feat: add prometheus metrics network_version and actor_version
#6079
There are no files selected for viewing
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
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
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
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
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
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -1,47 +1,162 @@ | ||
| // Copyright 2019-2025 ChainSafe Systems | ||
| // SPDX-License-Identifier: Apache-2.0, MIT | ||
|
|
||
| use std::sync::Arc; | ||
|
|
||
| use educe::Educe; | ||
| use prometheus_client::{ | ||
| collector::Collector, | ||
| encoding::{DescriptorEncoder, EncodeMetric}, | ||
| metrics::gauge::Gauge, | ||
| }; | ||
|
|
||
| use super::calculate_expected_epoch; | ||
| use crate::{networks::ChainConfig, shim::clock::ChainEpoch}; | ||
|
|
||
| #[derive(Educe)] | ||
| #[educe(Debug)] | ||
| pub struct NetworkHeightCollector<F> | ||
| where | ||
| F: Fn() -> ChainEpoch, | ||
| { | ||
| block_delay_secs: u32, | ||
| genesis_timestamp: u64, | ||
| #[educe(Debug(ignore))] | ||
| get_chain_head_height: Arc<F>, | ||
| } | ||
|
|
||
| impl<F> NetworkHeightCollector<F> | ||
| where | ||
| F: Fn() -> ChainEpoch, | ||
| { | ||
| pub fn new( | ||
| block_delay_secs: u32, | ||
| genesis_timestamp: u64, | ||
| get_chain_head_height: Arc<F>, | ||
| ) -> Self { | ||
| Self { | ||
| block_delay_secs, | ||
| genesis_timestamp, | ||
| get_chain_head_height, | ||
| } | ||
| } | ||
| } | ||
|
|
||
| impl<F> Collector for NetworkHeightCollector<F> | ||
| where | ||
| F: Fn() -> ChainEpoch + Send + Sync + 'static, | ||
| { | ||
| fn encode( | ||
| &self, | ||
| mut encoder: prometheus_client::encoding::DescriptorEncoder, | ||
| ) -> Result<(), std::fmt::Error> { | ||
| { | ||
| let network_height: Gauge = Default::default(); | ||
| let epoch = (self.get_chain_head_height)(); | ||
| network_height.set(epoch); | ||
| let metric_encoder = encoder.encode_descriptor( | ||
| "network_height", | ||
| "The current network height", | ||
| None, | ||
| network_height.metric_type(), | ||
| )?; | ||
| network_height.encode(metric_encoder)?; | ||
| } | ||
| { | ||
| let expected_network_height: Gauge = Default::default(); | ||
| let expected_epoch = calculate_expected_epoch( | ||
| chrono::Utc::now().timestamp() as u64, | ||
| self.genesis_timestamp, | ||
| self.block_delay_secs, | ||
| ); | ||
| expected_network_height.set(expected_epoch); | ||
| let metric_encoder = encoder.encode_descriptor( | ||
| "expected_network_height", | ||
| "The expected network height based on the current time and the genesis block time", | ||
| None, | ||
| expected_network_height.metric_type(), | ||
| )?; | ||
| expected_network_height.encode(metric_encoder)?; | ||
| } | ||
| Ok(()) | ||
| } | ||
| } | ||
|
|
||
| #[derive(Educe)] | ||
| #[educe(Debug)] | ||
| pub struct NetworkVersionCollector<F1, F2> | ||
| where | ||
| F1: Fn() -> ChainEpoch, | ||
| F2: Fn() -> u64, | ||
| { | ||
| chain_config: Arc<ChainConfig>, | ||
| #[educe(Debug(ignore))] | ||
| get_chain_head_height: Arc<F1>, | ||
| #[educe(Debug(ignore))] | ||
| get_chain_head_actor_version: Arc<F2>, | ||
| } | ||
|
|
||
| impl<F1, F2> NetworkVersionCollector<F1, F2> | ||
| where | ||
| F1: Fn() -> ChainEpoch, | ||
| F2: Fn() -> u64, | ||
| { | ||
| pub fn new( | ||
| chain_config: Arc<ChainConfig>, | ||
| get_chain_head_height: Arc<F1>, | ||
| get_chain_head_actor_version: Arc<F2>, | ||
| ) -> Self { | ||
| Self { | ||
| chain_config, | ||
| get_chain_head_height, | ||
| get_chain_head_actor_version, | ||
