This repository has been archived by the owner on Dec 7, 2023. It is now read-only.
🚀 Pulling upstream into main #23
Open
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…perlane-xyz#2646) ### Description just opening for image for now ### 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: Daniel Savu <23065004+daniel-savu@users.noreply.github.com>
…z#2543) ### Description This PR sets chain signer keys in the external secret. related to hyperlane-xyz#2458 ### 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: josedev-union <josedev-union@users.noreply.github.com> Co-authored-by: Nam Chu Hoai <nambrot@googlemail.com>
### Description Adds addresses of the rc validator keys for solana, nautilus and their testnet counterparts ### 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 -->
…yperlane-xyz#2647) ### Description This change adds the possibility to configure a subfolder inside an S3 bucket, to avoid having to spin up an entire bucket for hyperlane. ### Drive-by changes Nothing ### Related issues None, chatted about it on Slack with @nambrot ### Backward compatibility Yes I made the field optional in the `RawCheckpointSyncerConf` so that current config are still valid and will continue to read/write at the root of the bucket. That being said, I'm new to this code so I hope I got it right ### Testing None --------- Co-authored-by: Nam Chu Hoai <nambrot@googlemail.com>
### Description Process user used in Dockerfile can now read and write to any volume mounted under the `/data/` directory. ### Drive-by changes N/A ### Related issues Fixes: 2691 ### Backward compatibility Yes ### Testing Manual I've built the image locally and tested that we can mount any kind of docker volume under the `/data/` directory
### Description This does two things: 1) It parses the agent-specific configs for the relayer, validator, and scraper. 2) It removes the rigid structures that had been in place for parsing that were called out earlier and instead parses the serde json values directly. This PR does not cut over to the new versions just yet. That can happen once we start integrating it into the testing and deployment pipelines. ### Drive-by changes - Derive more was added and used where it made sense - The old config macro was removed ### Related issues - Progress on hyperlane-xyz#2215 ### Backward compatibility Yes ### Testing None
### Description Depends on hyperlane-xyz#2583 Indexes IGP payments related to the relayer's data pda address. Unless this address is specified in the config (`sealevel.relayer_account `), no filtering is applied and all IGP payments are stored in the local database. <!-- What's included in this PR? --> ### Drive-by changes - Sets `HYP_BASE_GASPAYMENTENFORCEMENT` in `run-locally` for the relayer, to test that it correctly indexes the IGP payment before submitting the message - A new config section (`sealevel`) is added to the relayer - The `MessageIndexer` trait is replaced with `SequenceIndexer<HyperlaneMessage>`, renaming `fetch_count_at_tip` to `sequence_at_tip`. `SequenceIndexer` is now common to both the message and igp indexers. - The `parse_addr` macro is modified so it can be reused when parsing the sealevel relayer address config too - `rust/utils/sealevel-test.bash` is included because I was using it to test locally, but I can remove it if the sealevel e2e test already does all the steps there @mattiecnvr - Performs a `try_into` conversion that can be removed once hyperlane-xyz#2610 is done ### Related issues - Fixes hyperlane-xyz#2501 ### 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? --> e2e tests but the pipeline is failing, likely fixed by hyperlane-xyz#2602 --------- Co-authored-by: Trevor