Skip to content

Additional collator revamp metrics#12283

Draft
tdimitrov wants to merge 7 commits into
tsv-revamp-loggingfrom
tsv-revamp-metrics
Draft

Additional collator revamp metrics#12283
tdimitrov wants to merge 7 commits into
tsv-revamp-loggingfrom
tsv-revamp-metrics

Conversation

@tdimitrov

@tdimitrov tdimitrov commented Jun 5, 2026

Copy link
Copy Markdown
Contributor

On top of #12282

Additinal metrics for the collator protocol revamp:

  • _validator_advertisements_total - number of received advertisements, by outcome (accepted, rejected + the exact reason as a label)
  • _validator_collations_seconded_total - collations sent to backing for secondig
  • _validator_slashes_total - collator slashes
  • _validator_approved_peer_signals_total - number of ApprovedPeer UMP signals received, grouped by processing outcome (approved, rejected + reason)
  • _validator_assigned_paras - how many paras are currently assigned
  • _validator_connected_collators - number of currently connected collators, grouped by score

Regarding _validator_advertisements_total

All different outcomes are tracked so that we can deduce from the metric why advertisements are getting rejected.

Regarding _validator_connected_collators

Collators are grouped in three different bands:

  • "0" - zero score, severely slashed or brand new collator.
  • "1-99" - low score, These are relatively new collators building up reputation.
  • "100-999" - medium score, established collators.
  • "1000+" - high score, highly trusted collators.

Reviewers, please pay attention to the following questions:

  • Is advertisement tracking an overkill? I was torn between simple accepted/rejected and the variant currently in the PR. I think having separate error cases is a bit better but I'm not sure we'll actually use them for monitoring.
  • Validator score - similar question. I was considering counting 0 reputation peers and 1+ reputation peers but went for the 4 bands.

Partially addresses #10402

@tdimitrov tdimitrov requested review from alindima and sandreim June 5, 2026 13:19
@tdimitrov tdimitrov added T8-polkadot This PR/Issue is related to/affects the Polkadot network. A4-backport-unstable2604 Pull request must be backported to the unstable2604 release branch A4-backport-stable2606 Pull request must be backported to the stable2606 release branch labels Jun 5, 2026
@tdimitrov

Copy link
Copy Markdown
Contributor Author

/cmd prdoc --audience node_dev --bump patch

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

A4-backport-stable2606 Pull request must be backported to the stable2606 release branch A4-backport-unstable2604 Pull request must be backported to the unstable2604 release branch T8-polkadot This PR/Issue is related to/affects the Polkadot network.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant