magicblock-labs / magicblock-labs/redsuite
Validate crash recovery through a durable checkpoint
@bzawisto is already working on this.
Since Sep 11, 2026.
- Dominant language
- Rust
- Stars
- 1
- Forks
- 0
- Avg merge
- 3h 58m
- Merged PRs (30d)
- 48
Description
## Outcome
Every transaction inside a known durable checkpoint must survive a crash.
The current restart workload permits missing confirmed tail entries. This scenario must establish its protected transaction set before the crash rather than remove missing entries from expectations afterward.
## Scope
Submit a known sequence containing successful mutations and executed failures. Wait until an archived checkpoint containing that sequence is complete, and record the checkpoint boundary and expected state.
Continue submitting distinguishable tail transactions, hard-kill MBV, and restart on the same storage. Repeat across several checkpoints.
Inspect every transaction in the protected set, not a sample. Classify later-tail outcomes separately and do not modify or restore database files manually.
## Acceptance criteria
- [ ] Every checkpointed transaction remains queryable with its expected success/failure outcome.
- [ ] Recovered state and transaction evidence agree, including rollback of failed program changes and permitted fee effects.
- [ ] Tail loss cannot shrink the previously established durable baseline.
- [ ] MBV accepts and executes fresh transactions after recovery.
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Assessment
This issue has not been assessed yet.