It seems `post_validator_duties_sync` is the only api which doesn't have its own metric in `duties_service`, this PR adds `metrics::VALIDATOR_DUTIES_SYNC_HTTP_POST` for completeness.
## Issue Addressed
Addresses #4117
## Proposed Changes
See https://github.com/ethereum/keymanager-APIs/pull/58 for proposed API specification.
## TODO
- [x] ~~Add submission to BN~~
- removed, see discussion in [keymanager API](https://github.com/ethereum/keymanager-APIs/pull/58)
- [x] ~~Add flag to allow voluntary exit via the API~~
- no longer needed now the VC doesn't submit exit directly
- [x] ~~Additional verification / checks, e.g. if validator on same network as BN~~
- to be done on client side
- [x] ~~Potentially wait for the message to propagate and return some exit information in the response~~
- not required
- [x] Update http tests
- [x] ~~Update lighthouse book~~
- not required if this endpoint makes it to the standard keymanager API
Co-authored-by: Paul Hauner <paul@paulhauner.com>
Co-authored-by: Jimmy Chen <jimmy@sigmaprime.io>
## Issue Addressed
NA
## Proposed Changes
In #4024 we added metrics to expose the latency measurements from a VC to each BN. Whilst playing with these new metrics on our infra I realised it would be great to have a single metric to make sure that the primary BN for each VC has a reasonable latency. With the current "metrics for all BNs" it's hard to tell which is the primary.
## Additional Info
NA
## Issue Addressed
Closes#3963 (hopefully)
## Proposed Changes
Compute attestation selection proofs gradually each slot rather than in a single `join_all` at the start of each epoch. On a machine with 5k validators this replaces 5k tasks signing 5k proofs with 1 task that signs 5k/32 ~= 160 proofs each slot.
Based on testing with Goerli validators this seems to reduce the average time to produce a signature by preventing Tokio and the OS from falling over each other trying to run hundreds of threads. My testing so far has been with local keystores, which run on a dynamic pool of up to 512 OS threads because they use [`spawn_blocking`](https://docs.rs/tokio/1.11.0/tokio/task/fn.spawn_blocking.html) (and we haven't changed the default).
An earlier version of this PR hyper-optimised the time-per-signature metric to the detriment of the entire system's performance (see the reverted commits). The current PR is conservative in that it avoids touching the attestation service at all. I think there's more optimising to do here, but we can come back for that in a future PR rather than expanding the scope of this one.
The new algorithm for attestation selection proofs is:
- We sign a small batch of selection proofs each slot, for slots up to 8 slots in the future. On average we'll sign one slot's worth of proofs per slot, with an 8 slot lookahead.
- The batch is signed halfway through the slot when there is unlikely to be contention for signature production (blocks are <4s, attestations are ~4-6 seconds, aggregates are 8s+).
## Performance Data
_See first comment for updated graphs_.
Graph of median signing times before this PR:
Graph of update attesters metric (includes selection proof signing) before this PR:
Median signing time after this PR (prototype from 12:00, updated version from 13:30):
99th percentile on signing times (bounded attestation signing from 16:55, now removed):
Attester map update timing after this PR:
Selection proof signings per second change:
## Link to late blocks
I believe this is related to the slow block signings because logs from Stakely in #3963 show these two logs almost 5 seconds apart:
> Feb 23 18:56:23.978 INFO Received unsigned block, slot: 5862880, service: block, module: validator_client::block_service:393
> Feb 23 18:56:28.552 INFO Publishing signed block, slot: 5862880, service: block, module: validator_client::block_service:416
The only thing that happens between those two logs is the signing of the block:
0fb58a680d/validator_client/src/block_service.rs (L410-L414)
Helpfully, Stakely noticed this issue without any Lighthouse BNs in the mix, which pointed to a clear issue in the VC.
## TODO
- [x] Further testing on testnet infrastructure.
- [x] Make the attestation signing parallelism configurable.
## Issue Addressed
NA
## Proposed Changes
Adds a service which periodically polls (11s into each mainnet slot) the `node/version` endpoint on each BN and roughly measures the round-trip latency. The latency is exposed as a `DEBG` log and a Prometheus metric.
The `--latency-measurement-service` has been added to the VC, with the following options:
- `--latency-measurement-service true`: enable the service (default).
- `--latency-measurement-service`: (without a value) has the same effect.
- `--latency-measurement-service false`: disable the service.
## Additional Info
Whilst looking at our staking setup, I think the BN+VC latency is contributing to late blocks. Now that we have to wait for the builders to respond it's nice to try and do everything we can to reduce that latency. Having visibility is the first step.
