333 Commits (d1bf212bf7dfa8e146950c3760fddad5fc102720)

Author	SHA1	Message	Date
Danil Nemirovsky	d1bf212bf7	feat: Report git commit hash as agent metadata (#4282) ... ### Description * We are storing git commit hash as an environment variable when we compile the source code of an agent. * Validator uses git commit hash as a agent version in `metadata_latest.json` file. * File `metadata_latest.json` is overwritten every time when validator is restarted. * Git commit hash is extracted by `git` util, so, it should be available during build time. * If the directory with the source code does not contain `.git` folder, git commit hash will be defaulted to `VERGEN_IDEMPOTENT_OUTPUT`. Agent Metadata is propagated to Relayer and Scraper, but is not used there. We can use it in the future, reporting git hash commit in the logs, for example. ### Backward compatibility Yes ### Testing * Tested locally with E2E Tests. * Tested against S3 by deploying docker image into k8s --------- Co-authored-by: Danil Nemirovsky <4614623+ameten@users.noreply.github.com>	4 months ago
Paul Balaji	103e6b3e12	feat: deploy to `zircuit` (#4265) ... - feat: deploy to `zircuit` - update registry + agent config	4 months ago
Daniel Savu	a49f52cc9c	feat: deploy solana, eclipsemainnet (#4238) ... ### Description <!-- What's included in this PR? --> ### Drive-by changes <!-- Are there any minor or drive-by changes also included? --> ### Related issues <!-- - Fixes #[issue number here] --> ### Backward compatibility <!-- Are these changes backward compatible? Are there any infrastructure implications, e.g. changes that would prohibit deploying older commits using this infra tooling? Yes/No --> ### Testing <!-- What kind of testing have these changes undergone? None/Manual/Unit Tests --> --------- Co-authored-by: pbio <10051819+paulbalaji@users.noreply.github.com>	4 months ago
Paul Balaji	ff3c9897ea	chore: migrate `test1` from `13371` to `133711` (#4252) ... - chore: migrate `test1` from `13371` to `133711` - due to clash with immutablezkevm - drive-by registry update + agent config update --------- Signed-off-by: pbio <10051819+paulbalaji@users.noreply.github.com>	4 months ago
Mohammed Hussan	8e942d3c6b	feat: deploy cheesechain, worldchain, xlayer (#4207) ... ### Description - Support deployment to cheesechain, worldchain and zlayer ### Drive-by changes - update `CosmosChainSchema` to add `chain_type` to fix breaking change introduced by 6854133f98	4 months ago
Daniel Savu	b643fba90a	feat: use multiple `mpsc` channels instead of `broadcast` for async interface (#4190) ... ### Description Follow up to https://github.com/hyperlane-xyz/hyperlane-monorepo/pull/4180 Lowers txid broadcast channel capacity from 1M to 30k and refactors it to use a vec of `mpsc` under the hood. ### Context The heap profiler showed that each new chain takes up 100MB of memory because of the txid broadcast channel. In addition to this, the tokio broadcast channel is a ring buffer where new items simply overwrite old items, making it hard to safely lower the capacity. As such, to be able to safely lower the capacity of this channel, usage of `tokio`'s `broadcast` channel is replaced with a `Vec` of `mpsc`s, which offer an async `send` interface for backpressure. So instead of overwriting old items, senders will block on sends, which makes sense in our case as we absolutely don't want to miss items. The submitter is slower than the indexer based on empirical observation, so blocking on channel size doesn't bottleneck throughput. I noticed the max number of channel items we can reach is 29k, by running the RC relayer after an 8 day break, during which around 160k messages were submitted. Lowering the capacity to 30k should cause us to only encounter backpressure in very rare cases, and the footprint of this channel would lower from 100MB to 3MB. ### Drive-by changes The delivery indexer in the scraper is no longer passed a txid receiver, since message deliveries are not expected to occur in the same tx as message dispatches. ### Related issues - Fixes https://github.com/hyperlane-xyz/issues/issues/1293 ### Testing E2E, by adding invariants to make sure txid indexing works --------- Co-authored-by: Trevor Porter <tkporter4@gmail.com>	4 months ago
Daniel Savu	6140e6f397	feat: Relayer heap profiling (#4180) ... ### Description During the scalability audit we've identified relayer memory usage as a cause of concern. This PR runs the profiler each time e2e is run and outputs a report called `dhat-heap.json` in the current dir. That file can be visualised with a dhat viewer, such as the one hosted by the author [here](https://nnethercote.github.io/dh_view/dh_view.html). It adds a conditionally compiled heap memory profiler ([dhat-rs](https://docs.rs/dhat/latest/dhat/)), which is only included when both the `memory-profiling` feature is enabled and the `memory-profiling` cargo profile is used. The report can order the profiling results by total lifetime allocated bytes, peak allocated bytes, total allocation count, etc. Below is the result of sorting by total lifetime allocated bytes, which reveals that the biggest memory hogs are the hook indexing channel [senders](https://github.com/hyperlane-xyz/hyperlane-monorepo/blob/main/rust/hyperlane-base/src/contract_sync/mod.rs#L52) (one per origin chain), taking up 100+ MB. Its size can most likely be reduced from 1M to 1k.  ### Drive-by changes Adds the remaining IDs in the channel as a field to the hook indexing log, to know if it's safe to lower the channel size. ### Related issues - Fixes https://github.com/hyperlane-xyz/issues/issues/1293 ### Backward compatibility Yes, since everything is conditionally compiled / isolated in a new cargo profile. ### Testing Manual	4 months ago
Daniel Savu	97588b6fc1	fix(relayer): don't submit reverting messages in batches (#4169) ... ### Description Depends on https://github.com/hyperlane-xyz/hyperlane-monorepo/pull/4159 Some messages in a Multicall would revert because previous messages would invalidate them temporarily. The IGP allowance of those messages would be deducted, and if this happened more than once, those messages would become stuck. This PR fixes those unnecesary spends on reverting `process` txs by using `.call()` to simulate the outcome of a Multicall tx. Reverting messages are removed from the batch and submitted serially, and if the batch is only left with one non-reverting message, even that one is submitted serially (rather than through a batch of one message). Messages that did get submitted have their igp allowance deducted and status updated in `update_sent_ops_state`. ### Drive-by changes <!-- Are there any minor or drive-by changes also included? --> ### Related issues - Fixes https://github.com/hyperlane-xyz/issues/issues/1294 ### Backward compatibility <!-- Are these changes backward compatible? Are there any infrastructure implications, e.g. changes that would prohibit deploying older commits using this infra tooling? Yes/No --> ### Testing Tested by manually setting the result of `call` to failed for all messages, to see if they are submitted serially. In prod, we can track the `Submitted transaction batch` log, which now includes any excluded messages from the batch.	4 months ago
Daniel Savu	bf1dad7f77	feat(relayer): new status label for IGP not found (#4106) ... ### Description Adds new status label to relayer opqueue for operations whose gas payments were not found. This now creates a distinction between operations without a payment and those with insufficient payment. Helps better uncover cases where hook indexing misses events because there is no receipt yet	5 months ago
