[Merged by Bors] - Improve debugging experience for builder proposals #3725

paulhauner · 2022-11-15T01:43:16Z

Issue Addressed

NA

Proposed Changes

This PR sets out to improve the logging/metrics experience when interacting with the builder. Namely, it:

Adds/changes metrics (see "Metrics Changes" section).
Adds new logs which show the duration of requests to the builder/local EL.
Refactors existing logs for consistency and so that the parent_hash is include in all relevant logs (we can grep for this field when trying to trace the flow of block production).

Additionally, when I was implementing this PR I noticed that we skip some verification of the builder payload in the scenario where the builder return Ok but the local EL returns with Err. Namely, we were skipping the bid signature and other values like parent hash and prev randao. In this PR I've changed it so we always check these values and reject the bid if they're incorrect. With these changes, we'll sometimes choose to skip a proposal rather than propose something invalid -- that's the only side-effect to the changes that I can see.

Metrics Changes

Changed: execution_layer_request_times:
- method = "get_blinded_payload_local": time taken to get a payload from a local EE.
- method = "get_blinded_payload_builder": time taken to get a blinded payload from a builder.
- method = "post_blinded_payload_builder": time taken to get a builder to reveal a payload they've previously supplied us.
execution_layer_get_payload_outcome
- outcome = "success": we successfully produced a payload from a builder or local EE.
- outcome = "failure": we were unable to get a payload from a builder or local EE.
New: execution_layer_builder_reveal_payload_outcome
- outcome = "success": a builder revealed a payload from a signed, blinded block.
- outcome = "failure": the builder did not reveal the payload.
New: execution_layer_get_payload_source
- type = "builder": we used a payload from a builder to produce a block.
- type = "local": we used a payload from a local EE to produce a block.
New: execution_layer_get_payload_builder_rejections has a reason field to describe why we rejected a payload from a builder.
New: execution_layer_payload_bids tracks the bid (in gwei) from the builder or local EE (local EE not yet supported, waiting on EEs to expose the value). Can only record values that fit inside an i64 (roughly 9 million ETH).

Additional Info

NA

paulhauner · 2022-11-15T06:03:30Z

This might be a good one for @realbigsean, since he's familiar with this code? I'm keen to hear your thoughts on the extra checks for builder-only proposals 🙏

michaelsproul

Looks great on the whole, will be very useful for trying out some new relays ;)

beacon_node/execution_layer/src/lib.rs

michaelsproul · 2022-11-25T04:16:41Z

beacon_node/execution_layer/src/lib.rs

+                                "relay_block_hash" => ?header.block_hash(),
+                                "local_block_hash" => %local.block_hash(),


Is the long-form for the relay block hash here intentional, as it's most likely to be selected?

Ah yeah, I actually intended to make both of these long-form. I figured that long-form would make it easier for us to search third-party services. I find it a little frustrating when we have short-form and then need to cross-reference with something like beaconcha.in.

I've gone and changed these in 26118a4. I ended up changing all roots/hashes in this file to be Display, hopefully this is not perceived as over-zealous.

I'm onboard with the long hashes! Agree 100% about the annoyances of short hashes.

I think you got the impls switched in 26118a4 though. The Debug implementation is the long-form one

beacon_node/execution_layer/src/lib.rs

michaelsproul · 2022-11-28T03:33:20Z

Related features that might be cool to add to this PR (or a follow-up):

Count metrics for the number of builder proposals vs local proposals (vs misses?)
Counts for each reason that a builder proposal is not made (an IntCounterVec).

Co-authored-by: Michael Sproul <micsproul@gmail.com>

paulhauner · 2022-11-28T08:08:38Z

Related features that might be cool to add to this PR (or a follow-up):

I added a bunch of new metrics in 339698b and updated the description to describe them. Good idea.

EDIT: a fixed a bug in those metrics in 55e0b9e.

EDIT 2: In 54390be I swapped to tracking the bid in gwei, since we'd be stuck at ~9 ETH when trying to fit wei inside an i64.

michaelsproul

Looking good!

