Skip to content

feat(docker): bake gouda.genesis.json; bump to gouda chains#105

Open
rswanson wants to merge 2 commits into
mainfrom
feat/gouda-image
Open

feat(docker): bake gouda.genesis.json; bump to gouda chains#105
rswanson wants to merge 2 commits into
mainfrom
feat/gouda-image

Conversation

@rswanson
Copy link
Copy Markdown
Member

@rswanson rswanson commented May 18, 2026

Changes

  • signet-sidecar deps bumped to crates.io releases
    • init4-bin-base 0.21 → 0.23.1
    • signet-blobber, signet-host-rpc, signet-node, signet-node-config, signet-rpc 0.18.0 → 0.19.0
  • Add .github/genesis/gouda.genesis.json (mirrored from upstream node-components) and bake it into all three Dockerfiles at /network_configs/gouda.genesis.json, matching the existing parmigiana pattern.

Verification

  • cargo +nightly fmt --check
  • cargo clippy --workspace --all-targets -- -D warnings
  • RUSTDOCFLAGS="-D warnings" cargo doc --workspace --no-deps
  • cargo test -p signet-sidecar

@Evalir
Copy link
Copy Markdown
Member

Evalir commented May 18, 2026

Stale gouda.genesis.json mirror — wrong host.startTimestamp

.github/genesis/gouda.genesis.json at line 54 still has "startTimestamp": 1779051536 (the gouda-on-host deploy time). The canonical file in node-components was fixed to 1765226348 (the parmigiana host genesis anchor) in 538b2f6"gouda.genesis.json: host.startTimestamp 1779051536 -> 1765226348 (parmigiana host genesis anchor — gouda runs on parmigiana host)".

The mirror was added once in 62c258b and never re-synced, even though deps were bumped three times after that (dca13d1, 89e521b, 0de3da6). Three independent sources, two values:

Source host.startTimestamp
signet @ feat/gouda-image:.github/genesis/gouda.genesis.json 1779051536 ⚠️
node-components @ feat/gouda-genesis:crates/genesis/src/gouda.genesis.json 1765226348
signet-sdk @ feat/gouda-known-chain:chains/gouda.rs::HOST_START_TIMESTAMP 1765226348

Impact: the documented CHAIN_NAME=gouda path is safe — it uses the embedded GOUDA_GENESIS_JSON from the binary, which was built against the fixed node-components rev 752fa4a. The trap is GenesisSpec::Custom reading from disk via ROLLUP_GENESIS_JSON_PATH=/network_configs/gouda.genesis.json (the baked file): that path hits the wrong slot-calc anchor. Anyone copying the baked file as a "reference" inherits the same bug.

Fix: re-mirror from node-components/crates/genesis/src/gouda.genesis.json, or drop the baking entirely if the embedded const is intended as the only source of truth.

…nesis.json

signet-sidecar moves to fully crates.io-published versions:
- init4-bin-base 0.21 -> 0.23.1
- signet-blobber / signet-host-rpc / signet-node /
  signet-node-config / signet-rpc 0.18.0 -> 0.19.0

Adds .github/genesis/gouda.genesis.json (mirrored from upstream
node-components) and bakes it into all three Dockerfiles at
/network_configs/gouda.genesis.json.

signet/ and signet-exex/ unchanged: their reth and signet-host-reth
deps remain git-only upstream (reth not published by paradigmxyz;
signet-host-reth marked publish = false in node-components).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
@rswanson rswanson force-pushed the feat/gouda-image branch from 0de3da6 to 37e52ef Compare May 21, 2026 19:45
…pstream

Local mirror was at 1779051536 (gouda-on-host deploy time, never the
slot-calc anchor); upstream node-components has 1765226348 (parmigiana
host genesis anchor — gouda runs on parmigiana host). The mirror was
added at 62c258b and never re-synced through three subsequent dep bumps.

The embedded GOUDA_GENESIS_JSON in the binary is unaffected (built
against fixed node-components rev 752fa4a). This only fixes the on-disk
baked file used by GenesisSpec::Custom via ROLLUP_GENESIS_JSON_PATH.

Diff vs node-components/crates/genesis/src/gouda.genesis.json @ 752fa4a
is now empty.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
@rswanson
Copy link
Copy Markdown
Member Author

Fixed in 47f1ead. Diff vs node-components/crates/genesis/src/gouda.genesis.json @ 752fa4a is now empty (diff returns clean). Only host.startTimestamp differed: 17790515361765226348.

@rswanson rswanson marked this pull request as ready for review May 21, 2026 19:48
@rswanson rswanson requested a review from a team as a code owner May 21, 2026 19:48
Copy link
Copy Markdown
Contributor

@Fraser999 Fraser999 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Should we include a patch-level version bump on each of the three binaries? (there are no backwards-incompatible env var changes afaik).

Otherwise this LGTM.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants