What
Re-run the full transaction smoke suite against the dev environment after it was rebooted from a clean build, and confirm that the previously-working transaction types continue to work plus that the bugs the recent merges targeted are actually fixed.
Status
In progress. Partial pass against the rebooted dev nodes is encouraging on the bug-fix side, but blocked on a chain-level stall that has not been touched for ~14 hours.
What is confirmed working
- Native pay on
dev.node2.demos.sh:53650: broadcast succeeds, transaction is included, balance and nonce advance. No Transaction hash mismatch — the bug that motivated the diagnostic PR earlier is gone on the post-merge build.
- Validator stake landed at block 174; the funded address is now an active validator (status 2, valid from block 177).
- SDK v4.0.6 from npm talks to the rebooted node end-to-end through
Demos.connect / confirm / broadcast and the getTransactionStatus polling method.
What is blocked by a deployment-level stall
After the first successful stake the chain stalled at block 182 on both dev.node2 and dev.node3. Both nodes still report:
- Last block 182, not advancing across ~14 hours of polling
- Mempool size 0 on both
getValidators returns 1 on node2 (my staked entry) but 0 on node3 — they disagree on the validator set
The disagreement on the validator set across the two-node deployment looks like the proximate cause: with only one node seeing a validator and the other seeing none, no quorum is achievable and block production halts. Subsequent broadcasts (second-run native pay and stake) returned valid: true but never appeared in either mempool.
This is a deployment / consensus-side artefact, not a code regression in the merged PRs — native pay and stake both validated cleanly while the chain was producing blocks, and the L2PS native canonicalisation fix from PR #926 cannot be exercised end-to-end while the chain is halted.
Next step
Flag the dev environment stall to the deploy owner so consensus can be unstuck — likely the genesis-baked validator on node3 needs to be registered in the validator set, or replication of the validator-set table between node2 and node3 needs to converge. Once chain advances again, re-run battery to exercise governance propose/vote and L2PS native transfer to verify PR #926 end-to-end.
What
Re-run the full transaction smoke suite against the dev environment after it was rebooted from a clean build, and confirm that the previously-working transaction types continue to work plus that the bugs the recent merges targeted are actually fixed.
Status
In progress. Partial pass against the rebooted dev nodes is encouraging on the bug-fix side, but blocked on a chain-level stall that has not been touched for ~14 hours.
What is confirmed working
dev.node2.demos.sh:53650: broadcast succeeds, transaction is included, balance and nonce advance. NoTransaction hash mismatch— the bug that motivated the diagnostic PR earlier is gone on the post-merge build.Demos.connect/confirm/broadcastand thegetTransactionStatuspolling method.What is blocked by a deployment-level stall
After the first successful stake the chain stalled at block 182 on both
dev.node2anddev.node3. Both nodes still report:getValidatorsreturns 1 on node2 (my staked entry) but 0 on node3 — they disagree on the validator setThe disagreement on the validator set across the two-node deployment looks like the proximate cause: with only one node seeing a validator and the other seeing none, no quorum is achievable and block production halts. Subsequent broadcasts (second-run native pay and stake) returned
valid: truebut never appeared in either mempool.This is a deployment / consensus-side artefact, not a code regression in the merged PRs — native pay and stake both validated cleanly while the chain was producing blocks, and the L2PS native canonicalisation fix from PR #926 cannot be exercised end-to-end while the chain is halted.
Next step
Flag the dev environment stall to the deploy owner so consensus can be unstuck — likely the genesis-baked validator on node3 needs to be registered in the validator set, or replication of the validator-set table between node2 and node3 needs to converge. Once chain advances again, re-run battery to exercise governance propose/vote and L2PS native transfer to verify PR #926 end-to-end.