Paul Balaji	0dedbf5a02	feat: deploy to new chains (endurance, fuse, zora) (#4148) ... - feat: deploy to new chains (endurance, fuse, zora) - https://github.com/hyperlane-xyz/issues/issues/1308 - also includes https://github.com/hyperlane-xyz/issues/issues/1305 - registry PR https://github.com/hyperlane-xyz/hyperlane-registry/pull/104	5 months ago
Daniel Savu	8c8f39abfe	feat(relayer): ListOperations api endpoint (#4089) ... ### Description Adds a `list_operations` API endpoint to the relayer, which returns all operations in its prepare queue. Example output from `GET http://0.0.0.0:9092/list_operations?destination_domain=13371` ``` [ { "app_context": null, "message": { "body": [ 18, 52 ], "destination": 13371, "nonce": 1, "origin": 13373, "recipient": "0x000000000000000000000000927b167526babb9be047421db732c663a0b77b11", "sender": "0x000000000000000000000000927b167526babb9be047421db732c663a0b77b11", "version": 3 }, "num_retries": 13, "status": { "Retry": "CouldNotFetchMetadata" }, "submitted": false, "type": "PendingMessage" }, { "app_context": null, "message": { "body": [ 18, 52 ], "destination": 13371, "nonce": 2, "origin": 13372, "recipient": "0x000000000000000000000000927b167526babb9be047421db732c663a0b77b11", "sender": "0x000000000000000000000000927b167526babb9be047421db732c663a0b77b11", "version": 3 }, "num_retries": 13, "status": { "Retry": "CouldNotFetchMetadata" }, "submitted": false, "type": "PendingMessage" } ] ``` ### Drive-by changes Makes `PendingOperation` serializable, which means switching from JSON format to e.g. CSV (for easily filtering / aggregating in excel) should be very easy. ### Related issues <!-- - Fixes #[issue number here] --> ### Backward compatibility Yes ### Testing Manual and Unit Tests	5 months ago
Paul Balaji	d316772243	feat: deploy to BOB, Mantle, Taiko (#4051) ... resolves https://github.com/hyperlane-xyz/issues/issues/1289 - feat: deploy to BOB, Mantle, Taiko - drive-by: fix deploying to specific chain - drive-by: adding AVS operators for eth - drive-by: support non-evm remotes in mainnet3 core config --------- Signed-off-by: pbio <10051819+paulbalaji@users.noreply.github.com>	5 months ago
Daniel Savu	d90d8f86b4	chore: reduce op-queue verbosity (#4092) ... ### Description The gcp bill has been shooting up, and it's most likely due to the increased number of undeliverable messages being added to and removed from the queue, causing lots of `debug` level logs on `push` and `pop_many` to be emitted. This PR changes the verbosity to `trace` - we'll have lower visibility over the lifetime of a message, but there should be sufficient logs remaining to tell us what's going on (such as [operation prepared](30630c4fa6/rust/agents/relayer/src/msg/op_submitter.rs (L240)), [confirming a pre-submitted one](30630c4fa6/rust/agents/relayer/src/msg/op_submitter.rs (L260)), [submitting single op](30630c4fa6/rust/agents/relayer/src/msg/op_submitter.rs (L313)), [submitting batch](30630c4fa6/rust/agents/relayer/src/msg/op_submitter.rs (L456)), and [op confirmed](30630c4fa6/rust/agents/relayer/src/msg/op_submitter.rs (L386))). @paulbalaji suggested measuring log counts, so here they are: - in the last 6h, we had 42M `hyperlane` relayer logs, of which 22M where `push` logs and 1.1M were `pop_many` (no `pop`s) - if we make push ones trace, 5.5% of the remaining logs would be pop_many ### Drive-by changes <!-- Are there any minor or drive-by changes also included? --> ### Related issues <!-- - Fixes #[issue number here] --> ### Backward compatibility <!-- Are these changes backward compatible? Are there any infrastructure implications, e.g. changes that would prohibit deploying older commits using this infra tooling? Yes/No --> ### Testing <!-- What kind of testing have these changes undergone? None/Manual/Unit Tests -->	5 months ago
Trevor Porter	6f4ef05d41	feat: faster way of figuring out how many new scraper rows were inserted (#4073) ... ### Description ### Context Full context / thread: https://discord.com/channels/935678348330434570/1254871503825141923 Copied here: so this is kinda interesting basically the gas payment indexing in the scraper task is consistently prioritizing hook-based indexing work over block range indexing work this only manifests in the scraper and for arbitrum because: there are a lot of arbitrum origin messages on the scraper, each gas payment indexing tick takes a very long time because of some very suboptimal queries. store_payments can take like 10-15 seconds! You can see batches of logs Found log(s) in index range https://cloudlogging.app.goo.gl/Bo9Q7YwyziSqjsEu7 that correspond with the gas payment contract sync block height (green line) advancing https://abacusworks.grafana.net/goto/poW4KOQIg?orgId=1  but we're still indexing gas payments bc of hook indexing - there are tons of Found log(s) for tx id https://cloudlogging.app.goo.gl/QvDaBuf67SjjKviw6 In the scraper's store_payments we bizarrely end up performing 2 very expensive queries with the calls to payments_count - it gets the count of gas payments for the provided domain so it can figure out how many new gas payments were actually inserted https://github.com/hyperlane-xyz/hyperlane-monorepo/blob/main/rust/agents/scraper/src/db/payment.rs#L27-L64 These queries vary a bunch in how expensive they can be - sometimes each one is like 1-2 seconds, sometimes it gets up to 10 seconds each though https://cloudlogging.app.goo.gl/cDFNH3oxh6WWV3Go8  easiest fix is probably to figure out how to make store_payments less expensive - we probably don't need to be making these calls to payments_count to figure out which ones are new which ill do atm there's maybe a decent argument to having hook indexing and range indexing not contend with one another but we should be able to kick that can also an annoying thing is https://github.com/hyperlane-xyz/hyperlane-monorepo/blob/main/rust/hyperlane-base/src/contract_sync/mod.rs#L155 this log not showing up, probably some weird instrumentation setup or something makes debugging harder ### The fix We want to keep supporting the possibility that we've double-indexed a data type, and to only return the number of new rows inserted. We want to keep each indexing task as non-conflicting as possible - we know that only one scraper is running at a time, but that a task for each chain and data type is executing concurrently. It's possible for e.g. two gas payment indexing tasks to try to insert into the same table concurrently, but it'll be for different domains. Insert statement concurrency issues aren't an issue because the only chance for conflict is the auto-incrementing primary key id, but this is all handled properly by database internals. Insert results only give the latest ID resulting from the insertion. E.g. if no matter how many rows were inserted, it gives the latest ID from that batch, but no info about how many were in the batch. We could do something like get the latest id in the table for all domains (this is cheap) before, then do the insert, and then get the latest id again, but this is susceptible to race conditions if concurrent tasks get an insertion in during that time as well. Shoving all these into a transaction doesn't seem to solve the problem either unless there's a lock on the table. However, sea-orm doesn't support table level locks, just select locks. We could also use a CTE to perform