Porter <trkporter@ucdavis.edu>
…p transfers in tooling (hyperlane-xyz#2700) ### Description Sadly, there's no programmatic way to submit transactions to Squads via an API or anything. As a workaround, if `-k` is passed a Pubkey and not a path to a Keypair as the payer, transactions will instead be logged in a base58 serialized format that can then be copied into Squads. Instructions for exactly how to do this are found in our Notion So this includes: * Allow for the payer to just be a Pubkey without a Keypair * Some commands & functions to allow for ownership transfer ### Drive-by changes * There was a whole directory `rust/chains/hyperlane-sealevel/src/solana` that somehow got into main, I deleted this * Cleaned up some logs to be formatted more nicely ### Related issues part of hyperlane-xyz/issues#545 ### Backward compatibility backward compatible ### Testing Transferred ownership of devnet --------- Co-authored-by: Daniel Savu <23065004+daniel-savu@users.noreply.github.com>
### Description - Add support for multi-protocol to the HelloWorld app and Kathy service - Create a new MultiProtocolCore ### Related issues hyperlane-xyz#2503 ### Backward compatibility Yes ### Testing - Tested new Kathy abstraction on existing EVM while we wait for Sealevel Kathy to be unblocked - Tested using local config stubs for both EVM and Sealevel
### Description - Migrating `SealevelTokenAdapter` to monorepo for providing a common abstraction - Common serialization code under `sdk/sealevel` ### Related issues - hyperlane-xyz#2652 ### Backward compatibility Yes ### Testing Manual --------- Co-authored-by: J M Rossy <jm.rossy@gmail.com>
### Description - Script to monitor token balances for Nautilus-Zebec-Solana warp route (hardcoded) - Helm command to deploy warp route monitoring script to Kubernetes ### Drive-by changes - Updated default version to testnet3 for kathy and liquidity-layer ### Related issues - issue hyperlane-xyz/issues#549 ### Backward compatibility Yes ### Testing Manual
### Description - Currently if you do keyboard interrupt while running the deploy script, none of the already deployed addresses get written to the artifacts. - Register a SIGINT handler in the `deployWithArtifacts` script ### Drive-by changes None ### Related issues None ### Backward compatibility Yes ### Testing Manual
### Description Tooling to get / set gas oracle data for sealevel, part of hyperlane-xyz/issues#568 The PR adds two IGP oracle commands. If the values to be set are specified, then the command performs a write using the key configured. Otherwise, it performs a read. Unsure whether this is the best UX, happy to change it. Usage example: ```bash # Get gas values cargo run --bin hyperlane-sealevel-client igp destination-gas-overhead --environments-dir ./sealevel/environments --environment testnet3 --chain-name solanadevnet --remote-domain 88002 # Set gas values. Note the added `--gas-overhead 159736` flag cargo run --bin hyperlane-sealevel-client igp destination-gas-overhead --environments-dir ./sealevel/environments --environment testnet3 --chain-name solanadevnet --remote-domain 88002 --gas-overhead 159736 ``` There's also a small warp route util added to read the destination gas: ```bash cargo run --bin hyperlane-sealevel-client warp-route destination-gas --program-id 84DAG9VX5CvefQrBUGh9Gp66kmkat2voaM4MB2YwpGW1 --destination-domain 88002 ``` <!-- 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 Manual <!-- What kind of testing have these changes undergone? None/Manual/Unit Tests -->
### Description Fixes a current deployment issue due to some changes in the infra deployment. Also caught an error in the schema for the matching list type and updated it.
Removes some unnecessary logs from the sealevel igp indexing fn. Also instruments the eta calculator in response to hyperlane-xyz#2723 ### 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 -->
### 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 -->
### Description - Build on hyperlane-xyz#2684, migrates the EvmAdapter and missing utils from the warp UI - Remove the token App as it's unused and redundant with adapters - Add IGP serialization code for hyperlane-xyz#2546 - Simplify multi-protocol adapters in SDK - Make token adapters extend base adapters - Move token serialization code to token package ### Drive-by changes - Update Sepolia RPC which was causing errors ### Related issues hyperlane-xyz#2652 ### Backward compatibility Yes ### Testing Tested these in warp UI --------- Co-authored-by: Yorke Rhodes <yorke@hyperlane.xyz>
Simplify mailbox and make interfaces payable Update version constant Keep message ID events
Use mailbox callback to authenticate messages in hooks where necessary (hyperlane-xyz#2609) Co-authored-by: Kunal Arora <55632507+aroralanuk@users.noreply.github.com>
- IGP as a standalone hook, implementing postDispatch to call payForGas directly - Setting a DEFAULT_GAS_USAGE if metadata not specified and message.senderAddress() as refund address if not specified. - None Fixes hyperlane-xyz/issues#511 Yes, same interface as the previous IGP but for Mailbox V3 Unit Tests --------- Co-authored-by: Yorke Rhodes <yorke@hyperlane.xyz>
…-xyz#2632) - Updated the OP Stack tests for the Mailbox V3 transient storage version - Allowing OPStackHook to send msg.value at the time of message delivery (uses bit masking) - Added LibBit library, will be useful for all the auth hooks which can send `msg.value` - None - Fixes breaking OP Stack tests for hyperlane-xyz/issues#513 - Also fixes hyperlane-xyz#2410 No Unit tests --------- Co-authored-by: Yorke Rhodes <yorke@hyperlane.xyz> Remove e2e and add yarn build to CI Adding protocol fees (hyperlane-xyz#2640) - Adding protocol fee as a hook - None V3 Yes Fuzz tests --------- Co-authored-by: Yorke Rhodes <yorke@hyperlane.xyz>
- fixes GasRouter expectRevert message - added TestMerkleRootHook for proof() and fixes MerkleRootMultisig test - fixes ERC5164 tests post v3 changes - added contract name to mailbox and abstractHook error messages - removed unnecessary tests for "message too large" - downgrade slither in actions to 0.3.0 because of their recent "missing inheritance" bug on main - V3 Yes Unit tests
- `quoteDispatch` added to `IPostDispatchHook` interface and the
relevant hooks:
- StaticProtocolFee
- OPStackHook
- InterchainGasPaymaster
- MerkleTreeHook
- PausableHook (no tests)
- DomainRoutingHook (no tests)
- ConfigFallbackDomainRoutingHook
- ConfigurableDomainRoutingHook (no tests)
- `expectEmit` -> `expectCall`
- Quote in V3
Yes
Uint tests
### Description - AggregationHook for v3 ### Drive-by changes None ### Related issues v3 hyperlane-xyz/issues#514 ### Backward compatibility Yes ### Testing Always be fuzzing
### Description This PR adds a slightly better error message if there is no `./configs` folder for the CLI invocation
### Description deploys agents with new release ### Drive-by changes n/a ### Related issues n/a ### Backward compatibility ye ### Testing deployed
### Description Update to latest ethers-rs (forked) version with hyperlane-xyz/ethers-rs#9 ### Drive-by changes n/a ### Related issues n/a ### Backward compatibility n/a ### Testing Planning to deploy with it, but the ethers-rs version is well tested
### Description Deploys w/ hyperlane-xyz#2998 ### Drive-by changes - Had to remove mantapacific from the list of scraper chains because it's not present in the DB domain table atm ### Related issues n/a ### Backward compatibility n/a ### Testing Deployed
### Description Updates the values given the current market conditions ### Drive-by changes Add a sleep for avoid getting rate limited
### Description Uses a version of our ethers fork with hyperlane-xyz/ethers-rs#10, which adds back a change that I had reverted from hyperlane-xyz/ethers-rs#8 This is required to fix bsc / bsctestnet txs, where the base fee is 0 and we want to estimate the priority fee ### Drive-by changes n/a ### Related issues n/a ### Backward compatibility ye ### Testing deployed and saw successful bsc / bsctestnet msgs
### Description - fix to filter out igp in merging the given and SDK artifacts ### Drive-by changes - None ### Related issues - None ### Backward compatibility Yes ### Testing Manual --------- Co-authored-by: Yorke Rhodes <yorke@hyperlane.xyz>
…fact (hyperlane-xyz#2993) ### Description - changing `multisigIsm` to`interchainSecurityModule` for core deployments artifact ### 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 -->
