Skip to content

Fix Docker build cache issue for deposit contract files #8488

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Closed
jimmygchen wants to merge 1 commit into from
Closed

Fix Docker build cache issue for deposit contract files #8488

jimmygchen wants to merge 1 commit into from

Conversation

@jimmygchen
Copy link
Member

@jimmygchen jimmygchen commented Nov 28, 2025 •
edited
Loading

Description

Fixes Docker build failures when using cache mounts. The build was failing with errors like:

error: couldn't read `common/deposit_contract/src/../contracts/v0.12.1_validator_registration.json`: No such file or directory (os error 2)
  --> common/deposit_contract/src/lib.rs:47:24

The deposit contract crate uses a build script (build.rs) to download contract ABI and bytecode files from GitHub, storing them in common/deposit_contract/contracts/. These files are then included at compile time via include_bytes!().

The Docker cache mount for /lighthouse/target caused Cargo to skip re-running the build script on subsequent builds, but the generated contract files weren't persisted in the cache. This resulted in compilation failures because include_bytes!() couldn't find the required files.

Workaround right now is to run docker builder prune -a -f

The fix adds a cache mount for the contracts directory, ensuring generated files persist across builds.

@jimmygchen
Fix Docker build cache issue for deposit contract files
a8d574b 
  The deposit contract build script downloads files and stores them in
  common/deposit_contract/contracts/, but these generated files were not
  persisted when using Docker cache mounts for the target directory.

  When Docker reused cached build artifacts, the contract files were
  missing, causing build failures with include_bytes!() errors.

  This fix adds a cache mount for the contracts directory, ensuring the
  generated files persist across builds alongside other build artifacts.
  When the TAG constant in build.rs changes, the Docker COPY layer
  invalidates and the cache rebuilds correctly.
@jimmygchen jimmygchen added ready-for-review The code is ready for review devex labels Nov 28, 2025
@michaelsproul
Copy link
Member

michaelsproul commented Nov 28, 2025

I think the correct fix is for the build script to put the files in target. Build scripts aren't meant to write to the source directory

@michaelsproul michaelsproul mentioned this pull request Dec 2, 2025
Merged
@jimmygchen
Copy link
Member Author

jimmygchen commented Dec 3, 2025

will go with #8518 instead which implements what @michaelsproul suggested above, it's a much better approach.

@jimmygchen jimmygchen closed this Dec 3, 2025
@jimmygchen jimmygchen deleted the fix-docker-build-cache branch December 3, 2025 03:22
mergify bot pushed a commit that referenced this pull request Dec 3, 2025
@michaelsproul
Move deposit contract artifacts to /target (#8518)
41ba135 
Alternative to:

- #8488


  Refactors deposit_contract crate to comply with Rust build conventions by placing generated artifacts in the build output directory.


Co-Authored-By: Michael Sproul <[email protected]>
kevaundray added a commit to eth-act/lighthouse that referenced this pull request Dec 13, 2025
@kevaundray @jimmygchen
@0xMushow @chong-he @michaelsproul @macladson @eserilev @pawanjay176
Merge unstable into optional-proofs (#2)
0fe4341 
* Update local testnet scripts for the fulu fork (sigp#8489)

* Remove `fulu-devnet-3` testing on CI
* Delete `scripts/local_testnet/network_params_das.yaml` and consolidate it into the main `network_params.yaml` file we use on CI
* Delete enclave before building image, so it doesn't cause slow image building.


  


Co-Authored-By: Jimmy Chen <[email protected]>

* Fix data columns sorting when reconstructing blobs (sigp#8510)

Closes sigp#8509


  


Co-Authored-By: Antoine James <[email protected]>

* Instrument attestation signing. (sigp#8508)

We noticed attestation signing taking 2+ seconds on some of our hoodi nodes and the current traces doesn't provide enough details. This PR adds a few more spans to the `attestation_duty_cycle` code path in the VC.

Before:
<img width="842" height="154" alt="image" src="https://github.com/user-attachments/assets/cfc5c8c0-e6f2-4f56-a8e4-65001af4a005" />

After:
<img width="496" height="217" alt="image" src="https://github.com/user-attachments/assets/c91cfa88-af1b-456e-8c64-625809eb6881" />


  


Co-Authored-By: Jimmy Chen <[email protected]>

* Always use committee index 0 when getting attestation data (sigp#8171)

* sigp#8046


  Split the function `publish_attestations_and_aggregates` into `publish_attestations` and `handle_aggregates`, so that for attestations, only 1 task is spawned.


Co-Authored-By: Tan Chee Keong <[email protected]>

Co-Authored-By: chonghe <[email protected]>

Co-Authored-By: Michael Sproul <[email protected]>

Co-Authored-By: Michael Sproul <[email protected]>

* Move deposit contract artifacts to /target (sigp#8518)

Alternative to:

- sigp#8488


  Refactors deposit_contract crate to comply with Rust build conventions by placing generated artifacts in the build output directory.


Co-Authored-By: Michael Sproul <[email protected]>

* Move beacon state endpoints to a separate module. (sigp#8529)

Part of the http api refactor to move endpoint handlers to separate modules.

This should improve code maintainability, incremental compilation time and rust analyzer performance.


  


Co-Authored-By: Jimmy Chen <[email protected]>

* Refactor `consensus/types` (sigp#7827)

Organize and categorize `consensus/types` into modules based on their relation to key consensus structures/concepts.
This is a precursor to a sensible public interface.