as a way to perform the select & insertion atomically and return the value of the select - but again sea-orm doesn't seem to provide a nice way of doing this from what I could tell. Instead, my suggested fix is: 1. Get the latest id in the table relating to the specific domain (this is cheap, < 100ms) 2. Perform the insertion 3. Count how many new rows for the specific domain now exist whose id is > the id from (1). This is also cheap, < 100ms. If there is a very long time since the last insertion for the domain, this could take longer, but in practice this doesn't seem to be the case. If this couple order of magnitude improvement still poses issues in the future, we can consider other alternatives, probably involving locking or fancier CTE queries. Some example queries and how long they take: Before (2x of these would occur!): <img width="477" alt="Screen Shot 2024-06-28 at 12 36 41 PM" src="https://github.com/hyperlane-xyz/hyperlane-monorepo/assets/20362969/5cc8ac35-74fb-4847-9c6f-448deb253a0c"> Now: 1. Max id for a given domain <img width="494" alt="Screen Shot 2024-06-28 at 12 37 00 PM" src="https://github.com/hyperlane-xyz/hyperlane-monorepo/assets/20362969/56dbb3ff-b540-458c-809c-56ebfbfd3b0f"> 2. Subsequent count but with an ID filter - note I actually changed this query to look for IDs > 3600000, which includes 55300 rows and it's still just 102ms! <img width="632" alt="Screen Shot 2024-06-28 at 12 37 49 PM" src="https://github.com/hyperlane-xyz/hyperlane-monorepo/assets/20362969/11593180-682f-4867-9867-97fd93512802"> ### Drive-by changes - Updated the neutron image along with a deploy there ### Related issues ### Backward compatibility ### Testing Ran some sorta ad-hoc unit tests to test my new query by running against the prod db with a tokio::task. Tested for a new domain and an existing one	5 months ago
Daniel Savu	aecb65a986	feat: persist status labels (#4043) ... ### Description Builds on top of https://github.com/hyperlane-xyz/hyperlane-monorepo/pull/4003 to persist updates to an operation's status to the db, so we no longer default to `FirstPrepareAttempt` upon restarting the relayer ### Drive-by changes Moves `hyperlane_db` from `hyperlane_base` to `hyperlane_core` so it can be added to the `PendingOperation` trait. `PendingMessage` already stores a reference to the db in `ctx`, so I figured it doesn't break any assumptions. ### Related issues <!-- - Fixes #[issue number here] --> ### Backward compatibility Yes ### Testing Unit tests encoding of `PendingOperationStatus`, need to manually tests e2e	5 months ago
Daniel Savu	d213e32341	feat: relayer queue operation labels (#4003) ... ### Description - adds status labels to operations in the relayer queues. In practice, the most useful ones are those for when a message is retried due to an error. Labels are defined as enums to keep the cardinality finite, the size of prometheus metrics could get very large - whenever a message is pushed to a queue, the interface requires specifying the new status as a `Option<PendingOperationStatus>`. If `None`, the status is left as before (such as in the case when an operation isn't ready to be retried). - a limitation is that labels aren't stored in the DB, so messages are loaded with the `FirstPrepareAttempt` label. I just triggered retries for destinations I wanted to get labels for, but we could persist to the db in a follow-up PR - Example prep queue labels for inevm: <img width="621" alt="Screenshot 2024-06-21 at 18 05 58" src="https://github.com/hyperlane-xyz/hyperlane-monorepo/assets/23065004/f53b2858-371f-4b6b-86d9-5c21409d641f"> ### Drive-by changes - removes `make_op_try` as it doesn't really add anything that couldn't be done with regular Rust - bumps the `strum` version because there was a bug in the previous one, preventing inner enum fields from being converted to strings ### Related issues <!-- - Fixes #[issue number here] --> ### Backward compatibility <!-- Yes ### Testing Manual	5 months ago
Daniel Savu	718d2c27da	chore: don't reset retry count when processing retry API reqs (#4037) ... ### Description When messages are retried due to the Message Retry API, we bring messages to the front of the queue (by changing their `next_attempt_after`), but also reset their retry count. For messages that are unprocessable due to various reasons (and hence have very high `next_attempt_after`), resetting the retry count will cause the relayer to waste time instead of trying to process newer messages. This PR makes it such that the retry API will cause messages to be retried immediately once, but won't boost their priority in the queue afterwards (because `num_retries` isn't changed). ### Drive-by changes <!-- Are there any minor or drive-by changes also included? --> ### Related issues <!-- - Fixes #[issue number here] --> ### Backward compatibility <!-- Are these changes backward compatible? Are there any infrastructure implications, e.g. changes that would prohibit deploying older commits using this infra tooling? Yes/No --> ### Testing <!-- What kind of testing have these changes undergone? None/Manual/Unit Tests -->	5 months ago
Trevor Porter	a109937f96	chore: some throughput tweaks (#4036) ... ### Description Context: https://discord.com/channels/935678348330434570/1254036693217054794 ### Drive-by changes <!-- Are there any minor or drive-by changes also included? --> ### Related issues <!-- - Fixes #[issue number here] --> ### Backward compatibility <!-- Are these changes backward compatible? Are there any infrastructure implications, e.g. changes that would prohibit deploying older commits using this infra tooling? Yes/No --> ### Testing <!-- What kind of testing have these changes undergone? None/Manual/Unit Tests -->	5 months ago
Trevor Porter	c84a1dc73b	feat: address blacklisting in the relayer (#4000) ... ### Description See context here https://discord.com/channels/935678348330434570/1236041564673806427/1252918894755319920 and https://discord.com/channels/935678348330434570/1252953507833708605/1252954456400592997 This is a rudimentary first step toward blacklisting addresses. Future options are outlined in https://discord.com/channels/935678348330434570/1236041564673806427/1252918894755319920. Eventually we can also fetch these addresses directly within the agent from an updated list that we cache and re-fetch occasionally. For now, this just relies on configured addresses. The approach is: - Allow hex addresses to be configured. These are intentionally treated as a Vec<u8> instead of an H160 / H256. This is because treating them as H160 is a leak of Ethereum-specific logic into non-chain-specific code in the relayer, and treating them as H256 would pad Ethereum addresses with a bunch of zeroes which would diminish the accuracy of our heuristic implemented in this. The subsequence stuff is a bit unfortunate because it's O(n^2) but 🤷‍♂️ - The heuristic is as follows - if any of the configured addresses are found as a subsequence in the message sender, recipient, or body, it's thrown out by the message processor, and the message will never make its way to the op_submitter. This heuristic for the body is to prevent warp route transfers that include that address. We could maybe convert Vec<u8> to H256 for the sender and recipient, I'm open to this - I just went with the substring approach bc it felt easiest tbh. ### Drive-by changes <!-- Are there any minor or drive-by changes also included? --> ### Related issues - Fixes https://github.com/hyperlane-xyz/issues/issues/1285 ### Backward compatibility <!-- Are these changes backward compatible? Are there any infrastructure implications, e.g. changes that would prohibit deploying older commits using this infra tooling? Yes/No --> ### Testing <!-- What kind of testing have these changes undergone? None/Manual/Unit Tests -->	6 months ago
