## Issue Addressed
Fixes#1864 and a bunch of other closed but unresolved issues.
## Proposed Changes
Allows the deposit caching to recover from `NonConsecutive` deposit errors by resetting the last processed block to the last valid deposit's block number. Still not sure of the underlying cause of this error, but this should recover the cache so we don't need `--eth1-purge-cache` anymore 🎉
A huge thanks to @one-three-three-seven for reproducing the error and providing the data that helped testing out the fix 🙌
Still needs a few more tests.
## Proposed Changes
Expand the set of paths tracked by the HTTP API metrics to include all paths hit by the validator client.
These paths were only partially updated for Altair, so we were missing some of the sync committee and v2 APIs.
## Issue Addressed
NA
## Proposed Changes
I used these logs when debugging a spurious failure with Infura and thought they might be nice to have around permanently.
There's no changes to functionality in this PR, just some additional `debug!` logs.
## Additional Info
NA
## Issue Addressed
Deprecates the step parameter in the blocks by range request
## Proposed Changes
- Modifies the BlocksByRangeRequest type to remove the step parameter and everywhere we took it into account before
- Adds a new type to still handle coding and decoding of requests that use the parameter
## Additional Info
I went with a deprecation over the type itself so that requests received outside `lighthouse_network` don't even need to deal with this parameter. After the deprecation period just removing the Old blocks by range request should be straightforward
## Issue Addressed
Following up from https://github.com/sigp/lighthouse/pull/3223#issuecomment-1158718102, it has been observed that the validator client uses vastly more memory in some compilation configurations than others. Compiling with Cross and then putting the binary into an Ubuntu 22.04 image seems to use 3x more memory than compiling with Cargo directly on Debian bullseye.
## Proposed Changes
Enable malloc metrics for the validator client. This will hopefully allow us to see the difference between the two compilation configs and compare heap fragmentation. This PR doesn't enable malloc tuning for the VC because it was found to perform significantly worse. The `--disable-malloc-tuning` flag is repurposed to just disable the metrics.
## Issue Addressed
Closes https://github.com/sigp/lighthouse/issues/2944
## Proposed Changes
Remove snapshots from the cache during sync rather than cloning them. This reduces unnecessary cloning and memory fragmentation during sync.
## Additional Info
This PR relies on the fact that the `block_delay` cache is not populated for blocks from sync. Relying on block delay may have the side effect that a change in `block_delay` calculation could lead to: a) more clones, if block delays are added for syncing blocks or b) less clones, if blocks near the head are erroneously provided without a `block_delay`. Case (a) would be a regression to the current status quo, and (b) is low-risk given we know that the snapshot cache is current susceptible to misses (hence `tree-states`).
## Issue Addressed
#2820
## Proposed Changes
The problem is that validator_monitor_prev_epoch metrics are updated only if there is EpochSummary present in summaries map for the previous epoch and it is not the case for the offline validator. Ensure that EpochSummary is inserted into summaries map also for the offline validators.
## Issue Addressed
Partly resolves https://github.com/sigp/lighthouse/issues/3032
## Proposed Changes
Extracts some of the functionality of #3094 into a separate PR as the original PR requires a bit more work.
Do not unnecessarily penalize peers when we fail to validate received execution payloads because our execution layer is offline.
## Issue Addressed
Which issue # does this PR address?
## Proposed Changes
Please list or describe the changes introduced by this PR.
## Additional Info
Please provide any additional information. For example, future considerations
or information useful for reviewers.
## Issue Addressed
NA
## Proposed Changes
- Uses a `Vec` in `SingleEpochParticipationCache` rather than `HashMap` to speed up processing times at the cost of memory usage.
- Cache the result of `integer_sqrt` rather than recomputing for each validator.
- Cache `state.previous_epoch` rather than recomputing it for each validator.
### Benchmarks
Benchmarks on a recent mainnet state using #3252 to get timing.