Do we have this running on any test infra yet?

beacon_node/execution_layer/src/lib.rs

Co-authored-by: Michael Sproul <micsproul@gmail.com>

michaelsproul

Approved!

beacon_node/execution_layer/src/lib.rs

paulhauner · 2022-11-29T05:22:21Z

I fixed a sneaky bug here: c6256e1

paulhauner · 2022-11-29T05:51:25Z

Alrighty, I'm running this on some Prater nodes to good result! Thanks for the review ☺️

bors r+

## Issue Addressed NA ## Proposed Changes This PR sets out to improve the logging/metrics experience when interacting with the builder. Namely, it: - Adds/changes metrics (see "Metrics Changes" section). - Adds new logs which show the duration of requests to the builder/local EL. - Refactors existing logs for consistency and so that the `parent_hash` is include in all relevant logs (we can grep for this field when trying to trace the flow of block production). Additionally, when I was implementing this PR I noticed that we skip some verification of the builder payload in the scenario where the builder return `Ok` but the local EL returns with `Err`. Namely, we were skipping the bid signature and other values like parent hash and prev randao. In this PR I've changed it so we *always* check these values and reject the bid if they're incorrect. With these changes, we'll sometimes choose to skip a proposal rather than propose something invalid -- that's the only side-effect to the changes that I can see. ## Metrics Changes - Changed: `execution_layer_request_times`: - `method = "get_blinded_payload_local"`: time taken to get a payload from a local EE. - `method = "get_blinded_payload_builder"`: time taken to get a blinded payload from a builder. - `method = "post_blinded_payload_builder"`: time taken to get a builder to reveal a payload they've previously supplied us. - `execution_layer_get_payload_outcome` - `outcome = "success"`: we successfully produced a payload from a builder or local EE. - `outcome = "failure"`: we were unable to get a payload from a builder or local EE. - New: `execution_layer_builder_reveal_payload_outcome` - `outcome = "success"`: a builder revealed a payload from a signed, blinded block. - `outcome = "failure"`: the builder did not reveal the payload. - New: `execution_layer_get_payload_source` - `type = "builder"`: we used a payload from a builder to produce a block. - `type = "local"`: we used a payload from a local EE to produce a block. - New: `execution_layer_get_payload_builder_rejections` has a `reason` field to describe why we rejected a payload from a builder. - New: `execution_layer_payload_bids` tracks the bid (in gwei) from the builder or local EE (local EE not yet supported, waiting on EEs to expose the value). Can only record values that fit inside an i64 (roughly 9 million ETH). ## Additional Info NA

bors · 2022-11-29T08:19:19Z

paulhauner · 2022-12-02T22:15:54Z

@michaelsproul I just got woken up in the middle of the night because a block was not revealed by a builder but was actually broadcast 😅 Perhaps you're right about dropping that log level.

michaelsproul · 2022-12-02T22:38:20Z

Let's drop it to ERROR! That won't wake you up, right?

@michaelsproul

## Issue Addressed NA ## Proposed Changes In #3725 I introduced a `CRIT` log for unrevealed payloads, against @michaelsproul's [advice](#3725 (comment)). After being woken up in the middle of the night by a block that was not revealed to the BN but *was* revealed to the network, I have capitulated. This PR implements @michaelsproul's suggestion and reduces the severity to `ERRO`. Additionally, I have dropped a `CRIT` to an `ERRO` for when a block is published late. The block in question was indeed published late on the network, however now that we have builders that can slow down block production I don't think the error is "actionable" enough to warrant a `CRIT` for the user. ## Additional Info NA