Daniel Savu	3bb9d0a767	fix(cosmos): validator checkpoint & agent indexing improvements (#3958) ... - propagates RPC errors to the indexing logic, so errors cause sleeps instead of overwhelming the RPCs. This also greatly reduces cosmos log noisiness: from 9.5k logs per min, to 2.5k logs per min. - checks for published checkpoints in reverse, to prioritize recent messages among those from before the validator was spun up - fixes a validator bug where errors in checkpoint publishing would cause the entire checkpoint queue to be reiterated through, from scratch - Builds on top of https://github.com/hyperlane-xyz/hyperlane-monorepo/pull/3887 so we can test this out on osmosis ### Description <!-- What's included in this PR? --> ### Drive-by changes <!-- Are there any minor or drive-by changes also included? --> ### Related issues <!-- - Fixes #[issue number here] --> ### Backward compatibility <!-- Are these changes backward compatible? Are there any infrastructure implications, e.g. changes that would prohibit deploying older commits using this infra tooling? Yes/No --> ### Testing <!-- What kind of testing have these changes undergone? None/Manual/Unit Tests -->	6 months ago
Trevor Porter	bbbe93206f	chore: bump default ISM max depth from 5 to 7 (#3989) ... ### Description Needed for some Renzo messages that hit the max depth ### Drive-by changes - drive by to update the agent config with the updated registry ### Related issues <!-- - Fixes #[issue number here] --> ### Backward compatibility <!-- Are these changes backward compatible? Are there any infrastructure implications, e.g. changes that would prohibit deploying older commits using this infra tooling? Yes/No --> ### Testing <!-- What kind of testing have these changes undergone? None/Manual/Unit Tests -->	6 months ago
Paul Balaji	5cc64eb095	feat: deploy to new chains (fraxtal, linea, sei) (#3887) ... https://github.com/hyperlane-xyz/issues/issues/1279 - adds `fraxtal`, `linea`, `sei` - supports `osmosis` as a remote - ~drive-by hack to skip reconfiguring the same gas oracle (cc @tkporter)~ - drive-by fix to bring the logic of estimating tx + adding 10% buffer into the ism factory (as originally seen in the multiprovider handleDeploy function) - drive-by fix for creating custom safes - drive-by support of `osmosis` as a remote chain - drive-by addition of `osmosis` to the rust agent --------- Signed-off-by: Paul Balaji <paul@hyperlane.xyz> Co-authored-by: Trevor Porter <tkporter4@gmail.com> Co-authored-by: nambrot <nambrot@googlemail.com>	6 months ago
Daniel Savu	5a31e7b5d3	chore: debug slow relayer startup (#3932) ... ### Description Opened this PR to debug slow relayer startup, but it actually looks like that's already been fixed by hook indexing. The requirement for fast startup is that the processor sees new messages and then one of the following happens: - (1) if the message is meant for an unknown domain, it drops it ([this log](6d132b02a3/rust/agents/relayer/src/msg/processor.rs (L269))), otherwise - it forwards the message to the submitter, which then: - (2) adds it straight to the confirm queue because it's already been submitted ([here](6d132b02a3/rust/agents/relayer/src/msg/pending_message.rs (L159))) - (3) successfully builds metadata ([here](6d132b02a3/rust/agents/relayer/src/msg/op_submitter.rs (L231))) - (4) in some cases building metadata fails because the ism doesn't exist The relayer logs either of the four cases (in most cases (2), because we're running a second relayer for now). To be even more confident in fast startup, I ran a single relayer and restarted it. New messages were successfully processed according to one of the four cases above, and the prep queues didn't increase ### Drive-by changes <!-- Are there any minor or drive-by changes also included? --> ### Related issues <!-- - Fixes #[issue number here] --> ### Backward compatibility <!-- Are these changes backward compatible? Are there any infrastructure implications, e.g. changes that would prohibit deploying older commits using this infra tooling? Yes/No --> ### Testing <!-- What kind of testing have these changes undergone? None/Manual/Unit Tests -->	6 months ago
Daniel Savu	826b4ae57f	feat: batching igp (#3870) ... ### Description Deducts IGP cost from senders of messages within a batch, based on the total gas used by the batch and the estimates for submitting messages individually. The formula is `gas_used_by_operation = gas_used_by_tx * (operation_estimated_gas / total_estimated_cost)` Note that the individual estimate have a buffer value added to them (currently 75k), which will slightly skew proportions, though by a negligible amount. For example, given these estimates in a batch: 150k, 160k, 180k, if we add 80k to each, we go from e.g. 0.312 for the first message to 0.319 -> 2.25% error which isn't so bad. The error can be larger if operations within a batch have very different estimates, but realistically within a 5% range based on back-of-the-napkin calculations. ### Drive-by changes - The batching feature introduced an bug whereby gas expenditure would only be deducted in the `confirm` step. Now this is done in the `submit` step to account for txs that revert - Sealevel e2e can now be disabled for faster iteration! I've not done the cleanest job, but even the fact that we have this will reduce the time to run e2e locally by more than half. To use it, set `SEALEVEL_ENABLED=false` when running `run-locally` - e2e uses a new utility called `get_matching_lines`, that can be used to count (and in the future _parse_) logs, to reconstruct the state of the relayer and have more expressive correctness checks. This is used to make sure that gas is deducted for all messages, including those in batches. ### Related issues - Fixes https://github.com/hyperlane-xyz/hyperlane-monorepo/issues/3709 ### Backward compatibility Yes ### Testing E2E, using `get_matching_lines`	6 months ago
Daniel Savu	942aba8e55	feat(evm): collaborative indexing through txid sharing (#3833) ... ### Description - improved indexing reliability by sharing channels between indexing tasks, such that they inform one another when they come across a hyperlane-related transaction - the receiving end will then get the receipt of that tx and filter it for events relevant to it - this adds some flexibility to indexing since we can stop relying on watermark cursor entirely: as long as we have one task using sequence indexing, all others can rely on being informed by their channels about what txs to index - this PR is only for the EVM chains, but can be easily extended ### Drive-by changes - Enforces gas payments for all chains in e2e (previously this only used to be the case for sealevel) - The channel size is configured per type, similarly to the indexing cursor, in `contract_sync/cursors/mod.rs` - Reduces log verbosity is some places, e.g. by manually implementing `Debug` to exclude fields that aren't relevant to debugging (maybe there's even a macro annotation to exclude, but I haven't checked) - Renames `MpmcReceiver` -> `BroadcastReceiver` ### Related issues - Fixes https://github.com/hyperlane-xyz/hyperlane-monorepo/issues/3267 ### Backward compatibility <!-- Are these changes backward compatible? Are there any infrastructure implications, e.g. changes that would prohibit deploying older commits using this infra tooling? Yes/No --> ### Testing e2e tested with the watermark cursor manually disabled	6 months ago
