Third regression fix for the plasma-consensus 0.15.0→1.1.0 bump (issue #5309). Prior fixes: --nat flag (PR #5316, merged) and restoring trusted_peers (PR #90, merged) — but #90's restored [[network.trusted_peers]] array-of-tables (with trusted_only left true) still fails 1.1.0's schema validation:
plasma-mainnet-node-1 | Error: Invalid config schema: Schema compliance: 'Invalid Toml config file. /tmp/non-validator.toml [{"address":"...","peer_id":"..."},...] is not of type "object"'.
Root cause (verified against upstream, not guessed): fetched the official plasma-consensus 1.1.0 template — https://raw.githubusercontent.com/PlasmaLaboratories/node-templates/main/config/mainnet/non-validator.toml. It has trusted_only = false and no trusted_peers field at all for an observer node; bootstrap_nodes + discovery is the whole story. Every other section of our plasma/mainnet/non-validator.toml (bootstrap_nodes.0/1/2, bls_peer_ids, api, chain.aquila, chain.static_committee.*) already matches this upstream template byte-for-byte — only trusted_only and the leftover trusted_peers block (both carried over from the pre-1.1.0 0.15.0 config) were wrong.
Fix: set trusted_only = false; delete the three [[network.trusted_peers]] blocks entirely. Nothing else changed — diff verified byte-for-byte against the upstream template after the change.
Impact of not merging:plasma-mainnet-node-1 (observer) on rpc-de-27 stays isolated/crash-looping; the paired reth has been stalled at latest_block=33349025 since 2026-09-28T00:48Z (~13.5h now).
Rollout plan: after merge, redeploy rpc-de-27 only (single host), then verify: plasma-mainnet-node-1 container Up with no crashloop, logs show it connecting via discovery/bootstrap, reth resumes advancing past latest_block=33349025. Fleet propagation follows as a separate, non-gating step once verified.
Commit: 3a4bbcb166597cae8c5151f73ddba7faf48cb92d on branch plasma-consensus-trusted-only-fix.
Third regression fix for the plasma-consensus 0.15.0→1.1.0 bump (issue #5309). Prior fixes: `--nat` flag (PR #5316, merged) and restoring `trusted_peers` (PR #90, merged) — but #90's restored `[[network.trusted_peers]]` array-of-tables (with `trusted_only` left `true`) still fails 1.1.0's schema validation:
```
plasma-mainnet-node-1 | Error: Invalid config schema: Schema compliance: 'Invalid Toml config file. /tmp/non-validator.toml [{"address":"...","peer_id":"..."},...] is not of type "object"'.
```
**Root cause (verified against upstream, not guessed):** fetched the official plasma-consensus 1.1.0 template — `https://raw.githubusercontent.com/PlasmaLaboratories/node-templates/main/config/mainnet/non-validator.toml`. It has `trusted_only = false` and **no `trusted_peers` field at all** for an observer node; bootstrap_nodes + discovery is the whole story. Every other section of our `plasma/mainnet/non-validator.toml` (`bootstrap_nodes.0/1/2`, `bls_peer_ids`, `api`, `chain.aquila`, `chain.static_committee.*`) already matches this upstream template byte-for-byte — only `trusted_only` and the leftover `trusted_peers` block (both carried over from the pre-1.1.0 0.15.0 config) were wrong.
**Fix:** set `trusted_only = false`; delete the three `[[network.trusted_peers]]` blocks entirely. Nothing else changed — diff verified byte-for-byte against the upstream template after the change.
**Impact of not merging:** `plasma-mainnet-node-1` (observer) on `rpc-de-27` stays isolated/crash-looping; the paired reth has been stalled at `latest_block=33349025` since 2026-09-28T00:48Z (~13.5h now).
**Rollout plan:** after merge, redeploy `rpc-de-27` only (single host), then verify: `plasma-mainnet-node-1` container Up with no crashloop, logs show it connecting via discovery/bootstrap, reth resumes advancing past `latest_block=33349025`. Fleet propagation follows as a separate, non-gating step once verified.
Commit: `3a4bbcb166597cae8c5151f73ddba7faf48cb92d` on branch `plasma-consensus-trusted-only-fix`.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Third regression fix for the plasma-consensus 0.15.0→1.1.0 bump (issue #5309). Prior fixes:
--natflag (PR #5316, merged) and restoringtrusted_peers(PR #90, merged) — but #90's restored[[network.trusted_peers]]array-of-tables (withtrusted_onlylefttrue) still fails 1.1.0's schema validation:Root cause (verified against upstream, not guessed): fetched the official plasma-consensus 1.1.0 template —
https://raw.githubusercontent.com/PlasmaLaboratories/node-templates/main/config/mainnet/non-validator.toml. It hastrusted_only = falseand notrusted_peersfield at all for an observer node; bootstrap_nodes + discovery is the whole story. Every other section of ourplasma/mainnet/non-validator.toml(bootstrap_nodes.0/1/2,bls_peer_ids,api,chain.aquila,chain.static_committee.*) already matches this upstream template byte-for-byte — onlytrusted_onlyand the leftovertrusted_peersblock (both carried over from the pre-1.1.0 0.15.0 config) were wrong.Fix: set
trusted_only = false; delete the three[[network.trusted_peers]]blocks entirely. Nothing else changed — diff verified byte-for-byte against the upstream template after the change.Impact of not merging:
plasma-mainnet-node-1(observer) onrpc-de-27stays isolated/crash-looping; the paired reth has been stalled atlatest_block=33349025since 2026-09-28T00:48Z (~13.5h now).Rollout plan: after merge, redeploy
rpc-de-27only (single host), then verify:plasma-mainnet-node-1container Up with no crashloop, logs show it connecting via discovery/bootstrap, reth resumes advancing pastlatest_block=33349025. Fleet propagation follows as a separate, non-gating step once verified.Commit:
3a4bbcb166597cae8c5151f73ddba7faf48cb92don branchplasma-consensus-trusted-only-fix.🤖 Generated with Claude Code
merge-bot (del-018): classified B (hand-maintained rpc file(s), single scope). Auto-merge after a 72h objection window - comment 'hold' to block.