## Issue Addressed NA ## Proposed Changes This PR sets out to improve the logging/metrics experience when interacting with the builder. Namely, it: - Adds/changes metrics (see "Metrics Changes" section). - Adds new logs which show the duration of requests to the builder/local EL. - Refactors existing logs for consistency and so that the `parent_hash` is include in all relevant logs (we can grep for this field when trying to trace the flow of block production). Additionally, when I was implementing this PR I noticed that we skip some verification of the builder payload in the scenario where the builder return `Ok` but the local EL returns with `Err`. Namely, we were skipping the bid signature and other values like parent hash and prev randao. In this PR I've changed it so we *always* check these values and reject the bid if they're incorrect. With these changes, we'll sometimes choose to skip a proposal rather than propose something invalid -- that's the only side-effect to the changes that I can see. ## Metrics Changes - Changed: `execution_layer_request_times`: - `method = "get_blinded_payload_local"`: time taken to get a payload from a local EE. - `method = "get_blinded_payload_builder"`: time taken to get a blinded payload from a builder. - `method = "post_blinded_payload_builder"`: time taken to get a builder to reveal a payload they've previously supplied us. - `execution_layer_get_payload_outcome` - `outcome = "success"`: we successfully produced a payload from a builder or local EE. - `outcome = "failure"`: we were unable to get a payload from a builder or local EE. - New: `execution_layer_builder_reveal_payload_outcome` - `outcome = "success"`: a builder revealed a payload from a signed, blinded block. - `outcome = "failure"`: the builder did not reveal the payload. - New: `execution_layer_get_payload_source` - `type = "builder"`: we used a payload from a builder to produce a block. - `type = "local"`: we used a payload from a local EE to produce a block. - New: `execution_layer_get_payload_builder_rejections` has a `reason` field to describe why we rejected a payload from a builder. - New: `execution_layer_payload_bids` tracks the bid (in gwei) from the builder or local EE (local EE not yet supported, waiting on EEs to expose the value). Can only record values that fit inside an i64 (roughly 9 million ETH). ## Additional Info NA

@michaelsproul

## Issue Addressed NA ## Proposed Changes In sigp#3725 I introduced a `CRIT` log for unrevealed payloads, against @michaelsproul's [advice](sigp#3725 (comment)). After being woken up in the middle of the night by a block that was not revealed to the BN but *was* revealed to the network, I have capitulated. This PR implements @michaelsproul's suggestion and reduces the severity to `ERRO`. Additionally, I have dropped a `CRIT` to an `ERRO` for when a block is published late. The block in question was indeed published late on the network, however now that we have builders that can slow down block production I don't think the error is "actionable" enough to warrant a `CRIT` for the user. ## Additional Info NA

## Issue Addressed NA ## Proposed Changes This PR sets out to improve the logging/metrics experience when interacting with the builder. Namely, it: - Adds/changes metrics (see "Metrics Changes" section). - Adds new logs which show the duration of requests to the builder/local EL. - Refactors existing logs for consistency and so that the `parent_hash` is include in all relevant logs (we can grep for this field when trying to trace the flow of block production). Additionally, when I was implementing this PR I noticed that we skip some verification of the builder payload in the scenario where the builder return `Ok` but the local EL returns with `Err`. Namely, we were skipping the bid signature and other values like parent hash and prev randao. In this PR I've changed it so we *always* check these values and reject the bid if they're incorrect. With these changes, we'll sometimes choose to skip a proposal rather than propose something invalid -- that's the only side-effect to the changes that I can see. ## Metrics Changes - Changed: `execution_layer_request_times`: - `method = "get_blinded_payload_local"`: time taken to get a payload from a local EE. - `method = "get_blinded_payload_builder"`: time taken to get a blinded payload from a builder. - `method = "post_blinded_payload_builder"`: time taken to get a builder to reveal a payload they've previously supplied us. - `execution_layer_get_payload_outcome` - `outcome = "success"`: we successfully produced a payload from a builder or local EE. - `outcome = "failure"`: we were unable to get a payload from a builder or local EE. - New: `execution_layer_builder_reveal_payload_outcome` - `outcome = "success"`: a builder revealed a payload from a signed, blinded block. - `outcome = "failure"`: the builder did not reveal the payload. - New: `execution_layer_get_payload_source` - `type = "builder"`: we used a payload from a builder to produce a block. - `type = "local"`: we used a payload from a local EE to produce a block. - New: `execution_layer_get_payload_builder_rejections` has a `reason` field to describe why we rejected a payload from a builder. - New: `execution_layer_payload_bids` tracks the bid (in gwei) from the builder or local EE (local EE not yet supported, waiting on EEs to expose the value). Can only record values that fit inside an i64 (roughly 9 million ETH). ## Additional Info NA