## Issue Addressed
Cleans up all the remnants of 4844 in capella. This makes sure when 4844 is reviewed there is nothing we are missing because it got included here
## Proposed Changes
drop a bomb on every 4844 thing
## Additional Info
Merge process I did (locally) is as follows:
- squash merge to produce one commit
- in new branch off unstable with the squashed commit create a `git revert HEAD` commit
- merge that new branch onto 4844 with `--strategy ours`
- compare local 4844 to remote 4844 and make sure the diff is empty
- enjoy
Co-authored-by: Paul Hauner <paul@paulhauner.com>
This fixes issues with certain metrics scrapers, which might error if the content-type is not correctly set.
## Issue Addressed
Fixes https://github.com/sigp/lighthouse/issues/3437
## Proposed Changes
Simply set header: `Content-Type: text/plain` on metrics server response. Seems like the errored branch does this correctly already.
## Additional Info
This is needed also to enable influx-db metric scraping which work very nicely with Geth.
## Issue Addressed
Resolves#2521
## Proposed Changes
Add a metric that indicates the next attestation duty slot for all managed validators in the validator client.
## Issue Addressed
https://github.com/sigp/lighthouse/issues/3091
Extends https://github.com/sigp/lighthouse/pull/3062, adding pre-bellatrix block support on blinded endpoints and allowing the normal proposal flow (local payload construction) on blinded endpoints. This resulted in better fallback logic because the VC will not have to switch endpoints on failure in the BN <> Builder API, the BN can just fallback immediately and without repeating block processing that it shouldn't need to. We can also keep VC fallback from the VC<>BN API's blinded endpoint to full endpoint.
## Proposed Changes
- Pre-bellatrix blocks on blinded endpoints
- Add a new `PayloadCache` to the execution layer
- Better fallback-from-builder logic
## Todos
- [x] Remove VC transition logic
- [x] Add logic to only enable builder flow after Merge transition finalization
- [x] Tests
- [x] Fix metrics
- [x] Rustdocs
Co-authored-by: Mac L <mjladson@pm.me>
Co-authored-by: realbigsean <sean@sigmaprime.io>
## Issue Addressed
Lays the groundwork for builder API changes by implementing the beacon-API's new `register_validator` endpoint
## Proposed Changes
- Add a routine in the VC that runs on startup (re-try until success), once per epoch or whenever `suggested_fee_recipient` is updated, signing `ValidatorRegistrationData` and sending it to the BN.
- TODO: `gas_limit` config options https://github.com/ethereum/builder-specs/issues/17
- BN only sends VC registration data to builders on demand, but VC registration data *does update* the BN's prepare proposer cache and send an updated fcU to a local EE. This is necessary for fee recipient consistency between the blinded and full block flow in the event of fallback. Having the BN only send registration data to builders on demand gives feedback directly to the VC about relay status. Also, since the BN has no ability to sign these messages anyways (so couldn't refresh them if it wanted), and validator registration is independent of the BN head, I think this approach makes sense.
- Adds upcoming consensus spec changes for this PR https://github.com/ethereum/consensus-specs/pull/2884
- I initially applied the bit mask based on a configured application domain.. but I ended up just hard coding it here instead because that's how it's spec'd in the builder repo.
- Should application mask appear in the api?
Co-authored-by: realbigsean <sean@sigmaprime.io>
## 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
#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
#3103
## Proposed Changes
Parse `http-address` and `metrics-address` as `IpAddr` for both the beacon node and validator client to support IPv6 addresses.
Also adjusts parsing of CORS origins to allow for IPv6 addresses.
## Usage
You can now set `http-address` and/or `metrics-address` flags to IPv6 addresses.
For example, the following:
`lighthouse bn --http --http-address :: --metrics --metrics-address ::1`
will expose the beacon node HTTP server on `[::]` (equivalent of `0.0.0.0` in IPv4) and the metrics HTTP server on `localhost` (the equivalent of `127.0.0.1` in IPv4)
The beacon node API can then be accessed by:
`curl "http://[server-ipv6-address]:5052/eth/v1/some_endpoint"`
And the metrics server api can be accessed by:
`curl "http://localhost:5054/metrics"` or by `curl "http://[::1]:5054/metrics"`
## Additional Info
On most Linux distributions the `v6only` flag is set to `false` by default (see the section for the `IPV6_V6ONLY` flag in https://www.man7.org/linux/man-pages/man7/ipv6.7.html) which means IPv4 connections will continue to function on a IPv6 address (providing it is appropriately mapped). This means that even if the Lighthouse API is running on `::` it is also possible to accept IPv4 connections.