Yorke Rhodes	49f41d9759	chore: Merge branch 'main' into cli-2.0 (#3863) ... Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com> Co-authored-by: Kunal Arora <55632507+aroralanuk@users.noreply.github.com> Co-authored-by: Connor McEwen <connor.mcewen@gmail.com> Co-authored-by: Trevor Porter <tkporter4@gmail.com> Co-authored-by: byeongsu-hong <hong@byeongsu.dev> Co-authored-by: Avi Atkin <103125634+avious00@users.noreply.github.com> Co-authored-by: Daniel Savu <23065004+daniel-savu@users.noreply.github.com> Co-authored-by: J M Rossy <jm.rossy@gmail.com> Co-authored-by: Ali Alaoui <aalaoui2001@gmail.com> Co-authored-by: Nam Chu Hoai <nambrot@googlemail.com> Co-authored-by: Paul Balaji <paul@hyperlane.xyz> Co-authored-by: omahs <73983677+omahs@users.noreply.github.com>	6 months ago
omahs	806f9842de	fix: duplicate word typos (#3854)	6 months ago
Daniel Savu	214f503e53	feat: forward-backward message processor (#3775) ... ### Description - adds logic for iterating message nonces in the processor task in forward-backward fashion - removes all manual calls that were updating the next nonce to query. This is done automatically in the iterator struct now - stores the highest known message nonce in the db, which is used to initialize the iterator ### Drive-by changes - Converts the concrete HyperlaneDb type stored in the processor to a trait obj, to enable mocking db responses ### Related issues - Fixes https://github.com/hyperlane-xyz/hyperlane-monorepo/issues/3796 - Fixes https://github.com/hyperlane-xyz/hyperlane-monorepo/issues/3816 ### Backward compatibility Yes - if there is no db entry for the new prefix, the processor starts from nonce zero, so no migration is required ### Testing Unit testing + e2e	6 months ago
Daniel Savu	27aabf238c	feat: Relayer tokio task instrumentation (#3760) ... ### Description - Started off adding tokio-metrics but then realised those are quite general, so while we do have instrumentation it's not exposed in our metrics endpoint - switched to adding [tokio-console](https://github.com/tokio-rs/console/tree/main), which does give insight into the lifetime of specific tasks, so we can check which ones take up a long time during relayer startup. These are only visible at the `dependencyTrace` log level, so don't affect performance in the `hyperlane` context. ### Drive-by changes <!-- Are there any minor or drive-by changes also included? --> ### Related issues - Helps debug https://github.com/hyperlane-xyz/hyperlane-monorepo/issues/3454 and any future performance issues - Does half the work for https://github.com/hyperlane-xyz/hyperlane-monorepo/issues/3239 (still need to expose these in the metrics endpoint and import the grafana template) ### Backward compatibility <!-- Are these changes backward compatible? Are there any infrastructure implications, e.g. changes that would prohibit deploying older commits using this infra tooling? Yes/No --> ### Testing <!-- What kind of testing have these changes undergone? None/Manual/Unit Tests -->	6 months ago
Trevor Porter	03ca2d1df7	Update gas oracles, deploy agents (#3755) ... ### Description - Updates gas oracles to make messages to Scroll slightly cheaper - After running into issues configuring non-EVM remotes in the IGP / gas oracle tooling (see https://discord.com/channels/935678348330434570/1238439687165579264), added a temporary band aid of just try/catching. We'll want to fix this at some point in the future - Deploys agents with a new image - Updates the agent JSON configs with some params we were running the agents with in prod via env vars - Adjusts kathy frequency on testnet to send every 1.5hrs instead of 7hrs - Stops running eclipsetestnet agents - apparently there was a reset ### Drive-by changes Some drive-by logging changes ### Related issues n/a ### Backward compatibility <!-- Are these changes backward compatible? Are there any infrastructure implications, e.g. changes that would prohibit deploying older commits using this infra tooling? Yes/No --> ### Testing <!-- What kind of testing have these changes undergone? None/Manual/Unit Tests -->	7 months ago
Daniel Savu	5d96847ffb	Multithreaded relayer (#3748) ... ### Description - Sets the tokio runtime flavor to be multithreaded and hardcodes the worker threads to 20. The latter usually defaults to the number of CPUs in the system so 20 may be high, but it worked fine in RC. Improved performance metrics: - time to first `Ingesting leaf` log: **1:30 mins (multithreaded) vs 5:51 mins (single threaded)** - time to first message in the submit queue of Optimism (one of the slowest chains to boot up for): **8:30 mins vs 20 mins** ### Drive-by changes Sets more descriptive `span`s when instrumenting the contract syncs ### Related issues <!-- - Fixes #[issue number here] --> ### Backward compatibility Yes ### Testing Manually, by checking the RC logs	7 months ago
Daniel Savu	be43b0fd91	feat: use batching configs on all execution environments (#3723) ... ### Description While we only have a tx batching implementation available for EVM chains, we still parallelize the `prepare` and `confirm` steps in the serial submitter according to the `maxBatchSize` config item. To reap the benefits of this change, we need to make these configs universal, because they were only defined within `h_eth::ChainConnectionConf`. This PR adds slight duplication because it adds the batching config on each individual connection config, but it's far simpler than including it in the `ChainConf` struct. The latter would've required changing all the trait impls of `BuildableWithProvider` to include the batching config. ### Drive-by changes Renames `serial_submitter` to `op_submitter` since it's no longer serial. ### Related issues Fixes https://github.com/hyperlane-xyz/hyperlane-monorepo/issues/3722 ### Backward compatibility Yes ### Testing Manually - added the configs to cosmos e2e and checked the logs	7 months ago
Daniel Savu	f46cf7708e	Indexing Type Wrapper (#3610) ... ### Description Depends on https://github.com/hyperlane-xyz/hyperlane-monorepo/pull/3544. Wraps all types to be indexed in an `Indexed<T>` struct in the relayer. This allows for using sequece-aware cursors for data that doesn't contain a sequence but can have it inferred from metadata (as is the case for the IGP on sealevel). `Indexed<T>` stores the sequence as an `Option`, so the db can have additional sequence data stored in it when the sequence is defined - this is how `process_indexed_gas_payment` is implemented now. In addition, the IGP payments and Deliveries are configured to be indexed with a `SequenceAware` cursor on sealevel (see `cursors/mod.rs`). ### Drive-by changes Implements the custom `Encode` and `Decode` traits for `IndexingDecorator`, which can be generalized to `Option<T>` where `T: Encode + Decode`. ### Related issues <!-- - Fixes #[issue number here] --> ### Backward compatibility <!-- Are these changes backward compatible? Are there any infrastructure implications, e.g. changes that would prohibit deploying older commits using this infra tooling? Yes/No --> ### Testing E2E	7 months ago