Squashed commit of the following: commit ad08d07 Author: Paul Hauner <paul@paulhauner.com> Date: Mon Dec 5 16:51:06 2022 +1100 Remove crits for late block commit 8e85d62 Author: Paul Hauner <paul@paulhauner.com> Date: Mon Dec 5 16:48:43 2022 +1100 Downgrade log for payload reveal failure commit 84392d6 Author: Michael Sproul <michael@sigmaprime.io> Date: Fri Dec 2 00:07:43 2022 +0000 Delete DB schema migrations for v11 and earlier (sigp#3761) ## Proposed Changes Now that the Gnosis merge is scheduled, all users should have upgraded beyond Lighthouse v3.0.0. Accordingly we can delete schema migrations for versions prior to v3.0.0. ## Additional Info I also deleted the state cache stuff I added in sigp#3714 as it turned out to be useless for the light client proofs due to the one-slot offset. commit 18c9be5 Author: Mac L <mjladson@pm.me> Date: Thu Dec 1 06:03:53 2022 +0000 Add API endpoint to count statuses of all validators (sigp#3756) ## Issue Addressed sigp#3724 ## Proposed Changes Adds an endpoint to quickly count the number of occurances of each status in the validator set. ## Usage ```bash curl -X GET "http://localhost:5052/lighthouse/ui/validator_count" -H "accept: application/json" | jq ``` ```json { "data": { "active_ongoing":479508, "active_exiting":0, "active_slashed":0, "pending_initialized":28, "pending_queued":0, "withdrawal_possible":933, "withdrawal_done":0, "exited_unslashed":0, "exited_slashed":3 } } ``` commit 2211504 Author: Michael Sproul <michael@sigmaprime.io> Date: Wed Nov 30 05:22:58 2022 +0000 Prioritise important parts of block processing (sigp#3696) ## Issue Addressed Closes sigp#2327 ## Proposed Changes This is an extension of some ideas I implemented while working on `tree-states`: - Cache the indexed attestations from blocks in the `ConsensusContext`. Previously we were re-computing them 3-4 times over. - Clean up `import_block` by splitting each part into `import_block_XXX`. - Move some stuff off hot paths, specifically: - Relocate non-essential tasks that were running between receiving the payload verification status and priming the early attester cache. These tasks are moved after the cache priming: - Attestation observation - Validator monitor updates - Slasher updates - Updating the shuffling cache - Fork choice attestation observation now happens at the end of block verification in parallel with payload verification (this seems to save 5-10ms). - Payload verification now happens _before_ advancing the pre-state and writing it to disk! States were previously being written eagerly and adding ~20-30ms in front of verifying the execution payload. State catchup also sometimes takes ~500ms if we get a cache miss and need to rebuild the tree hash cache. The remaining task that's taking substantial time (~20ms) is importing the block to fork choice. I _think_ this is because of pull-tips, and we should be able to optimise it out with a clever total active balance cache in the state (which would be computed in parallel with payload verification). I've decided to leave that for future work though. For now it can be observed via the new `beacon_block_processing_post_exec_pre_attestable_seconds` metric. Co-authored-by: Michael Sproul <micsproul@gmail.com> commit b4f4c0d Author: Divma <divma@protonmail.com> Date: Wed Nov 30 03:21:35 2022 +0000 Ipv6 bootnodes (sigp#3752) ## Issue Addressed our bootnodes as of now support only ipv4. this makes it so that they support ipv6 ## Proposed Changes - Adds code necessary to update the bootnodes to run on dual stack nodes and therefore contact and store ipv6 nodes. - Adds some metrics about connectivity type of stored peers. It might have been nice to see some metrics over the sessions but that feels out of scope right now. ## Additional Info - some code quality improvements sneaked in since the changes seemed small - I think it depends on the OS, but enabling mapped addresses on an ipv6 node without dual stack support enabled could fail silently, making these nodes effectively ipv6 only. In the future I'll probably change this to use two sockets, which should fail loudly commit 3534c85 Author: GeemoCandama <geemo@tutanota.com> Date: Tue Nov 29 08:19:27 2022 +0000 Optimize finalized chain sync by skipping newPayload messages (sigp#3738) ## Issue Addressed sigp#3704 ## Proposed Changes Adds is_syncing_finalized: bool parameter for block verification functions. Sets the payload_verification_status to Optimistic if is_syncing_finalized is true. Uses SyncState in NetworkGlobals in BeaconProcessor to retrieve the syncing status. ## Additional Info I could implement FinalizedSignatureVerifiedBlock if you think it would be nicer. commit a2969ba Author: Paul Hauner <paul@paulhauner.com> Date: Tue Nov 29 05:51:42 2022 +0000 Improve debugging experience for builder proposals (sigp#3725) ## Issue Addressed NA ## Proposed Changes This PR sets out to improve the logging/metrics experience when interacting with the builder. Namely, it: - Adds/changes metrics (see "Metrics Changes" section). - Adds new logs which show the duration of requests to the builder/local EL. - Refactors existing logs for consistency and so that the `parent_hash` is include in all relevant logs (we can grep for this field when trying to trace the flow of block production). Additionally, when I was implementing this PR I noticed that we skip some verification of the builder payload in the scenario where the builder return `Ok` but the local EL returns with `Err`. Namely, we were skipping the bid signature and other values like parent hash and prev randao. In this PR I've changed it so we *always* check these values and reject the bid if they're incorrect. With these changes, we'll sometimes choose to skip a proposal rather than propose something invalid -- that's the only side-effect to the changes that I can see. ## Metrics Changes - Changed: `execution_layer_request_times`: - `method = "get_blinded_payload_local"`: time taken to get a payload from a local EE. - `method = "get_blinded_payload_builder"`: time taken to get a blinded payload from a builder. - `method = "post_blinded_payload_builder"`: time taken to get a builder to reveal a payload they've previously supplied us. - `execution_layer_get_payload_outcome` - `outcome = "success"`: we successfully produced a payload from a builder or local EE. - `outcome = "failure"`: we were unable to get a payload from a builder or local EE. - New: `execution_layer_builder_reveal_payload_outcome` - `outcome = "success"`: a builder revealed a payload from a signed, blinded block. - `outcome = "failure"`: the builder did not reveal the payload. - New: `execution_layer_get_payload_source` - `type = "builder"`: we used a payload from a builder to produce a block. - `type = "local"`: we used a payload from a local EE to produce a block. - New: `execution_layer_get_payload_builder_rejections` has a `reason` field to describe why we rejected a payload from a builder. - New: `execution_layer_payload_bids` tracks the bid (in gwei) from the builder or local EE (local EE not yet supported, waiting on EEs to expose the value). Can only record values that fit inside an i64 (roughly 9 million ETH). ## Additional Info NA commit 99ec9d9 Author: kevinbogner <kevbogner@gmail.com> Date: Mon Nov 28 10:05:43 2022 +0000 Add Run a Node guide (sigp#3681) ## Issue Addressed Related to sigp#3672 ## Proposed Changes - Added a guide to run a node. Mainly, copy and paste from 'Merge Migration' and 'Checkpoint Sync'. - Ranked it high in ToC: - Introduction - Installation - Run a Node - Become a Validator ... - Hid 'Merge Migration' in ToC. ## Additional Info - Should I add/rephrase/delete something? - Now there is some redundancy: - 'Run a node' and 'Checkpoint Sync' contain similar information. - Same for 'Run a node' and 'Become a Validator'. Co-authored-by: kevinbogner <114221396+kevinbogner@users.noreply.github.com> Co-authored-by: Michael Sproul <micsproul@gmail.com> commit 2779017 Author: Age Manning <Age@AgeManning.com> Date: Mon Nov 28 07:36:52 2022 +0000 Gossipsub fast message id change (sigp#3755) For improved consistency, this mixes in the topic into our fast message id for more consistent tracking of messages across topics. commit c881b80 Author: Mac L <mjladson@pm.me> Date: Mon Nov 28 00:22:53 2022 +0000 Add CLI flag for gui requirements (sigp#3731) ## Issue Addressed sigp#3723 ## Proposed Changes Adds a new CLI flag `--gui` which enables all the various flags required for the gui to function properly. Currently enables the `--http` and `--validator-monitor-auto` flags. commit 969ff24 Author: Mac L <mjladson@pm.me> Date: Fri Nov 25 07:57:11 2022 +0000 Add CLI flag to opt in to world-readable log files (sigp#3747) ## Issue Addressed sigp#3732 ## Proposed Changes Add a CLI flag to allow users to opt out of the restrictive permissions of the log files. ## Additional Info This is not recommended for most users. The log files can contain sensitive information such as validator indices, public keys and API tokens (see sigp#2438). However some users using a multi-user setup may find this helpful if they understand the risks involved. commit e9bf7f7 Author: antondlr <anton@delaruelle.net> Date: Fri Nov 25 07:57:10 2022 +0000 remove commas from comma-separated kv pairs (sigp#3737) ## Issue Addressed Logs are in comma separated kv list, but the values sometimes contain commas, which breaks parsing commit d5a2de7 Author: Giulio rebuffo <giulio.rebuffo@gmail.com> Date: Fri Nov 25 05:19:00 2022 +0000 Added LightClientBootstrap V1 (sigp#3711) ## Issue Addressed Partially addresses sigp#3651 ## Proposed Changes Adds server-side support for light_client_bootstrap_v1 topic ## Additional Info This PR, creates each time a bootstrap without using cache, I do not know how necessary a cache is in this case as this topic is not supposed to be called frequently and IMHO we can just prevent abuse by using the limiter, but let me know what you think or if there is any caveat to this, or if it is necessary only for the sake of good practice. Co-authored-by: Pawan Dhananjay <pawandhananjay@gmail.com>