### Description Add artifact selection step to warp deploy Improve chain selection for kurtosis agent deploy ### Drive-by changes Add chain metadata consts validation unit test ### Related issues Fixes hyperlane-xyz/issues#774 ### Backward compatibility Small break in kurtosis deploy args, likely won't be noticed. ### Testing Manual
hyperlane-xyz#3005) ### Description - CLI was using a static start block numbers for core chains or getting the latest block number from a PI chain for agent config which is redundant and dangerous. Instead, I updated to using the `mailbox.deployedBlock()` which should precede all other indexable contract deployment and is hence safer. ### Drive-by changes - filtering the agent configs in CLI is redundant since the `HyperlaneDeploymentArtifactsSchema` requires all the specified entries and if `writeAgentConfig` gets an artifacts which doesn't contain these, we should throw an error and not filter them. ### Related issues - related to hyperlane-xyz/issues#736 ### Backward compatibility Yes ### Testing Manual b/w anvil1 and anvil2
### Description Deploying with new release ### 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 -->
…hains to match addresses in MultiProtocolApp (hyperlane-xyz#3001) ### Description Fixes 2 issues: 1. Estimates gas in Kathy by explicitly specifying the `from` address due to this bug with PolygonZkEvm when using a non-zero value 0xPolygonHermez/zkevm-node#2869 2. Now that we've added neutron & mantapacific as mainnet chains but we don't have helloworld deployments on these chains, Kathy was trying to send to & from these chains. To fix, I changed the constructor of `MultiProtocolApp` to intersect the multiProvider to only work with the chains specified in `addresses` ### Drive-by changes n/a ### Related issues n/a ### Backward compatibility ye ### Testing Ran kathy locally successfully & sent from polygonzkevm
### Description Deploy Kathy with hyperlane-xyz#3001 ### Drive-by changes Small fix to deploy kathy successfully - because the relayer chains doesn't include neutron and kathy will try to create a multiprovider for all chains the env, we need to use the environment chain names as the helm `hyperlane.chains` value ### 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 -->
) ### Description Turns out bridging from L1 to Base has been broken the whole time :( it's because the version of @eth-optimism/sdk we were using was 1.8.0 (according to yarn.lock) which was released in late 2022 https://github.com/ethereum-optimism/optimism/releases/tag/%40eth-optimism%2Fsdk%401.8.0, far before Base was launched. So we'd always get this error in key funder because we rely on the optimism SDK knowing about Base's chain ID: ``` {"chain":"base","error":"Error: cannot get contract AddressManager for unknown L2 chain ID 8453, you must provide an address\n at getOEContract (/hyperlane-monorepo/node_modules/@eth-optimism/sdk/src/utils/contracts.ts:58:11)\n at getAllOEContracts (/hyperlane-monorepo/node_modules/@eth-optimism/sdk/src/utils/contracts.ts:121:46)\n at new CrossChainMessenger (/hyperlane-monorepo/node_modules/@eth-optimism/sdk/src/cross-chain-messenger.ts:170:39)\n at ContextFunder.bridgeToOptimism (/hyperlane-monorepo/typescript/infra/scripts/funding/fund-keys-from-deployer.ts:652:33)\n at ContextFunder.bridgeToL2 (/hyperlane-monorepo/typescript/infra/scripts/funding/fund-keys-from-deployer.ts:637:23)\n at processTicksAndRejections (node:internal/process/task_queues:96:5)\n at async ContextFunder.bridgeIfL2 (/hyperlane-monorepo/typescript/infra/scripts/funding/fund-keys-from-deployer.ts:492:9)\n at async