However on Windows, this is not the case. The `v6only` flag is set to `true` so binding to `::` will only allow IPv6 connections.
[EIP-3030]: https://eips.ethereum.org/EIPS/eip-3030
[Web3Signer]: https://consensys.github.io/web3signer/web3signer-eth2.html
## Issue Addressed
Resolves#2498
## Proposed Changes
Allows the VC to call out to a [Web3Signer] remote signer to obtain signatures.
## Additional Info
### Making Signing Functions `async`
To allow remote signing, I needed to make all the signing functions `async`. This caused a bit of noise where I had to convert iterators into `for` loops.
In `duties_service.rs` there was a particularly tricky case where we couldn't hold a write-lock across an `await`, so I had to first take a read-lock, then grab a write-lock.
### Move Signing from Core Executor
Whilst implementing this feature, I noticed that we signing was happening on the core tokio executor. I suspect this was causing the executor to temporarily lock and occasionally trigger some HTTP timeouts (and potentially SQL pool timeouts, but I can't verify this). Since moving all signing into blocking tokio tasks, I noticed a distinct drop in the "atttestations_http_get" metric on a Prater node:
I think this graph indicates that freeing the core executor allows the VC to operate more smoothly.
### Refactor TaskExecutor
I noticed that the `TaskExecutor::spawn_blocking_handle` function would fail to spawn tasks if it were unable to obtain handles to some metrics (this can happen if the same metric is defined twice). It seemed that a more sensible approach would be to keep spawning tasks, but without metrics. To that end, I refactored the function so that it would still function without metrics. There are no other changes made.
## TODO
- [x] Restructure to support multiple signing methods.
- [x] Add calls to remote signer from VC.
- [x] Documentation
- [x] Test all endpoints
- [x] Test HTTPS certificate
- [x] Allow adding remote signer validators via the API
- [x] Add Altair support via [21.8.1-rc1](https://github.com/ConsenSys/web3signer/releases/tag/21.8.1-rc1)
- [x] Create issue to start using latest version of web3signer. (See #2570)
## Notes
- ~~Web3Signer doesn't yet support the Altair fork for Prater. See https://github.com/ConsenSys/web3signer/issues/423.~~
- ~~There is not yet a release of Web3Signer which supports Altair blocks. See https://github.com/ConsenSys/web3signer/issues/391.~~
## Proposed Changes
* Implement the validator client and HTTP API changes necessary to support Altair
Co-authored-by: realbigsean <seananderson33@gmail.com>
Co-authored-by: Michael Sproul <michael@sigmaprime.io>
## Issue Addressed
Resolves#2069
## Proposed Changes
- Adds a `--doppelganger-detection` flag
- Adds a `lighthouse/seen_validators` endpoint, which will make it so the lighthouse VC is not interopable with other client beacon nodes if the `--doppelganger-detection` flag is used, but hopefully this will become standardized. Relevant Eth2 API repo issue: https://github.com/ethereum/eth2.0-APIs/issues/64
- If the `--doppelganger-detection` flag is used, the VC will wait until the beacon node is synced, and then wait an additional 2 epochs. The reason for this is to make sure the beacon node is able to subscribe to the subnets our validators should be attesting on. I think an alternative would be to have the beacon node subscribe to all subnets for 2+ epochs on startup by default.
## Additional Info
I'd like to add tests and would appreciate feedback.
TODO: handle validators started via the API, potentially make this default behavior
Co-authored-by: realbigsean <seananderson33@gmail.com>
Co-authored-by: Michael Sproul <michael@sigmaprime.io>
Co-authored-by: Paul Hauner <paul@paulhauner.com>
## Issue Addressed
Resolves#2313
## Proposed Changes
Provide `BeaconNodeHttpClient` with a dedicated `Timeouts` struct.
This will allow granular adjustment of the timeout duration for different calls made from the VC to the BN. These can either be a constant value, or as a ratio of the slot duration.
Improve timeout performance by using these adjusted timeout duration's only whenever a fallback endpoint is available.
Add a CLI flag called `use-long-timeouts` to revert to the old behavior.
## Additional Info
Additionally set the default `BeaconNodeHttpClient` timeouts to the be the slot duration of the network, rather than a constant 12 seconds. This will allow it to adjust to different network specifications.
Co-authored-by: Paul Hauner <paul@paulhauner.com>
## Issue Addressed
Closes#2052
## Proposed Changes
- Refactor the attester/proposer duties endpoints in the BN
- Performance improvements
- Fixes some potential inconsistencies with the dependent root fields.
- Removes `http_api::beacon_proposer_cache` and just uses the one on the `BeaconChain` instead.
