Add MCE and PCIe AER sub-checks to check_syslogs

[gustcol]

Summary

• Add mce and pcie-aer sub-commands to the existing check_syslogs Click group
• Detect Machine Check Exceptions (MCE) and PCIe Advanced Error Reporting (AER) messages from dmesg
• Follow the established pattern of xid, link-flaps, and io-errors sub-checks
• Include documentation pages under check-syslogs/ and test coverage

Test plan

# Run the new tests
python -m pytest gcm/tests/health_checks_tests/test_check_syslogs_mce_pcie.py -v

# Run all check_syslogs tests
python -m pytest gcm/tests/health_checks_tests/test_check_syslogs.py gcm/tests/health_checks_tests/test_check_syslogs_mce_pcie.py -v

# Usage examples
health_checks check-syslogs mce [CLUSTER] app
health_checks check-syslogs pcie-aer [CLUSTER] app

Sample output

MCE check (clean):

OK: No MCE errors detected.

MCE check (errors found):

CRITICAL: 2 MCE error(s) detected.

PCIe AER check (corrected errors):

WARN: 1 PCIe AER corrected error(s) detected.

PCIe AER check (uncorrectable errors):

CRITICAL: 2 PCIe AER error(s) detected, including uncorrectable.

[github-actions]

CI Commands

The following CI workflows run automatically on every push and pull request:

Workflow	What it runs
GPU Cluster Monitoring Python CI	lint, tests, typecheck, format, deb build, pyoxidizer builds
Go packages CI	shelper tests, format, lint

---

The following commands can be used by maintainers to trigger additional tests that require access to secrets:

Command	Description	Requires approval?
/metaci tests	Runs Meta internal integration tests (pytest)	Yes — a maintainer must trigger the command and approve the deployment request
/metaci integration tests	Same as above (alias)	Yes

Note: Only repository maintainers (OWNER association) can trigger /metaci commands. After commenting the command, a maintainer must also navigate to the Actions tab and approve the deployment to the graph-api-access environment before the jobs will run. See the approval guidelines for what to approve or reject.

[meta-cla]

Hi @gustcol!

Thank you for your pull request and welcome to our community.

Action Required

In order to merge any pull request (code, docs, etc.), we require contributors to sign our Contributor License Agreement, and we don't seem to have one on file for you.

Process

In order for us to review and merge your suggested changes, please sign at https://code.facebook.com/cla. If you are contributing on behalf of someone else (eg your employer), the individual CLA may not be sufficient and your employer may need to sign the corporate CLA.

Once the CLA is signed, our tooling will perform checks and validations. Afterwards, the pull request will be tagged with CLA signed. The tagging process may take up to 1 hour after signing. Please give it that time before contacting us about it.

If you have received this in error or have any questions, please contact us at cla@meta.com. Thanks!

[meta-cla]

Thank you for signing our Contributor License Agreement. We can now accept your code for this (and any) Meta Open Source project. Thanks!

[luccabb]

@gustcol same comment from #54, if you could add a small doc: https://github.com/facebookresearch/gcm/tree/main/website/docs/GCM_Health_Checks/health_checks

[gustcol]

Hi @luccabb, thanks for the feedback!

I've added the documentation page:

• website/docs/GCM_Health_Checks/health_checks/check-dmesg.md — covers overview, error detection patterns (Xid, MCE, PCIe AER), command-line options, exit conditions, and usage examples

The doc follows the same structure as existing health check pages like xid.md and check-syslogs/README.md.

Let me know if you'd like any adjustments!

[luccabb]

what do you think of:

1. keeping XIDs at: https://facebookresearch.github.io/gcm/docs/GCM_Health_Checks/health_checks/check-syslogs/xid/

2. moving Machine Check Exceptions (MCE), and PCIe AER errors to be their own checks under check-syslogs? https://facebookresearch.github.io/gcm/docs/GCM_Health_Checks/health_checks/check-syslogs/

[gustcol]

Hi @luccabb, great suggestion! I've refactored the PR:

• Removed the standalone check_dmesg command entirely
• Added mce and pcie-aer as new sub-commands under check_syslogs, following the same pattern as xid, link-flaps, and io-errors

Now the usage looks like:

# MCE check
health_checks check-syslogs mce [CLUSTER] app

# PCIe AER check
health_checks check-syslogs pcie-aer [CLUSTER] app

Each sub-check has its own:

• Protocol method in Syslog + implementation in SyslogImpl
• Processing function (process_mce_output, process_pcie_aer_output)
• HealthCheckName enum entry
• Feature flag killswitch
• Documentation page under check-syslogs/
• Test coverage (7 new tests, 22 total passing)

XIDs remain at their current location in check_syslogs xid as you suggested.