gracefullyHandleError (/hyperlane-monorepo/typescript/infra/scripts/funding/fund-keys-from-deployer.ts:774:5)\n at async /hyperlane-monorepo/typescript/infra/scripts/funding/fund-keys-from-deployer.ts:394:29\n at async Promise.all (index 9)","level":"error","message":"Error bridging to L2"} ``` The fix was just to upgrade to a newer optimism SDK version A mystery to me is that we'd get that log ^ but sometimes the key funder pod would show as having ran successfully. My best guess here is there was a race condition in `fund` when we had a variable local to the function called `failureOccurred` that many promises which were `Promise.all`'d would read and write to it via `failureOccurred ||= await someFallibleFn()`. I changed the logic here to not have multiple concurrent promises contend for the variable Also made a small change to not include mantapacific in the list of relayer keys for the Hyperlane context. It's not ever used, and had a downstream effect of us trying to fund the Kathy key on mantapacific, which we don't want to actually do ### Drive-by changes n/a ### Related issues n/a ### Backward compatibility ye ### Testing ran locally
### Description - enable configuring hooks in the CLI (merkle, igp, protocolFee, aggregation, routing) - use preset by default without prompting ### 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 -->
### Description https://eslint.org/docs/latest/rules/guard-for-in ### Drive-by changes Fix lint error in SDK Sealevel adapter ### Backward compatibility Yes
…xyz#3016) ### Description - CI fails otherwise ### 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 Manual
### Description - Improve shared context utility with conditional typing and core artifact handling - Improve UX of send and status commands with prompts - Prompt for key if required and not provided ### Drive-by Changes Update the token readme for hyperlane-xyz/issues#715 ### Backward compatibility Yes ### Testing Manual
) Fix for missing bit of hyperlane-xyz#3006
…-xyz#3021) ### Description - don't restrict user to having two chains for ism config - if the user accidentally picks two chains, we prompt them again to confirm if they don't want to use the hyperlane validators for their multisigconfig ### Drive-by changes None ### Related issues - fixes hyperlane-xyz/issues#811 ### Backward compatibility Yes ### Testing Manual between arbitrumgoerli and anvil1
Done: - scaffolding for fetching custom agent metrics - abstractions for building a metrics fetcher for a given VM - querying cosmos balances; e2e tested. - querying evm balances; e2e tested. - **Note that as a result, evm addresses are now no longer zero-padded when printed in the logs. This may break existing log queries** - fixed a nasty bug on ubuntu where wasmd (osmosisd dependency, part of the grpc query flow) would panic when a block is specified via `x-cosmos-block-height`. The fix was to bump the version of osmosisd from `19.0.0` to `20.5.0`. **Note that when running e2e on Mac OS, the osmosis version in use is still 19.0.0**. That's because we need a fork that publishes a darwin target binary (currently pointing [here](https://github.com/hashableric/osmosis/releases/download/v19.0.0-mnts/osmosisd-19.0.0-mnts-darwin-arm64.tar.gz)) For follow up PR: - sealevel balance querying I'm open to all renaming suggestions, I just tried to speed through and didn't ponder names too much Relates to hyperlane-xyz/issues#701 Closes hyperlane-xyz/issues#702 (because the balance becomes available in the metrics endpoint for polling)
Publishes sealevel balance as a relayer metric **Drive-by Changes** - Creates `HyperlaneSealevelError`, a `hyperlane-sealevel` specific type to avoid including sealevel errors in `hyperlane-core`'s `ChainCommunicationError`. It's pretty empty now, but we have a starting point now. This follows the approach used in `hyperlane-cosmos` too. - Includes an rpc provider instance in `SealevelProvider`, and uses it to add `get_balance` on `SealevelProvider`. **Testing** None, as sealevel e2e is currently disabled
There are way too many `No message found in DB for leaf index` logs to have them at debug level. Also the wording makes it seem like an error when it's not.