#### Without this PR
```
lcli skip-slots --state-path /tmp/state-0x3cdc.ssz --partial-state-advance --slots 32 --state-root 0x3cdc33cd02713d8d6cc33a6dbe2d3a5bf9af1d357de0d175a403496486ff845e --runs 10
[2022-06-09T08:21:02Z INFO lcli::skip_slots] Using mainnet spec
[2022-06-09T08:21:02Z INFO lcli::skip_slots] Advancing 32 slots
[2022-06-09T08:21:02Z INFO lcli::skip_slots] Doing 10 runs
[2022-06-09T08:21:02Z INFO lcli::skip_slots] State path: "/tmp/state-0x3cdc.ssz"
SSZ decoding /tmp/state-0x3cdc.ssz: 43ms
[2022-06-09T08:21:03Z INFO lcli::skip_slots] Run 0: 245.718794ms
[2022-06-09T08:21:03Z INFO lcli::skip_slots] Run 1: 245.364782ms
[2022-06-09T08:21:03Z INFO lcli::skip_slots] Run 2: 255.866179ms
[2022-06-09T08:21:04Z INFO lcli::skip_slots] Run 3: 243.838909ms
[2022-06-09T08:21:04Z INFO lcli::skip_slots] Run 4: 250.431425ms
[2022-06-09T08:21:04Z INFO lcli::skip_slots] Run 5: 248.68765ms
[2022-06-09T08:21:04Z INFO lcli::skip_slots] Run 6: 262.051113ms
[2022-06-09T08:21:05Z INFO lcli::skip_slots] Run 7: 264.293967ms
[2022-06-09T08:21:05Z INFO lcli::skip_slots] Run 8: 293.202007ms
[2022-06-09T08:21:05Z INFO lcli::skip_slots] Run 9: 264.552017ms
```
#### With this PR:
```
lcli skip-slots --state-path /tmp/state-0x3cdc.ssz --partial-state-advance --slots 32 --state-root 0x3cdc33cd02713d8d6cc33a6dbe2d3a5bf9af1d357de0d175a403496486ff845e --runs 10
[2022-06-09T08:57:59Z INFO lcli::skip_slots] Run 0: 73.898678ms
[2022-06-09T08:57:59Z INFO lcli::skip_slots] Run 1: 75.536978ms
[2022-06-09T08:57:59Z INFO lcli::skip_slots] Run 2: 75.176104ms
[2022-06-09T08:57:59Z INFO lcli::skip_slots] Run 3: 76.460828ms
[2022-06-09T08:57:59Z INFO lcli::skip_slots] Run 4: 75.904195ms
[2022-06-09T08:58:00Z INFO lcli::skip_slots] Run 5: 75.53077ms
[2022-06-09T08:58:00Z INFO lcli::skip_slots] Run 6: 74.745572ms
[2022-06-09T08:58:00Z INFO lcli::skip_slots] Run 7: 75.823489ms
[2022-06-09T08:58:00Z INFO lcli::skip_slots] Run 8: 74.892055ms
[2022-06-09T08:58:00Z INFO lcli::skip_slots] Run 9: 76.333569ms
```
## Additional Info
NA
## Description
Add a new lint to CI that attempts to detect calls to functions like `block_on` from async execution contexts. This lint was written from scratch exactly for this purpose, on my fork of Clippy: https://github.com/michaelsproul/rust-clippy/tree/disallow-from-async
## Additional Info
- I've successfully detected the previous two issues we had with `block_on` by running the linter on the commits prior to each of these PRs: https://github.com/sigp/lighthouse/pull/3165, https://github.com/sigp/lighthouse/pull/3199.
- The lint runs on CI with `continue-on-error: true` so that if it fails spuriously it won't block CI.
- I think it would be good to merge this PR before https://github.com/sigp/lighthouse/pull/3244 so that we can lint the extensive executor-related changes in that PR.
- I aim to upstream the lint to Clippy, at which point building a custom version of Clippy from my fork will no longer be necessary. I imagine this will take several weeks or months though, because the code is currently a bit hacky and will need some renovations to pass review.
## Issue Addressed
Newer versions of MDBX have removed Windows and macOS support, so this PR pins MDBX at the last working version to prevent an accidental regression via `cargo update`.
## Additional Info
This is a short-term solution, if our pinned version of MDBX turns out to be buggy we will need to consider backporting patches from upstream to our own fork.
## Issue Addressed
Failures in our CI integration tests for Geth.
## Proposed Changes
Only connect to the authenticated execution endpoints during execution tests.
This is necessary now that it is impossible to connect to the `engine` api on an unauthenticated endpoint.
See https://github.com/ethereum/go-ethereum/pull/24997
## Additional Info
As these tests break semi-regularly, I have kept logs enabled to ease future debugging.
I've also updated the Nethermind tests, although these weren't broken. This should future-proof us if Nethermind decides to follow suit with Geth
## Issue Addressed
currently we count a failed attempt for a syncing chain even if the peer is not at fault. This makes us do more work if the chain fails, and heavily penalize peers, when we can simply retry. Inspired by a proposal I made to #3094
## Proposed Changes
If a batch fails but the peer is not at fault, do not count the attempt
Also removes some annoying logs
## Additional Info
We still get a counter on ignored attempts.. just in case
## Issue Addressed
na
## Proposed Changes
Updates libp2p to https://github.com/libp2p/rust-libp2p/pull/2662
## Additional Info
From comments on the relevant PRs listed, we should pay attention at peer management consistency, but I don't think anything weird will happen.
This is running in prater tok and sin
## Issue Addressed
Fixes a timing issue that results in spurious fork choice notifier failures:
```
WARN Error signalling fork choice waiter slot: 3962270, error: ForkChoiceSignalOutOfOrder { current: Slot(3962271), latest: Slot(3962270) }, service: beacon
```
There’s a fork choice run that is scheduled to run at the start of every slot by the `timer`, which creates a 12s interval timer when the beacon node starts up. The problem is that if there’s a bit of clock drift that gets corrected via NTP (or a leap second for that matter) then these 12s intervals will cease to line up with the start of the slot. This then creates the mismatch in slot number that we see above.
Lighthouse also runs fork choice 500ms before the slot begins, and these runs are what is conflicting with the start-of-slot runs. This means that the warning in current versions of Lighthouse is mostly cosmetic because fork choice is up to date with all but the most recent 500ms of attestations (which usually isn’t many).
## Proposed Changes
Fix the per-slot timer so that it continually re-calculates the duration to the start of the next slot and waits for that.
A side-effect of this change is that we may skip slots if the per-slot task takes >12s to run, but I think this is an unlikely scenario and an acceptable compromise.
## Issue Addressed
Reduces the effect of late blocks on overall node buildup
## Proposed Changes
change the capacity of the channels used to send work for reprocessing in the beacon processor, and to send back to the main processor task, to be 75% of the capacity of the channel for receiving new events
## Additional Info
The issues we've seen suggest we should still evaluate node performance under stress, with late blocks being a big factor.
Other changes that could help:
1. right now we have a cap for queued attestations for reprocessing that applies to the sum of aggregated and unaggregated attestations. We could consider adding a separate cap that favors aggregated ones.
2. solving #2848
## Issue Addressed
Fix for the eth1 cache sync issue observed on Ropsten.
## Proposed Changes
Ropsten blocks are so infrequent that they broke our algorithm for downloading eth1 blocks. We currently try to download forwards from the last block in our cache to the block with block number [`remote_highest_block - FOLLOW_DISTANCE + FOLLOW_DISTANCE / ETH1_BLOCK_TIME_TOLERANCE_FACTOR`](6f732986f1/beacon_node/eth1/src/service.rs (L489-L492)). With the tolerance set to 4 this is insufficient because we lag by 1536 blocks, which is more like ~14 hours on Ropsten. This results in us having an incomplete eth1 cache, because we should cache all blocks between -16h and -8h. Even if we were to set the tolerance to 2 for the largest allowance, we would only look back 1024 blocks which is still more than 8 hours.
For example consider this block https://ropsten.etherscan.io/block/12321390. The block from 1536 blocks earlier is 14 hours and 20 minutes before it: https://ropsten.etherscan.io/block/12319854. The block from 1024 blocks earlier is https://ropsten.etherscan.io/block/12320366, 8 hours and 48 minutes before.
- This PR introduces a new CLI flag called `--eth1-cache-follow-distance` which can be used to set the distance manually.
- A new dynamic catchup mechanism is added which detects when the cache is lagging the true eth1 chain and tries to download more blocks within the follow distance in order to catch up.
## Issue Addressed
N/A
## Proposed Changes
Preemptively switch Nethermind integration tests to use the `master` branch along with the baked in `kiln` config.
## Additional Info
There have been some spurious timeouts across CI so this also increases the timeout to 20s.
## Issue Addressed
#3156
## Proposed Changes
Emit a `WARN` log whenever the value of `fee_recipient` as returned from the EE is different from the value of `suggested_fee_recipient` as set on the BN, for example by the `--suggested-fee-recipient` CLI flag.
## Additional Info
I have set the log level to `WARN` since it is legal behaviour (meaning it isn't really an error but is important to know when it is occurring).
If we feel like this behaviour is almost always undesired (caused by a misconfiguration or malicious EE) then an `ERRO` log would be more appropriate. Happy to change it in that case.
## Issue Addressed
N/A
## Proposed Changes
Use stable version of ubuntu base image in dockerfile instead of using latest. This will help in narrowing down issues with docker images.