NA This PR sets out to improve the logging/metrics experience when interacting with the builder. Namely, it: - Adds/changes metrics (see "Metrics Changes" section). - Adds new logs which show the duration of requests to the builder/local EL. - Refactors existing logs for consistency and so that the `parent_hash` is include in all relevant logs (we can grep for this field when trying to trace the flow of block production). Additionally, when I was implementing this PR I noticed that we skip some verification of the builder payload in the scenario where the builder return `Ok` but the local EL returns with `Err`. Namely, we were skipping the bid signature and other values like parent hash and prev randao. In this PR I've changed it so we *always* check these values and reject the bid if they're incorrect. With these changes, we'll sometimes choose to skip a proposal rather than propose something invalid -- that's the only side-effect to the changes that I can see. - Changed: `execution_layer_request_times`: - `method = "get_blinded_payload_local"`: time taken to get a payload from a local EE. - `method = "get_blinded_payload_builder"`: time taken to get a blinded payload from a builder. - `method = "post_blinded_payload_builder"`: time taken to get a builder to reveal a payload they've previously supplied us. - `execution_layer_get_payload_outcome` - `outcome = "success"`: we successfully produced a payload from a builder or local EE. - `outcome = "failure"`: we were unable to get a payload from a builder or local EE. - New: `execution_layer_builder_reveal_payload_outcome` - `outcome = "success"`: a builder revealed a payload from a signed, blinded block. - `outcome = "failure"`: the builder did not reveal the payload. - New: `execution_layer_get_payload_source` - `type = "builder"`: we used a payload from a builder to produce a block. - `type = "local"`: we used a payload from a local EE to produce a block. - New: `execution_layer_get_payload_builder_rejections` has a `reason` field to describe why we rejected a payload from a builder. - New: `execution_layer_payload_bids` tracks the bid (in gwei) from the builder or local EE (local EE not yet supported, waiting on EEs to expose the value). Can only record values that fit inside an i64 (roughly 9 million ETH). NA