- Move the code for the proposer/attester duties endpoints into separate files, for readability.
- Refactor the `DutiesService` in the VC
- Required to reduce the delay on broadcasting new blocks.
- Gets rid of the `ValidatorDuty` shim struct that came about when we adopted the standard API.
- Separate block/attestation duty tasks so that they don't block each other when one is slow.
- In the VC, use `PublicKeyBytes` to represent validators instead of `PublicKey`. `PublicKey` is a legit crypto object whilst `PublicKeyBytes` is just a byte-array, it's much faster to clone/hash `PublicKeyBytes` and this change has had a significant impact on runtimes.
- Unfortunately this has created lots of dust changes.
- In the BN, store `PublicKeyBytes` in the `beacon_proposer_cache` and allow access to them. The HTTP API always sends `PublicKeyBytes` over the wire and the conversion from `PublicKey` -> `PublickeyBytes` is non-trivial, especially when queries have 100s/1000s of validators (like Pyrmont).
- Add the `state_processing::state_advance` mod which dedups a lot of the "apply `n` skip slots to the state" code.
- This also fixes a bug with some functions which were failing to include a state root as per [this comment](072695284f/consensus/state_processing/src/state_advance.rs (L69-L74)). I couldn't find any instance of this bug that resulted in anything more severe than keying a shuffling cache by the wrong block root.
- Swap the VC block service to use `mpsc` from `tokio` instead of `futures`. This is consistent with the rest of the code base.
~~This PR *reduces* the size of the codebase 🎉~~ It *used* to reduce the size of the code base before I added more comments.
## Observations on Prymont
- Proposer duties times down from peaks of 450ms to consistent <1ms.
- Current epoch attester duties times down from >1s peaks to a consistent 20-30ms.
- Block production down from +600ms to 100-200ms.
## Additional Info
- ~~Blocked on #2241~~
- ~~Blocked on #2234~~
## TODO
- [x] ~~Refactor this into some smaller PRs?~~ Leaving this as-is for now.
- [x] Address `per_slot_processing` roots.
- [x] Investigate slow next epoch times. Not getting added to cache on block processing?
- [x] Consider [this](072695284f/beacon_node/store/src/hot_cold_store.rs (L811-L812)) in the scenario of replacing the state roots
Co-authored-by: pawan <pawandhananjay@gmail.com>
Co-authored-by: Michael Sproul <michael@sigmaprime.io>
## Proposed Changes
Prune the slashing protection database so that it doesn't exhibit unbounded growth. Prune by dropping attestations and blocks from more than 512 epochs ago, relying on the guards that prevent signing messages with slots or epochs less than the minimum recorded in the DB.
The pruning process is potentially time consuming, so it's scheduled to run only every 512 epochs, in the last 2/3rds of a slot. This gives it at least 4 seconds to run without impacting other signing, which I think should be sufficient. I've seen it run for several minutes (yikes!) on our Pyrmont nodes, but I suspect that 1) this will only occur on the first run when the database is still huge 2) no other production users will be impacted because they don't have enough validators per node.
Pruning also happens at start-up, as I figured this is a fairly infrequent event, and if a user is experiencing problems with the VC related to pruning, it's nice to be able to trigger it with a quick restart. Users are also conditioned to not mind missing a few attestations during a restart.
We need to include a note in the release notes that users may see the message `timed out waiting for connection` the first time they prune a huge database, but that this is totally fine and to be expected (the VC will miss those attestations in the meantime).
I'm also open to making this opt-in for now, although the sooner we get users doing it, the less painful it will be: prune early, prune often!
## Issue Addressed
- Resolves#1883
## Proposed Changes
This follows on from @blacktemplar's work in #2018.
- Allows the VC to connect to multiple BN for redundancy.
- Update the simulator so some nodes always need to rely on their fallback.
- Adds some extra deprecation warnings for `--eth1-endpoint`
- Pass `SignatureBytes` as a reference instead of by value.
## Additional Info
NA
Co-authored-by: blacktemplar <blacktemplar@a1.net>
## Issue Addressed
NA
## Proposed Changes
- Adds a HTTP server to the VC which provides Prometheus metrics.
- Moves the health metrics into the `lighthouse_metrics` crate so it can be shared between BN/VC.
- Sprinkle some metrics around the VC.
- Update the book to indicate that we now have VC metrics.
- Shifts the "waiting for genesis" logic later in the `ProductionValidatorClient::new_from_cli`
- This is worth attention during the review.
## Additional Info
- ~~`clippy` has some new lints that are failing. I'll deal with that in another PR.~~