Daniel Savu	301239265e	WIP: don't hardcode multicall gas limit (#3710) ... ### Description <!-- What's included in this PR? --> ### Drive-by changes <!-- Are there any minor or drive-by changes also included? --> ### Related issues <!-- - Fixes #[issue number here] --> ### Backward compatibility <!-- Are these changes backward compatible? Are there any infrastructure implications, e.g. changes that would prohibit deploying older commits using this infra tooling? Yes/No --> ### Testing <!-- What kind of testing have these changes undergone? None/Manual/Unit Tests -->	7 months ago
Daniel Savu	22f367ff6c	fix: sleep after submitting to cosmos chains (#3699) ... ### Description <!-- What's included in this PR? --> ### Drive-by changes <!-- Are there any minor or drive-by changes also included? --> ### Related issues <!-- - Fixes #[issue number here] --> ### Backward compatibility <!-- Are these changes backward compatible? Are there any infrastructure implications, e.g. changes that would prohibit deploying older commits using this infra tooling? Yes/No --> ### Testing <!-- What kind of testing have these changes undergone? None/Manual/Unit Tests -->	7 months ago
Daniel Savu	5aeb83a2e5	Relayer Message Batching to EVM (#3656) ... ### Description Status: - [x] op-queue scaffolding - [x] hyperlane-ethereum integration with multicall - [x] new config items - [x] actually build batch in the op-queue submit step - [x] e2e tests with batching configured - [x] dynamic batch sizes (batch as many messages as the prepare step has ready) - [x] parallelize metadata building - [ ] proportional IGP deduction ### Drive-by changes <!-- Are there any minor or drive-by changes also included? --> ### Related issues - Fixes #3628 - Fixes #3630 ### Backward compatibility <!-- Are these changes backward compatible? Are there any infrastructure implications, e.g. changes that would prohibit deploying older commits using this infra tooling? Yes/No --> ### Testing <!-- What kind of testing have these changes undergone? None/Manual/Unit Tests --> --------- Co-authored-by: Trevor Porter <tkporter4@gmail.com>	7 months ago
Trevor Porter	d6f25ed035	Deploy to Mode & Blast (#3653) ... ### Description <!-- What's included in this PR? --> ### Drive-by changes <!-- Are there any minor or drive-by changes also included? --> ### Related issues <!-- - Fixes #[issue number here] --> ### Backward compatibility <!-- Are these changes backward compatible? Are there any infrastructure implications, e.g. changes that would prohibit deploying older commits using this infra tooling? Yes/No --> ### Testing <!-- What kind of testing have these changes undergone? None/Manual/Unit Tests -->	7 months ago
Trevor Porter	700240eb8a	Ancient8 new ism, misc infra fixes (#3606) ... ### Description - Deploys ancient8 with a new ISM where we are not quorum - Pulls out warp config to live in a single file / function so that deploy.ts and check-deploy.ts are in sync. Otherwise fork.sh is broken for warp routes - Shuffles around the aw multisig serialization stuff - now they're segmented by environment and context and are written to `config/{env}/aw-validators/{context}.json`. This was in response to some confusing merging logic, see https://discord.com/channels/935678348330434570/1228377321082327141/1229464197297868910 ### Drive-by changes Needed to specify pino to work in pretty mode for warp monitoring :/ this is necessary because pino doesn't work otherwise. After #3582 we should change this ### Related issues <!-- - Fixes #[issue number here] --> ### Backward compatibility <!-- Are these changes backward compatible? Are there any infrastructure implications, e.g. changes that would prohibit deploying older commits using this infra tooling? Yes/No --> ### Testing <!-- What kind of testing have these changes undergone? None/Manual/Unit Tests -->	8 months ago
Daniel Savu	f4626eb4fe	Allow protocol-specific indexing (#3544) ... ### Description - Generalizes contract syncer hashmaps to store trait objects instead of concrete structs, since both `SequencedDataContractSync` and `WatermarkContractSync` implement `ContractSyncer` - The nice thing this enables is being able to instantiate a different type of syncer for the same contract, for different chain protocols (e.g. for IGP, we can use a sequenced syncer on sealevel and cosmwasm, and a watermark syncer on EVM). - The problem is that `ContractSyncer` is implemented on different trait bounds on `SequencedDataContractSync` and `WatermarkContractSync`. Most notably, the `SequencedDataContractSync` impl bounds the indexed type by the `Sequenced` trait. This means that if a type is to be indexed by more than one syncer strategy, it needs to implement the union of bounds of those strategies - **Blocker:** In the IGP, this means that `InterchainGasPayment` needs to implement `Sequenced` and we have no obvious way of getting that just from the type. Even if the sequence is derived from the type, it has to be a sensible value, since indexing reliability depends on it. - Similar to the point above, we now have `contract_syncs`, a single function that instantiates contract syncs for all chains. Since `contract_syncs` must be able to build a `ContractSyncer` trait obj for all syncer types, it bounds both `T` (the indexed type) and `D` (the log store type) with a union of bounds, so it may be too strict for some types in the future. - The scraper now uses forward-backward indexing instead of just forward indexing, because that simplified the refactor - there's now a convenience trait called `TryFromWithMetrics` that allows adding this bound: `SequenceIndexer<T>: TryFromWithMetrics<ChainConf>,`. This isn't really extendable if we use non-sequence indexers, but it's fine for now since this is used everywhere - replaces the macro-based contract syncer builder with generic fns Still todo: - add new `IndexingDecorator<T>` struct that can wrap `InterchainGasPayment`, so we can use sequence-aware indexing for the IGP - remove sequence logic from the watermark cursor ### Drive-by changes <!-- Are there any minor or drive-by changes also included? --> ### Related issues <!-- - Fixes #[issue number here] --> ### Backward compatibility <!-- Are these changes backward compatible? Are there any infrastructure implications, e.g. changes that would prohibit deploying older commits using this infra tooling? Yes/No --> ### Testing <!-- What kind of testing have these changes undergone? None/Manual/Unit Tests -->	8 months ago
Trevor Porter	26a26f4073	Agent transaction overrides, ensure agent configs are updated in CI (#3563) ... ### Description - Adds some basic transaction overrides for Ethereum chains to be used by the agents. This follows the existing config schema in chain metadata that we have been using on the TS side for a while. The overrides are as follows: - `transactionOverrides.gasPrice` - if specified, non EIP-1559 txs are used, and they use this gasPrice - `transactionOverrides.gasLimit` - if specified, this takes precedence over any gas estimation that's done and is the gas limit that's used for txs. (This is pretty niche and I don't imagine we'll ever wanna use this internally, but happens to cover the use case described [here](https://discord.com/channels/935678348330434570/984123861144600587/1224617234580766741), which is nice to cover) - `transactionOverrides.maxFeePerGas` - if specified and if the chain supports EIP 1559 txs, is used for the `maxFeePerGas` - `transactionOverrides.maxPriorityFeePerGas` - if specified and if the chain supports EIP 1559 txs, is used for the `maxPriorityFeePerGas` - Updates some of the transaction overrides in