@michaelsproul

NA In sigp#3725 I introduced a `CRIT` log for unrevealed payloads, against @michaelsproul's [advice](sigp#3725 (comment)). After being woken up in the middle of the night by a block that was not revealed to the BN but *was* revealed to the network, I have capitulated. This PR implements @michaelsproul's suggestion and reduces the severity to `ERRO`. Additionally, I have dropped a `CRIT` to an `ERRO` for when a block is published late. The block in question was indeed published late on the network, however now that we have builders that can slow down block production I don't think the error is "actionable" enough to warrant a `CRIT` for the user. NA

paulhauner added 2 commits November 15, 2022 12:16

Refactor get_blinded_payload

6f5455b

Add metrics and logs for payload unblinding

dd97a89

paulhauner added the work-in-progress PR is a work-in-progress label Nov 15, 2022

paulhauner added 2 commits November 15, 2022 14:20

Run cargo fmt

80f6a1e

Tidy, add buckets

34eab88

paulhauner added ready-for-review The code is ready for review and removed work-in-progress PR is a work-in-progress labels Nov 15, 2022

paulhauner marked this pull request as ready for review November 15, 2022 06:02

realbigsean self-requested a review November 16, 2022 02:57

paulhauner added the v3.4.0 Minor release following v3.3.0 label Nov 23, 2022

michaelsproul reviewed Nov 25, 2022

View reviewed changes

michaelsproul added waiting-on-author The reviewer has suggested changes and awaits thier implementation. and removed ready-for-review The code is ready for review labels Nov 25, 2022

paulhauner and others added 4 commits November 28, 2022 14:37

Apply suggestions from code review

71c670b

Co-authored-by: Michael Sproul <micsproul@gmail.com>

Print full hash/root in several places

26118a4

Handle low value payload when there's no alternative

0a86b24

Add more metrics

339698b

Fix metric tracking for get_payload

55e0b9e

paulhauner added ready-for-review The code is ready for review and removed waiting-on-author The reviewer has suggested changes and awaits thier implementation. labels Nov 28, 2022

Log bid in gwei

54390be

michaelsproul reviewed Nov 28, 2022

View reviewed changes

michaelsproul added waiting-on-author The reviewer has suggested changes and awaits thier implementation. and removed ready-for-review The code is ready for review labels Nov 28, 2022

paulhauner and others added 3 commits November 29, 2022 11:29

Update beacon_node/execution_layer/src/lib.rs

c9b7d00

Co-authored-by: Michael Sproul <micsproul@gmail.com>

Avoid warning with low-value builder payload

e0b2b1a

Move roots/hashes back to debug

e8bd2be

Merge branch 'unstable' into builder-metrics

57e0e11

michaelsproul approved these changes Nov 29, 2022

View reviewed changes

beacon_node/execution_layer/src/lib.rs Show resolved Hide resolved

beacon_node/execution_layer/src/lib.rs Show resolved Hide resolved

michaelsproul added ready-for-merge This PR is ready to merge. and removed waiting-on-author The reviewer has suggested changes and awaits thier implementation. labels Nov 29, 2022

Fix bug with gwei calculation

c6256e1

bors bot changed the title ~~Improve debugging experience for builder proposals~~ [Merged by Bors] - Improve debugging experience for builder proposals Nov 29, 2022

bors bot closed this Nov 29, 2022

paulhauner mentioned this pull request Dec 5, 2022

[Merged by Bors] - Reduce log severity for late and unrevealed blocks #3775

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[Merged by Bors] - Improve debugging experience for builder proposals #3725

[Merged by Bors] - Improve debugging experience for builder proposals #3725

paulhauner commented Nov 15, 2022 •

edited

Loading

paulhauner commented Nov 15, 2022

michaelsproul left a comment

michaelsproul Nov 25, 2022

paulhauner Nov 28, 2022

michaelsproul Nov 28, 2022

michaelsproul commented Nov 28, 2022

paulhauner commented Nov 28, 2022 •

edited

Loading

michaelsproul left a comment

michaelsproul left a comment

paulhauner commented Nov 29, 2022

paulhauner commented Nov 29, 2022

bors bot commented Nov 29, 2022

paulhauner commented Dec 2, 2022

michaelsproul commented Dec 2, 2022

		"relay_block_hash" => ?header.block_hash(),
		"local_block_hash" => %local.block_hash(),

[Merged by Bors] - Improve debugging experience for builder proposals #3725

[Merged by Bors] - Improve debugging experience for builder proposals #3725

Conversation

paulhauner commented Nov 15, 2022 • edited Loading

Issue Addressed

Proposed Changes

Metrics Changes

Additional Info

paulhauner commented Nov 15, 2022

michaelsproul left a comment

Choose a reason for hiding this comment

michaelsproul Nov 25, 2022

Choose a reason for hiding this comment

paulhauner Nov 28, 2022

Choose a reason for hiding this comment

michaelsproul Nov 28, 2022

Choose a reason for hiding this comment

michaelsproul commented Nov 28, 2022

paulhauner commented Nov 28, 2022 • edited Loading

michaelsproul left a comment

Choose a reason for hiding this comment

michaelsproul left a comment

Choose a reason for hiding this comment

paulhauner commented Nov 29, 2022

paulhauner commented Nov 29, 2022

bors bot commented Nov 29, 2022

paulhauner commented Dec 2, 2022

michaelsproul commented Dec 2, 2022

paulhauner commented Nov 15, 2022 •

edited

Loading

paulhauner commented Nov 28, 2022 •

edited

Loading