## Proposed Changes
Speed up epoch processing by around 10% by inlining methods from the `safe_arith` crate.
The Rust standard library uses `#[inline]` for the `checked_` functions that we're wrapping, so it makes sense for us to inline them too.
## Additional Info
I conducted a brief statistical test on the block at slot [3858336](https://beaconcha.in/block/3858336) applied to the state at slot 3858335, which requires an epoch transition. The command used for testing was:
```
lcli transition-blocks --testnet-dir ./common/eth2_network_config/built_in_network_configs/mainnet --no-signature-verification state.ssz block.ssz output.ssz
```
The testing found that inlining reduced the epoch transition time from 398ms to 359ms, a reduction of 9.77%, which was found to be statistically significant with a two-tailed t-test (p < 0.01). Data and intermediate calculations can be found here: https://docs.google.com/spreadsheets/d/1tlf3eFjz3dcXeb9XVOn21953uYpc9RdQapPtcHGH1PY
## Issue Addressed
We were logging `out_finalized_epoch` instead of `our_finalized_epoch`. I noticed this ages ago but only just got around to fixing it.
## Additional Info
I also reformatted the log line to respect the line length limit (`rustfmt` won't do it because it gets confused by the `;` in slog's log macros).
## Proposed Changes
It's reasonably often that we want to manually convert an attestation to indexed form. This PR adds an `lcli` command for doing this, using an SSZ state and a list of JSON attestations (as extracted from a JSON block) as input.
## Issue Addressed
NA
## Proposed Changes
Please list or describe the changes introduced by this PR.
## Additional Info
- Pending testing on our infra. **Please do not merge**
## Issue Addressed
#3212
## Proposed Changes
Move chain segments coming from back-fill syncing from highest priority to lowest
## Additional Info
If this does not solve the issue, next steps would be lowering the batch size for back-fill sync, and as last resort throttling the processing of these chain segments
## Issue Addressed
#3154
## Proposed Changes
Add three new metrics for the VC:
1. `vc_beacon_nodes_synced_count`
2. `vc_beacon_nodes_available_count`
3. `vc_beacon_nodes_total_count`
Their values mirror the values present in the following log line:
```
Apr 08 17:25:17.000 INFO Connected to beacon node(s) synced: 4, available: 4, total: 4, service: notifier
```
## Issue Addressed
Fixes an issue that @paulhauner found with the v2.3.0 release candidate whereby the fork choice runs introduced by #3168 tripped over each other during sync:
```
May 24 23:06:40.542 WARN Error signalling fork choice waiter slot: 3884129, error: ForkChoiceSignalOutOfOrder { current: Slot(3884131), latest: Slot(3884129) }, service: beacon
```
This can occur because fork choice is called from the state advance _and_ the per-slot task. When one of these runs takes a long time it can end up finishing after a run from a later slot, tripping the error above. The problem is resolved by not running either of these fork choice calls during sync.
Additionally, these parallel fork choice runs were causing issues in the database:
```
May 24 07:49:05.098 WARN Found a chain that should already have been pruned, head_slot: 92925, head_block_root: 0xa76c7bf1b98e54ed4b0d8686efcfdf853484e6c2a4c67e91cbf19e5ad1f96b17, service: beacon
May 24 07:49:05.101 WARN Database migration failed error: HotColdDBError(FreezeSlotError { current_split_slot: Slot(92608), proposed_split_slot: Slot(92576) }), service: beacon
```
In this case, two fork choice calls triggering the finalization processing were being processed out of order due to differences in their processing time, causing the background migrator to try to advance finalization _backwards_ 😳. Removing the parallel fork choice runs from sync effectively addresses the issue, because these runs are most likely to have different finalized checkpoints (because of the speed at which fork choice advances during sync). In theory it's still possible to process updates out of order if any other fork choice runs end up completing out of order, but this should be much less common. Fixing out of order fork choice runs in general is difficult as it requires architectural changes like serialising fork choice updates through a single thread, or locking fork choice along with the head when it is mutated (https://github.com/sigp/lighthouse/pull/3175).
## Proposed Changes
* Don't run per-slot fork choice during sync (if head is older than 4 slots)
* Don't run state-advance fork choice during sync (if head is older than 4 slots)
* Check for monotonic finalization updates in the background migrator. This is a good defensive check to have, and I'm not sure why we didn't have it before (we may have had it and wrongly removed it).
