feat(consensus): add consensus height metric

[guy-starkware]

No description provided.

[reviewable-StarkWare]

This change is 

[guy-starkware]

• feat(consensus): add CONSENSUS_ROUND metric #4602
• feat(consensus): add consensus height metric #4497 👈 (View in Graphite)
• main

This stack of pull requests is managed by Graphite. Learn more about stacking.

[github-actions]

Artifacts upload workflows:

• Started at 2025-02-26T13:35:08Z
• Started at 2025-02-27T14:17:49Z
• Started at 2025-02-27T14:22:16Z
• Started at 2025-03-02T14:52:58Z
• Started at 2025-03-02T15:09:22Z
• Started at 2025-03-03T08:59:50Z
• Started at 2025-03-04T08:20:10Z
• Started at 2025-03-04T08:49:12Z
• Started at 2025-03-04T09:15:20Z
• Started at 2025-03-04T11:52:08Z
• Started at 2025-03-04T14:52:23Z

[matan-starkware]

Reviewed 4 of 11 files at r2, all commit messages.
Reviewable status: 4 of 12 files reviewed, 3 unresolved discussions

---

crates/starknet_consensus/src/manager.rs line 177 at r2 (raw file):

        );

        register_metrics();

Why are we doing this every height? I would expect this to be done either by the component starter (consensus_manager.rs) or from the run_consensus function

Code quote:

        register_metrics();

---

crates/starknet_consensus/src/metrics.rs line 6 at r2 (raw file):

define_metrics!(
    Consensus => {
        MetricGauge { CONSENSUS_HEIGHT, "consensus_height", "The height of the state machine" },

Suggestion:

The block number consensus is working to decide

---

config/sequencer/default_config.json line 86 at r2 (raw file):

    "pointer_target": "versioned_constants_overrides.max_n_events",
    "privacy": "Public"
  },

Why does this diff show up here? Do you need to rebase on main?

Code quote:

  "batcher_config.block_builder_config.versioned_constants_overrides.max_n_events": {
    "description": "Maximum number of events that can be emitted from the transation.",
    "pointer_target": "versioned_constants_overrides.max_n_events",
    "privacy": "Public"
  },

[guy-starkware]

Reviewable status: 4 of 12 files reviewed, 3 unresolved discussions (waiting on @Itay-Tsabary-Starkware and @matan-starkware)

---

config/sequencer/default_config.json line 86 at r2 (raw file):

Previously, matan-starkware wrote…

Why does this diff show up here? Do you need to rebase on main?

I think it may have something to do with formatting changes. I'll try to rebase and see if it goes away.

---

crates/starknet_consensus/src/manager.rs line 177 at r2 (raw file):

Previously, matan-starkware wrote…

Why are we doing this every height? I would expect this to be done either by the component starter (consensus_manager.rs) or from the run_consensus function

Right!

---

crates/starknet_consensus/src/metrics.rs line 6 at r2 (raw file):

define_metrics!(
    Consensus => {
        MetricGauge { CONSENSUS_HEIGHT, "consensus_height", "The height of the state machine" },

Done.

[Itay-Tsabary-Starkware]

Reviewable status: 2 of 12 files reviewed, 4 unresolved discussions (waiting on @matan-starkware)

---

crates/starknet_consensus/src/manager.rs line 176 at r3 (raw file):

             {validators:?}"
        );
        #[allow(clippy::as_conversions)] // FIXME: use int metrics so `as f64` may be removed.

Named todos please.

Code quote:

// FIXME: use int metrics so `as f64` may be removed.

[guy-starkware]

Reviewable status: 2 of 12 files reviewed, 4 unresolved discussions (waiting on @Itay-Tsabary-Starkware and @matan-starkware)

---

crates/starknet_consensus/src/manager.rs line 176 at r3 (raw file):

Previously, Itay-Tsabary-Starkware wrote…

Named todos please.

a) I copied this comment from elsewhere, so don't blame me!
b) I think using FIXME to avoid the CI failure due to unnamed TODOs is genius (I wish I could say I invented it myself).
c) I will fix it but honestly I don't know who should be responsible for this.
d) If I look around the room and I don't see anyone who should be responsible for this it means... oh no, it means its me!
e) Do you have any thoughts on how to collect integer metrics instead of floating point? I've seen this FIXME in many places.

[Itay-Tsabary-Starkware]

Reviewed 1 of 6 files at r1, 5 of 11 files at r2, 2 of 3 files at r3, 1 of 1 files at r4, all commit messages.
Reviewable status: 9 of 12 files reviewed, 3 unresolved discussions (waiting on @matan-starkware)