mainnet3 / testnet4 to be more reasonable now that we'll be using them to submit transactions pretty frequently - Note that atm the transaction overrides are most easily applied to the agents via the config JSONs, and not via our K8s configuration. The upshot of this is if we wanna change a transaction override right now, the easiest way to do it will be to run `update-agent-config.ts` and then build a new image. Created https://github.com/hyperlane-xyz/hyperlane-monorepo/issues/3565 to track making this experience better - Part of this change means that it's more important to keep the `transactionOverrides` specified in the SDK / infra up to date with the agent config. So I ended up doing https://github.com/hyperlane-xyz/hyperlane-monorepo/issues/3325 along the way, which also required doing https://github.com/hyperlane-xyz/hyperlane-monorepo/issues/3311. - To fix the issue with arbitrum blocks, we now require the index.from to be specified in the chain metadata if the chain's technical stack is arbitrum nitro - A new script, `update-agent-config.ts` is added, and it's ran for mainnet3 and testnet4 in CI to make sure there's no diff (similar idea to what we do with prettier) - Along the way I renamed mainnet3_config.json and testnet4_config.json to mainnet_config.json and testnet_config.json. This has been a source of confusion with ppl in the past - in v2 we named them without any numeric suffixes, so we now just do that again ### Drive-by changes - There's a drive-by to poll more frequently for pending transactions (2 seconds). The default is 7 seconds in Ethers, and for many chains that have really quick block times, this may be a way for us to speed up us learning that a transaction is included - Don't enforce gas on the inevm routes, looks like there's some weirdness there that was reported by Nam ### Related issues - Fixes https://github.com/hyperlane-xyz/hyperlane-monorepo/issues/3562 - Fixes https://github.com/hyperlane-xyz/hyperlane-monorepo/issues/3325 - Fixes https://github.com/hyperlane-xyz/hyperlane-monorepo/issues/3311 ### Backward compatibility <!-- Are these changes backward compatible? Are there any infrastructure implications, e.g. changes that would prohibit deploying older commits using this infra tooling? Yes/No --> ### Testing <!-- What kind of testing have these changes undergone? None/Manual/Unit Tests -->	8 months ago
Noah Bayindirli 🥂	38358ececd	feat: deprecate Mumbai testnet (#3521) ... ### Description - Removes all reference to Mumbai testnet, since it is soon being [deprecated](https://polygon.technology/blog/polygon-pos-is-cooking-the-napoli-upgrade-means-better-ux-the-mumbai-testnet-takes-a-bow) by the Polygon team - Monorepo docker tag: https://console.cloud.google.com/gcr/images/abacus-labs-dev/global/hyperlane-monorepo@sha256:d4b44359cae9c4e3bdeb6f965c59d056f7fce72c42069adc0dea038ac600785a/details?authuser=5&project=abacus-labs-dev&tab=info - Agent docker tag: https://console.cloud.google.com/gcr/images/abacus-labs-dev/global/hyperlane-agent@sha256:a578d96f7f06721d0b6cd8095d6a1fa5d22f01b59eba3320ee3b1fe34ca3c5df/details?authuser=5&project=abacus-labs-dev&tab=info ### Drive-by changes - Adds missing Testnet domains, Fuji & BSC Testnet - Fix for failing `test-rs` CI check with ``` error[E0004]: non-exhaustive patterns: `chain::KnownHyperlaneDomain::Fuji` and `chain::KnownHyperlaneDomain::BinanceSmartChainTestnet` not covered --> hyperlane-core/src/utils.rs:226:15 | 226 | match $v { | ^^ patterns `chain::KnownHyperlaneDomain::Fuji` and `chain::KnownHyperlaneDomain::BinanceSmartChainTestnet` not covered ``` ### Related issues - Dependency on #3523 - Fixes #1181 ### Backward compatibility - [x] Backwards compatible - [x] Requires docker image tags to be added post-build - [x] Requires agent artifact cleanup in cloud account: https://www.notion.so/hyperlanexyz/Runbook-83c755f2652943289cf98cd1309487a8?pvs=4#77730653a12f409dacfbd427cab545d3 ### Testing * Build success * `yarn hyperlane chains list -e testnet` does not return `mumbai` * fork.sh script success locally	8 months ago
Daniel Savu	00fcaf74db	fix: warn on tx reverts; log cosmos tx results (#3497) ... ### Description <!-- What's included in this PR? --> ### Drive-by changes <!-- Are there any minor or drive-by changes also included? --> ### Related issues <!-- - Fixes #[issue number here] --> ### Backward compatibility <!-- Are these changes backward compatible? Are there any infrastructure implications, e.g. changes that would prohibit deploying older commits using this infra tooling? Yes/No --> ### Testing <!-- What kind of testing have these changes undergone? None/Manual/Unit Tests -->	8 months ago
Daniel Savu	6e39cf01af	Relayer retry endpoints (#3476) ... ### Description Adds a `/message_retry` relayer endpoint, which supports retrying OpQueue operations either by ID or by destination domain. Such a request will be sent to an `mpmc` channel's receiving end in the OpQueue, which is emptied when `OpQueue::pop` is called. Example calls: ``` GET http://127.0.0.1:60843/message_retry?destination_domain=42 GET http://127.0.0.1:60843/message_retry?message_id=0x46910b1329ee53c86a023b322e9ca1c17e5f9f0bee789c77b0abced0a173d714 ``` If the endpoint is called specifying both filters, **two** requests will be sent across the channel, one for each condition. Efficiency note: The entire queue is iterated over if there's at least one retry request in the channel. The good thing is that rebuilding the heap is `O(n)` so not too bad. ### Drive-by changes - Removed the usage of `enum_dispatch`, because it would have made unit tests messy. The drawback is that we have to work around object safety restrictions in `PendingOperation` - for instance it can't be cloned (but can be stored behind an `Arc` if we ever need this). - Added new methods to `PendingOperation`, either to match by ID, or to implement heap element prioritization without depending on the concrete `PendingMessage` type (as was done before). Additions: `id()`, `origin_domain()`, `priority()`. Don't think this conflicts with any operations we may add in the future (e.g. gas oracle updates). - Although `tokio`'s broadcast channel is multi-producer-multi-consumer by default, the consumer end isn't `Clone` - you instead need to have a producer to call `.subscribe()` on to get a new consumer, so I added an `MpmcChannel` struct to encapsulate keeping a producer around to get new consumers. - OpQueue is moved into its own file and has unit tests added for retries (covering broadcast to multi-queue, and both retry types). - Did a bit of agent `server `cleanup, including `EigenNodeApi` ### Related issues - Fixes https://github.com/hyperlane-xyz/hyperlane-monorepo/issues/3098 ### Backward compatibility Yes ### Testing Unit tests for the server route (channel transmitter logic) and for the OpQueue retries (channel receiver logic).	8 months ago