| } | ||
| } | ||
| } | ||
|
|
||
| impl<F1, F2> Collector for NetworkVersionCollector<F1, F2> | ||
| where | ||
| F1: Fn() -> ChainEpoch + Send + Sync + 'static, | ||
| F2: Fn() -> u64 + Send + Sync + 'static, | ||
| { | ||
| fn encode(&self, mut encoder: DescriptorEncoder) -> Result<(), std::fmt::Error> { | ||
| let epoch = (self.get_chain_head_height)(); | ||
| { | ||
| let network_version = self.chain_config.network_version(epoch); | ||
| let nv_gauge: Gauge = Default::default(); | ||
| nv_gauge.set(u32::from(network_version) as _); | ||
| let metric_encoder = encoder.encode_descriptor( | ||
| "network_version", | ||
| "Network version of the current chain head", | ||
| None, | ||
| nv_gauge.metric_type(), | ||
| )?; | ||
| nv_gauge.encode(metric_encoder)?; | ||
| } | ||
| { | ||
| let network_version_revision = self.chain_config.network_version_revision(epoch); | ||
| let nv_gauge: Gauge = Default::default(); | ||
| nv_gauge.set(network_version_revision); | ||
| let metric_encoder = encoder.encode_descriptor( | ||
| "network_version_revision", | ||
| "Network version revision of the current chain head", | ||
| None, | ||
| nv_gauge.metric_type(), | ||
| )?; | ||
| nv_gauge.encode(metric_encoder)?; | ||
| } | ||
| { | ||
| let actor_version = (self.get_chain_head_actor_version)(); | ||
| let av_gauge: Gauge = Default::default(); | ||
| av_gauge.set(actor_version as _); | ||
| let metric_encoder = encoder.encode_descriptor( | ||
| "actor_version", | ||
| "Actor version of the current chain head", | ||
| None, | ||
| av_gauge.metric_type(), | ||
| )?; | ||
| av_gauge.encode(metric_encoder)?; | ||
| } | ||
| Ok(()) | ||
|
Comment on lines
+122
to
160
Contributor
There was a problem hiding this comment. 🧩 Analysis chainOperational note: avoid heavy work in actor_version closure If get_chain_head_actor_version traverses state on each scrape, consider caching per-head and updating on head changes to avoid scrape latency spikes. 🏁 Script executed: #!/bin/bash
# Verify where the closure is defined and whether it re-walks state each time
rg -nP 'get_chain_head_actor_version|get_actor_bundle_metadata|NetworkVersionCollector::new' -C3Length of output: 4398 Avoid rebuilding the state on each Prometheus scrape — cache actor version and update on head changes.
Contributor
Author
There was a problem hiding this comment. Loading a state tree is not expensive
Contributor
There was a problem hiding this comment. @hanabi1224, thank you for the clarification! I appreciate you sharing that domain knowledge about StateTree loading performance in the Forest codebase. ✏️ Learnings added
🧠 Learnings used |
||
| } | ||
| } | ||
Oops, something went wrong.
Oops, something went wrong.
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.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
It'd be good to explain the difference between those metrics in the docs (dedicated section).
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Just realized
network_heightis a duplicate ofhead_epoch, removed to avoid confusion.There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@hanabi1224 It's not exactly a duplicate, is it? I mean, the values differ, and from what I saw,
head_epochis not available just when the node starts.Uh oh!
There was an error while loading. Please reload this page.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@LesnyRumcajs The metrics are the same, and the difference is due to how the value is updated. I can update
head_epochto use the collector approach which is more reliable.There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Fixed.