While this refactor is very opinionated, I am open to suggestions on module names, or type groupings if my current ones are inappropriate.


Co-Authored-By: Mac L <[email protected]>

* Move validator http endpoints to a separate module (sigp#8536)

Continuation of:
* sigp#8529

Moving `/validator` endpoints out of `http_api` to a separation module. This should improve code maintainability, incremental compilation time and rust analyzer performance.

This is a tedious but straight forward change, so we're going with a pair & insta-merge approach to avoid painful & slow async review. @michaelsproul and I paired on the first commit - I believe we are almost done, will pair with @pawanjay176 tomorrow to wrap it up and merge tomorrow. (cc @macladson )


  


Co-Authored-By: Jimmy Chen <[email protected]>

* Move beacon pool http api to its own separate module (sigp#8543)

Continuation of:
* sigp#8536

Moving `/beacon/pool` endpoints out of `http_api` to a separation module. This should improve code maintainability, incremental compilation time and rust analyzer performance.

This is a tedious but straight forward change, so we're going with a pair & insta-merge approach to avoid painful & slow async review


  


Co-Authored-By: Jimmy Chen <[email protected]>

* Reduce `eth2` dependency space  (sigp#8524)

Remove certain dependencies from `eth2`, and feature-gate others which are only used by certain endpoints.

| Removed | Optional | Dev only |
| -------- | -------- | -------- |
| `either` `enr` `libp2p-identity` `multiaddr` | `protoarray` `eth2_keystore` `eip_3076` `zeroize` `reqwest-eventsource` `futures` `futures-util` | `rand` `test_random_derive` |

This is done by adding an `events` feature which enables the events endpoint and its associated dependencies.
The `lighthouse` feature also enables its associated dependencies making them optional.

The networking-adjacent dependencies were removed by just having certain fields use a `String` instead of an explicit network type. This means the user should handle conversion at the call site instead. This is a bit spicy, but I believe `PeerId`, `Enr` and `Multiaddr` are easily converted to and from `String`s so I think it's fine and reduces our dependency space by a lot. The alternative is to feature gate these types behind a `network` feature instead.


Co-Authored-By: Mac L <[email protected]>

* Clarify `alloy` dependencies (sigp#8550)

Previously, we had a pinned version of `alloy` to fix some crate compatibility issues we encountered during the migration away from `ethers`. Now that the migration is complete we should remove the pin. This also updates alloy crates to their latest versions.


Co-Authored-By: Mac L <[email protected]>

* Remove `consensus/types` re-exports (sigp#8540)

There are certain crates which we re-export within `types` which creates a fragmented DevEx, where there are various ways to import the same crates.

```rust
// consensus/types/src/lib.rs
pub use bls::{
AggregatePublicKey, AggregateSignature, Error as BlsError, Keypair, PUBLIC_KEY_BYTES_LEN,
PublicKey, PublicKeyBytes, SIGNATURE_BYTES_LEN, SecretKey, Signature, SignatureBytes,
get_withdrawal_credentials,
};
pub use context_deserialize::{ContextDeserialize, context_deserialize};
pub use fixed_bytes::FixedBytesExtended;
pub use milhouse::{self, List, Vector};
pub use ssz_types::{BitList, BitVector, FixedVector, VariableList, typenum, typenum::Unsigned};
pub use superstruct::superstruct;
```

This PR removes these re-exports and makes it explicit that these types are imported from a non-`consensus/types` crate.


Co-Authored-By: Mac L <[email protected]>

* Fix testnet script (sigp#8557)

Fix an issue where a kurtosis testnet script was failing because no supernodes were provided


```
There was an error interpreting Starlark code
Evaluation error: fail: Fulu fork is enabled (epoch: 0) but no supernodes are configured, no nodes have 128 or more validators, and perfect_peerdas_enabled is not enabled. Either configure a supernode, ensure at least one node has 128+ validators, or enable perfect_peerdas_enabled in network_params with 16 participants.
at [github.com/ethpandaops/ethereum-package/main.star:83:57]: run
at [github.com/ethpandaops/ethereum-package/src/package_io/input_parser.star:377:17]: input_parser
at [0:0]: fail
```


  


Co-Authored-By: Eitan Seri-Levi <[email protected]>

Co-Authored-By: Pawan Dhananjay <[email protected]>

* Do not request attestation data when attestation duty is empty (sigp#8559)

Co-Authored-By: Tan Chee Keong <[email protected]>

* Rust 1.92 lints (sigp#8567)

Co-Authored-By: Eitan Seri-Levi <[email protected]>

---------

Co-authored-by: Jimmy Chen <[email protected]>
Co-authored-by: Jimmy Chen <[email protected]>
Co-authored-by: 0xMushow <[email protected]>
Co-authored-by: Antoine James <[email protected]>
Co-authored-by: chonghe <[email protected]>
Co-authored-by: Michael Sproul <[email protected]>
Co-authored-by: Michael Sproul <[email protected]>
Co-authored-by: Mac L <[email protected]>
Co-authored-by: Eitan Seri-Levi <[email protected]>
Co-authored-by: Pawan Dhananjay <[email protected]>
Co-authored-by: Tan Chee Keong <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

devex ready-for-review The code is ready for review

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@jimmygchen @michaelsproul