Divergence between CCF testing and IRL testing
Nobody has claimed this yet.
- Dominant language
- C++
- Stars
- 876
- Forks
- 260
- Avg merge
- 1d 11h
- Merged PRs (30d)
- 157
Description
Everywhere we have a node join a network, we currently start it with either a copy of the primary's snapshot directory (or after #7832 an empty directory) by default.
Both of these options are not quite right, although #7832 moves us closer.
What we observe is that teams running CCF generally have a snapshots folder from previous iterations of that node.
Hence we should make the following changes:
-
join_nodeshould replicate the snapshot directory of a random node -
start_in_recovery's_add_nodewhen passed an existing network should infer which node to start from, and reuse the ledger directories of the previous network.
Contributor guide
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.
Research direction
Start with the join_node path and start_in_recovery's _add_node handling for an existing network; review #7832 for the empty-directory behavior. Trace how snapshot and ledger directories are selected, then verify that joins replicate a random node's snapshot and recovery reuses the previous network's ledger directories.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp
- Domain
- distributed-systems, testing-qa
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 48/100