.yml	Updates Prometheus parameter names and adds prometheus_url parameter
.github/workflows/c-chain-reexecution-benchmark-container.yml	Updates Prometheus parameter names and adds prometheus_url parameter
.github/actions/c-chain-reexecution-benchmark/action.yml	Adds new prometheus-url input definition

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[aaronbuchwald]

This fixes the metrics: https://grafana-poc.avax-dev.network/d/Gl1I20mnk/c-chain?orgId=1&refresh=10s&var-filter=is_ephemeral_node%7C%3D%7Cfalse&var-filter=gh_repo%7C%3D%7Cava-labs%2Favalanchego&var-filter=gh_run_id%7C%3D%7C18017140372&var-filter=gh_run_attempt%7C%3D%7C1&var-filter=gh_job_id%7C%3D%7Cc-chain-reexecution&from=2025-09-25T18:32:03.000Z&to=2025-09-25T18:47:03.000Z&timezone=America%2FNew_York&var-datasource=P1809F7CD0C75ACF3&var-chain=C

[aaronbuchwald]

TODO: align underscores and dashes for parameters between re-execution and run monitored tmpnet command to reduce confusion.

[aaronbuchwald]

TODO: align underscores and dashes for parameters between re-execution and run monitored tmpnet command to reduce confusion.

Opened a PR here, but would prefer to keep kebab case as it appears to be more standard as mentioned in the candidate PR description: #4349

[maru-ava]

Have you seen metrics being collected with this PR where it wasn't collected from master? That would be awfully strange, because only the push URL is used for metrics collection. The non-push URL is only used to query Prometheus for a non-zero set of the expected labels, and afaik the re-execute jobs are not performing such a check.

[aaronbuchwald]

Have you seen metrics being collected with this PR where it wasn't collected from master? That would be awfully strange, because only the push URL is used for metrics collection. The non-push URL is only used to query Prometheus for a non-zero set of the expected labels, and afaik the re-execute jobs are not performing such a check.

Hmm, you're right this is working fine, so not sure what the reported issue is and suspect it's related to a unique setup.

[aaronbuchwald]

Closing for now as @maru-ava pointed out this is not the issue. May re-open with the suggested check that metrics have bene successfully collected, but not sure where this is.

[aaronbuchwald]

Closing for now as @maru-ava pointed out this is not the issue. May re-open with the suggested check that metrics have bene successfully collected, but not sure where this is.

Re-opened with the recommended check and populating network_uuid to fit tmpnet's expected set of labels.

[aaronbuchwald]

Strange, tmpnet.CheckMetricsExist(...) is now failing in an unexpected way.

I can see the metrics show up including when adding the network_uuid filter here: https://grafana-poc.avax-dev.network/d/Gl1I20mnk/c-chain?orgId=1&refresh=10s&var-filter=is_ephemeral_node%7C%3D%7Cfalse&var-filter=gh_repo%7C%3D%7Cava-labs%2Favalanchego&var-filter=gh_run_id%7C%3D%7C18042855154&var-filter=gh_run_attempt%7C%3D%7C1&var-filter=gh_job_id%7C%3D%7Cc-chain-reexecution&var-filter=network_uuid%7C%3D%7Cb82518dd-ebca-4c8b-a263-2d17611bcb85&from=2025-09-26T15:59:56.000Z&to=2025-09-26T16:14:56.000Z&timezone=America%2FNew_York&var-datasource=P1809F7CD0C75ACF3&var-chain=C

However, it fails to make a request to get the metrics and thus triggers a failure in CI: https://github.com/ava-labs/avalanchego/actions/runs/18042855154/job/51346204250?pr=4343#step:3:1486.

Running locally, this check works fine with the following output and using the shared Prometheus env vars:

[09-30|13:13:58.046] INFO c-chain-reexecution c/vm_reexecute_test.go:207 shutting down DB
[09-30|13:14:10.050] INFO prometheus tmpnet/check_monitoring.go:178 checking if metrics exist {"url": "https://prometheus-poc.avax-dev.network", "query": "count({network_uuid=\"9decc783-93f4-4b0b-8913-53f08d902580\"})"}
[09-30|13:14:10.137] INFO prometheus tmpnet/check_monitoring.go:45 collected count is non-zero {"type": "metrics", "count": 696}
BenchmarkReexecuteRange/[101,10000]-Config-default-Runner-dev-12                       1               190.3 mgas/s
PASS
ok      github.com/ava-labs/avalanchego/tests/reexecute/c       18.045s

I suspect this indicates that the environment variables (perhaps the username/password?) are not being set as expected within this CI run.