## Proposed Changes
Add documentation for the `lighthouse db migate` command, which users will be able to use to downgrade from Lighthouse v2.3.0 on non-merge networks (mainnet & Prater).
I think it's important to get this into the live instance of the book so we can link to it from the v2.3.0 release notes.
*This PR was adapted from @pawanjay176's work in #3197.*
## Issue Addressed
Fixes a regression in https://github.com/sigp/lighthouse/pull/3168
## Proposed Changes
https://github.com/sigp/lighthouse/pull/3168 added calls to `fork_choice` in `BeaconChain::per_slot_task` function. This leads to a panic as `per_slot_task` is called from an async context which calls fork choice, which then calls `block_on`.
This PR changes the timer to call the `per_slot_task` function in a blocking thread.
Co-authored-by: Pawan Dhananjay <pawandhananjay@gmail.com>
## Issue Addressed
Removes the build status badge from the main README.md. I don't think it actually serves a purpose and it also has the downside that a spurious failure gives us a red badge. For example, v2.2.1 failed with a [spurious failure](https://github.com/sigp/lighthouse/runs/5984392665?check_suite_focus=true) and I can't see a way to re-trigger that run. It will be red until our next release.
The same test suite runs when we merge into `unstable`, so those tests must have already passed in order for the commits to get onto `stable` (assuming our workflow is followed). Github will send notifications on failed CI, so we'll still be alerted to a failure without checking this badge.
## Issue Addressed
This fixes the low-hanging Clippy lints introduced in Rust 1.61 (due any hour now). It _ignores_ one lint, because fixing it requires a structural refactor of the validator client that needs to be done delicately. I've started on that refactor and will create another PR that can be reviewed in more depth in the coming days. I think we should merge this PR in the meantime to unblock CI.
## Issue Addressed
NA
## Proposed Changes
Adds the configuration for the upcoming merge of the Ropsten network, as per:
https://github.com/eth-clients/merge-testnets/pull/9
Use the Ropsten network with: `lighthouse --network ropsten`
## Additional Info
This is still a work-in-progress. We should wait for the eth-clients/merge-testnets PR to be approved before merging this into our `unstable`.
## Issue Addressed
Upcoming spec change https://github.com/ethereum/consensus-specs/pull/2878
## Proposed Changes
1. Run fork choice at the start of every slot, and wait for this run to complete before proposing a block.
2. As an optimisation, also run fork choice 3/4 of the way through the slot (at 9s), _dequeueing attestations for the next slot_.
3. Remove the fork choice run from the state advance timer that occurred before advancing the state.
## Additional Info
### Block Proposal Accuracy
This change makes us more likely to propose on top of the correct head in the presence of re-orgs with proposer boost in play. The main scenario that this change is designed to address is described in the linked spec issue.
### Attestation Accuracy
This change _also_ makes us more likely to attest to the correct head. Currently in the case of a skipped slot at `slot` we only run fork choice 9s into `slot - 1`. This means the attestations from `slot - 1` aren't taken into consideration, and any boost applied to the block from `slot - 1` is not removed (it should be). In the language of the linked spec issue, this means we are liable to attest to C, even when the majority voting weight has already caused a re-org to B.
### Why remove the call before the state advance?
If we've run fork choice at the start of the slot then it has already dequeued all the attestations from the previous slot, which are the only ones eligible to influence the head in the current slot. Running fork choice again is unnecessary (unless we run it for the next slot and try to pre-empt a re-org, but I don't currently think this is a great idea).
### Performance
Based on Prater testing this adds about 5-25ms of runtime to block proposal times, which are 500-1000ms on average (and spike to 5s+ sometimes due to state handling issues 😢 ). I believe this is a small enough penalty to enable it by default, with the option to disable it via the new flag `--fork-choice-before-proposal-timeout 0`. Upcoming work on block packing and state representation will also reduce block production times in general, while removing the spikes.
### Implementation
Fork choice gets invoked at the start of the slot via the `per_slot_task` function called from the slot timer. It then uses a condition variable to signal to block production that fork choice has been updated. This is a bit funky, but it seems to work. One downside of the timer-based approach is that it doesn't happen automatically in most of the tests. The test added by this PR has to trigger the run manually.
## Issue Addressed
@z3n-chada is currently getting a `PayloadIdUnavailable` error when connecting lighthouse to Erigon and it's difficult to discern why so this just logs out the response status from the EE when we hit an `PayloadIdUnavailable` error
Co-authored-by: realbigsean <sean@sigmaprime.io>
