Collation metrics: exclude drops of fork-based collations to improve metrics accuracy #9319
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
AndreiEres
changed the title
AndreiEres
changed the title
AndreiEres
changed the title
Contributor
Author
|
/cmd prdoc --audience node_dev --bump patch |
sandreim
reviewed
Aug 18, 2025
polkadot/node/network/collator-protocol/src/collator_side/mod.rs
Outdated
Show resolved
Hide resolved
polkadot/node/network/collator-protocol/src/collator_side/mod.rs
Outdated
Show resolved
Hide resolved
Sajjon
suggested changes
Aug 19, 2025
Contributor
Sajjon
left a comment
Sajjon
left a comment
There was a problem hiding this comment.
Did not get it all, but found some wrong doc comment use.
polkadot/node/network/collator-protocol/src/collator_side/metrics.rs
Outdated
Show resolved
Hide resolved
polkadot/node/network/collator-protocol/src/collator_side/metrics.rs
Outdated
Show resolved
Hide resolved
polkadot/node/network/collator-protocol/src/collator_side/metrics.rs
Outdated
Show resolved
Hide resolved
sandreim
approved these changes
Sep 11, 2025
| /// Returns true if the collation was included in a block before (or in) last finalized. | ||
| pub fn is_possibly_finalized(&self, last_finalized: BlockNumber) -> bool { | ||
| self.included_at | ||
| .map(|included_at| included_at <= last_finalized) |
Contributor
There was a problem hiding this comment.
If we see some colation included it doesn't really mean it is finalized, because we don't even know if it is on the fork that is getting finalized. However, this should still work fine, because eventually the collation should be backed/included eventually as it was clearly backed offchain. Only issue is that the relay parent expires.
Please add some docs about the limitations of the measurement we are doing.
AndreiEres
added
A4-backport-stable2503
|
Created backport PR for
Please cherry-pick the changes locally and resolve any conflicts. git fetch origin backport-9319-to-stable2503
git worktree add --checkout .worktree/backport-9319-to-stable2503 backport-9319-to-stable2503
cd .worktree/backport-9319-to-stable2503
git reset --hard HEAD^
git cherry-pick -x d51c532f07c7e3307718b95f5a1f8859e14949a0
git push --force-with-lease |
|
Created backport PR for
Please cherry-pick the changes locally and resolve any conflicts. git fetch origin backport-9319-to-stable2506
git worktree add --checkout .worktree/backport-9319-to-stable2506 backport-9319-to-stable2506
cd .worktree/backport-9319-to-stable2506
git reset --hard HEAD^
git cherry-pick -x d51c532f07c7e3307718b95f5a1f8859e14949a0
git push --force-with-lease |
paritytech-release-backport-bot bot
pushed a commit
that referenced
this pull request
Oct 1, 2025
…metrics accuracy (#9319) # Description The polkadot_parachain_collation_expired metric is an indicator for parachain block confidence. However, this metric has a critical issue: not every drop should be counted. Lookahead collators intentionally build collations on a relay chain block and its forks, so the drop of fork-based collations is an expected behaviour. If we count them, the drop metrics show a picture that is worse than in reality. To improve tracking accuracy, we should exclude legit drops The minor issue is also present in the expiry mechanism. It doesn't take into account that collation was moved to a different stage, e.g., from "fetched" to "backed", and can write a drop of fetched collation. To solve this issue we should: - Track relay parent finalization. - Record expiration metrics only when relay parent was finalized. - Exclude drops of fork-based collation from the metrics. - Send metrics only for collations that either finalized or dropped. --------- Co-authored-by: cmd[bot] <41898282+github-actions[bot]@users.noreply.github.com> (cherry picked from commit d51c532)
|
Successfully created backport PR for |
paritytech-release-backport-bot bot
pushed a commit
that referenced
this pull request
Oct 1, 2025
…metrics accuracy (#9319) # Description The polkadot_parachain_collation_expired metric is an indicator for parachain block confidence. However, this metric has a critical issue: not every drop should be counted. Lookahead collators intentionally build collations on a relay chain block and its forks, so the drop of fork-based collations is an expected behaviour. If we count them, the drop metrics show a picture that is worse than in reality. To improve tracking accuracy, we should exclude legit drops The minor issue is also present in the expiry mechanism. It doesn't take into account that collation was moved to a different stage, e.g., from "fetched" to "backed", and can write a drop of fetched collation. To solve this issue we should: - Track relay parent finalization. - Record expiration metrics only when relay parent was finalized. - Exclude drops of fork-based collation from the metrics. - Send metrics only for collations that either finalized or dropped. --------- Co-authored-by: cmd[bot] <41898282+github-actions[bot]@users.noreply.github.com> (cherry picked from commit d51c532)
|
Successfully created backport PR for |
AndreiEres
removed
A4-backport-stable2503
EgorPopelyaev
added a commit
that referenced
this pull request
Oct 1, 2025
Backport #9319 into `stable2509` from AndreiEres. See the [documentation](https://github.com/paritytech/polkadot-sdk/blob/master/docs/BACKPORT.md) on how to use this bot. <!-- # To be used by other automation, do not modify: original-pr-number: #${pull_number} --> Co-authored-by: Andrei Eres <[email protected]> Co-authored-by: cmd[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: Egor_P <[email protected]>
bee344
pushed a commit
that referenced
this pull request
Oct 7, 2025
…metrics accuracy (#9319) # Description The polkadot_parachain_collation_expired metric is an indicator for parachain block confidence. However, this metric has a critical issue: not every drop should be counted. Lookahead collators intentionally build collations on a relay chain block and its forks, so the drop of fork-based collations is an expected behaviour. If we count them, the drop metrics show a picture that is worse than in reality. To improve tracking accuracy, we should exclude legit drops The minor issue is also present in the expiry mechanism. It doesn't take into account that collation was moved to a different stage, e.g., from "fetched" to "backed", and can write a drop of fetched collation. To solve this issue we should: - Track relay parent finalization. - Record expiration metrics only when relay parent was finalized. - Exclude drops of fork-based collation from the metrics. - Send metrics only for collations that either finalized or dropped. --------- Co-authored-by: cmd[bot] <41898282+github-actions[bot]@users.noreply.github.com>
alvicsam
pushed a commit
that referenced
this pull request
Oct 17, 2025
…metrics accuracy (#9319) # Description The polkadot_parachain_collation_expired metric is an indicator for parachain block confidence. However, this metric has a critical issue: not every drop should be counted. Lookahead collators intentionally build collations on a relay chain block and its forks, so the drop of fork-based collations is an expected behaviour. If we count them, the drop metrics show a picture that is worse than in reality. To improve tracking accuracy, we should exclude legit drops The minor issue is also present in the expiry mechanism. It doesn't take into account that collation was moved to a different stage, e.g., from "fetched" to "backed", and can write a drop of fetched collation. To solve this issue we should: - Track relay parent finalization. - Record expiration metrics only when relay parent was finalized. - Exclude drops of fork-based collation from the metrics. - Send metrics only for collations that either finalized or dropped. --------- Co-authored-by: cmd[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
The polkadot_parachain_collation_expired metric is an indicator for parachain block confidence. However, this metric has a critical issue: not every drop should be counted.
Lookahead collators intentionally build collations on a relay chain block and its forks, so the drop of fork-based collations is an expected behaviour. If we count them, the drop metrics show a picture that is worse than in reality. To improve tracking accuracy, we should exclude legit drops
The minor issue is also present in the expiry mechanism. It doesn't take into account that collation was moved to a different stage, e.g., from "fetched" to "backed", and can write a drop of fetched collation.
To solve this issue we should: