Use rr for randomized fault injection
Open
Nobody has claimed this yet.
- Dominant language
- C++
- Stars
- 10.7k
- Forks
- 662
- Avg merge
- 2d 3h
- Merged PRs (30d)
- 2
Description
Complementary to @rocallahan's gecko chaos mode. There are bazillion other tools that do this, but implementing this in rr is particularly compelling because
- no PRELOAD hacks required; rr 100% precisely sees all syscalls (through ptrace and syscallbuf)
- low overhead
- and most importantly of all, rr is recording execution the whole damn time! :) So if a failure pops up, you can immediately switch to debugging the trace.
This could be done with relatively little pain by adding a fault injection hook to rec_process_event.cc, and then recording pseudo-events when faults are injected so that they can be replayed.
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 by reading rec_process_event.cc and tracing how rr observes syscalls through ptrace and syscallbuf. Determine where a fault-injection hook could fit and how pseudo-events would be recorded for replay. Done means injected faults can be reproduced when debugging the resulting trace.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp, linux
- Domain
- devtools, operating-systems
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 25/100