# Releases ## @hyperlane-xyz/cli@3.3.0 ### Minor Changes - 7e620c9: Allow CLI to accept hook as a config ### Patch Changes - f44589e: Improve warp and kurtosis deploy command UX - 2da6cce: Allow users to only configure validators for their chain - Don't restrict user to having two chains for ism config - If the user accidentally picks two chains, we prompt them again to confirm if they don't want to use the hyperlane validators for their multisigConfig - 9f2c7ce: Removing agentStartBlocks and using mailbox.deployedBlock() instead - 9705079: Improve UX of the send and status commands - c606b6a: Add figlet to CLI - Updated dependencies [7e620c9] - Updated dependencies [3501755] - Updated dependencies [9f2c7ce] - @hyperlane-xyz/sdk@3.3.0 - @hyperlane-xyz/utils@3.3.0 ## @hyperlane-xyz/core@3.3.0 ### Patch Changes - 3501755: Rename StaticProtocolFee hook to ProtocolFee for clarity - @hyperlane-xyz/utils@3.3.0 ## @hyperlane-xyz/helloworld@3.3.0 ### Patch Changes - Updated dependencies [7e620c9] - Updated dependencies [3501755] - Updated dependencies [9f2c7ce] - @hyperlane-xyz/sdk@3.3.0 - @hyperlane-xyz/core@3.3.0 ## @hyperlane-xyz/sdk@3.3.0 ### Patch Changes - 7e620c9: Allow CLI to accept hook as a config - 3501755: Rename StaticProtocolFee hook to ProtocolFee for clarity - 9f2c7ce: Removing agentStartBlocks and using mailbox.deployedBlock() instead - Updated dependencies [3501755] - @hyperlane-xyz/core@3.3.0 - @hyperlane-xyz/utils@3.3.0 ## @hyperlane-xyz/utils@3.3.0 ## @hyperlane-xyz/infra@3.3.0 ### Patch Changes - 7e620c9: Allow CLI to accept hook as a config - 9f2c7ce: Removing agentStartBlocks and using mailbox.deployedBlock() instead - Updated dependencies [7e620c9] - Updated dependencies [3501755] - Updated dependencies [9f2c7ce] - @hyperlane-xyz/sdk@3.3.0 - @hyperlane-xyz/helloworld@3.3.0 - @hyperlane-xyz/utils@3.3.0 Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
…hyperlane-xyz#3009) ### Description - `DefaultFallbackRoutingIsm` needs the mailbox address which meant `DefaultFallbackRoutingIsmFactory` needed the mailbox address which is not ideal. So, I switched to non-factory deployments for this specific ISM type. - CLI does the ISM deployment inside core deployment instead of ISM first and core then. - Using the coreDeployer's `cachedAddresses` to store the latest ie toplevel ISM. ### Drive-by changes - fixed bug in `ismFactory.fromAddressMap` to not use the multiprovider filtered by contract map. This is undesirable as CLI user can provide chain selection and artifacts which doesn't include all the chains in the configs (and when we call `multiprovider.getDomainId` for the origin which may now be missing from the filtered MP, we get an error). ### Related issues - fixes hyperlane-xyz#2895 ### Backward compatibility Yes ### Testing Manual through CLI
### Description
* Remove the `bank` role, which we haven't used since the inception of
abacus / hyperlane
* Big changes to `key-utils.ts` so that there's a single source of truth
on what kind of keys are used depending on the role & chain. Before this
was sprinkled in a few different places
* You can now get an object of `{ [chain]: { [role]: keys[] } }`, so
it's super clear what kind of key relates to which chain. For example,
before we would use the AWS-based relayer key for EVM chains, and then a
GCP-based relayer key for non-EVM chains. But this wasn't really honored
by key funder - it had no way of knowing to only fund the AWS relayer on
EVM chains, and only fund the GCP relayer on non-EVM chains. Same
situation for Kathy, where we want to use AWS keys for EVM chains but
the GCP key for non-EVM chains
* On v2 and prior to that we were using the AWS-based key for Kathy.
Originally, we also launched v3 this way. However it was changed on v3
to use the GCP key for Kathy, causing us to fund both types of addresses
on v3. This makes it more clear that we should be using the AWS-based
key for Kathy on EVM chains
### 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
-->
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
An automated PR to sync with upstream
Co-authored-by: josedev-union josedev-union@users.noreply.github.com