Daniel Savu	8ea0bde7c9	Use app context classifier in relayer submitter queues (#3385) ... ### Description Includes the `app_context` classification in `PendingMessage`, and adds trait methods on `PendingOperation` to require always having such a label on `OpQueue` operations. This is done by reusing the matching list logic from the validator checkpoint labels (https://github.com/hyperlane-xyz/hyperlane-monorepo/pull/3057). The nice thing is that this enables later support for retrying a group of `OpQueue` operations just by specifying the `app_context` label, without adding any new logic, since these labels are essentially matching list results. One downside to using `app_context` for retries is that the endpoint caller is constrained to only the matching lists defined by the relayer operator - however imo only the relayer operator that should be able to trigger retries. ### Drive-by changes The `OpQueue` type alias is converted to an actual struct, that stores the queue label (for metrics purposes), and also the `IntGaugeVec` metric: the generic group of metrics associated with that queue (basically only `submitter_queue_length` currently). ### Related issues - Fixes https://github.com/hyperlane-xyz/hyperlane-monorepo/issues/3240 ### Backward compatibility Yes ### Testing Manual, by spinning up a relayer for injective and inevm. Sample metrics, from `--metricAppContexts '[{"name": "injectivelabel", "matchingList": [{"destination_domain": 6909546 }] }, {"name": "inevmlabel", "matchingList": [{"destination_domain": 2525 }] }]'` ``` hyperlane_submitter_queue_length{agent="relayer",app_context="inevmlabel",hyperlane_baselib_version="0.1.0",queue_name="confirm_queue",remote="inevm"} 11 hyperlane_submitter_queue_length{agent="relayer",app_context="inevmlabel",hyperlane_baselib_version="0.1.0",queue_name="prepare_queue",remote="inevm"} 0 hyperlane_submitter_queue_length{agent="relayer",app_context="injectivelabel",hyperlane_baselib_version="0.1.0",queue_name="confirm_queue",remote="injective"} 63 hyperlane_submitter_queue_length{agent="relayer",app_context="injectivelabel",hyperlane_baselib_version="0.1.0",queue_name="prepare_queue",remote="injective"} 13281 ```	9 months ago
Paul Balaji	76bd730103	feat: remove goerli testnets (#3389)	9 months ago
Daniel Savu	ee1f11b4e4	fix(agents): add inevm to known domains and make igp indexing log war… (#3387) ... ### Description adds inevm to known domains and makes igp indexing log warn level ### Drive-by changes <!-- Are there any minor or drive-by changes also included? --> ### Related issues <!-- - Fixes #[issue number here] --> ### Backward compatibility <!-- Are these changes backward compatible? Are there any infrastructure implications, e.g. changes that would prohibit deploying older commits using this infra tooling? Yes/No --> ### Testing <!-- What kind of testing have these changes undergone? None/Manual/Unit Tests -->	9 months ago
Daniel Savu	dd8ac43c97	Validator task retries (#3361) ... ### Description - Makes validator tasks infallible by adding retries - This fixes a bug where certain tasks would return an error an shut down, affecting liveness. This was desired in the relayer since we don't want individual chain failures to affect the liveness of other chains. Now validator tasks either terminate or panic, and panicking will be propagated by `try_join_all`, causing the agent to shut down. - A thing to keep in mind in general is that agents will only terminate if a task panics. If it returns an `Err` but doesn't panic, the task won't be respawned. We should consider the implications of this in the scraper too. If this isn't desired, should consider using `select_all!` ### Drive-by changes - retry logic is moved from `rust/chains/hyperlane-cosmos/src/providers/rpc.rs` into `rust/hyperlane-core/src/rpc_clients/retry.rs`, so we're even closer to turning it into a retrying provider - changes several fn signatures within the validator to now return `ChainCommunicationError`s instead of `eyre::Report`s, for compatibility with the retry logic. Also makes `DbError` convertible to `ChainCommunicationError`, for the same reason. This achieves some progress on https://github.com/hyperlane-xyz/hyperlane-monorepo/issues/2878 - allows the backfill checkpoint submitter task to terminate, since `try_join_all` is tolerant of this (as described above) ### Related issues - Fixes https://github.com/hyperlane-xyz/hyperlane-monorepo/issues/3349 ### Backward compatibility <!-- Are these changes backward compatible? Are there any infrastructure implications, e.g. changes that would prohibit deploying older commits using this infra tooling? Yes/No --> ### Testing <!-- What kind of testing have these changes undergone? None/Manual/Unit Tests -->	9 months ago
Daniel Savu	e737d998ff	Cosmos block-by-block indexing (#3245) ... ### Description Implements cosmos block-by-block indexing, parallelizing where possible: the `block` and `block_results` calls, and the individual block queries within a range. Also reuses the existing `handle_txs` logic, to improve readability. ### Drive-by changes Adds `tokio::task::JoinError` to the `hyperlane-core` error enum ### Related issues - Fixes https://github.com/hyperlane-xyz/hyperlane-monorepo/issues/3234 - Added some hacky retrying logic to `get_logs_in_block`, since we were seeing lots of `503` http errors. In https://github.com/hyperlane-xyz/hyperlane-monorepo/issues/3264 we should make sure to remove this and propagate the error instead, to retry via the cursor / indexer logic. I edited the description of 3264 to reflect this ### Backward compatibility Yes ### Testing e2e --------- Co-authored-by: Trevor Porter <trkporter@ucdavis.edu>	9 months ago
Daniel Savu	2283f1b1d1	wip: deploy agents to plumetestnet (#3306) ... ### Description <!-- What's included in this PR? --> ### Drive-by changes <!-- Are there any minor or drive-by changes also included? --> ### Related issues <!-- - Fixes #[issue number here] --> ### Backward compatibility <!-- Are these changes backward compatible? Are there any infrastructure implications, e.g. changes that would prohibit deploying older commits using this infra tooling? Yes/No --> ### Testing <!-- What kind of testing have these changes undergone? None/Manual/Unit Tests --> --------- Co-authored-by: Trevor Porter <trkporter@ucdavis.edu> Co-authored-by: -f <kunalarora1729@gmail.com>	9 months ago
Trevor Porter	9a83eb4d57	Sequence-aware indexing refactor (#3262) ... ### Description Part 1 of an indexing refactor. This mostly focuses on the sequence-aware cursors. Changes: - Some renames, e.g. `SequenceIndexer` -> `SequenceAwareIndexer`, `sequence_and_tip` -> `latest_sequence_count_and_tip` - Moved cursors out of `contract_sync/cursor.rs` into `contract_sync/cursors/*` - The goal was to make the the Forward sequence-aware cursor, the Backward sequence-aware, and the ForwardBackward sequence aware cursors more resilient, easier to understand, and better tested. - General strategy for doing this was to: - Only move onto a new range to index if the logs passed into the cursor's `update` function justify this - If updated with any weird data, like gaps in logs, rewind the cursor and retry previously queried ranges I'd recommend starting the review in the following order: 1. `rust/hyperlane-base/src/contract_sync/cursors/sequence_aware/forward.rs` - where `ForwardSequenceAwareSyncCursor` & tests live 2. `rust/hyperlane-base/src/contract_sync/cursors/sequence_aware/backward.rs` - where `BackwardSequenceAwareSyncCursor` & tests live 3. `rust/hyperlane-base/src/contract_sync/cursors/sequence_aware/mod.rs` - where `ForwardBackwardSequenceAwareSyncCursor` lives ### Drive-by changes <!-- Are there any minor or drive-by changes also included? --> ### Related issues <!-- - Fixes #[issue number here] --> ### Backward compatibility <!-- Are these changes backward compatible? Are there any infrastructure implications, e.g. changes that would prohibit deploying older commits using this infra tooling? Yes/No --> ### Testing <!-- What kind of testing have these changes undergone? None/Manual/Unit Tests -->	10 months ago