DanConwayDev
d3d905010c
The previous design held the per-(submitter, identifier) mutex for the
entire receive-pack pipeline — `git init --bare`, the pack upload from
the client, `git-receive-pack`, per-ref validation, and the zero-ref
cleanup. Two concurrent pushes to the same `/prs/<submitter>/<id>.git`
path were therefore fully serialised end-to-end, with the second push's
HTTP connection blocked on the mutex for the entire duration of the
first push (including its pack upload over the wire). With multiple
agents or CI jobs sharing a contributor identity this could produce
HTTP/proxy timeouts and apparently-stuck pushes for no good reason —
git's own ref locking would normally handle intra-push concurrency on
the same path just fine.
Switch to a `PrsPathState { mu: Mutex<()>, in_flight: AtomicUsize }`
per path. The receive handler now takes the mutex only briefly:
* at the start of the request to run `git init --bare` and bump
`in_flight` from 0 to 1, then release the mutex,
* at the end of the request to drop `in_flight` and, if it lands on
zero with the repo at zero refs, `rm -rf` the bare directory.
`git-receive-pack` and per-ref validation run with no per-path lock
held, so concurrent pushes by different agents to the same path proceed
in parallel.
Off-push cleanup paths (the PR-event policy when it discards a scoped
placeholder whose incoming event mismatches, and the purgatory expiry
sweep when a scoped placeholder times out) take the same mutex briefly,
delete the dangling ref, then remove the bare directory only when
`in_flight.load() == 0` and `list_refs` is empty. With both reads
performed under the mutex that gates `in_flight` mutations, a repo
deletion can never race a push that is mid-receive: either the push is
still in init/register and we wait on the mutex, or the push has
already incremented `in_flight` (so we read non-zero and skip), or the
push has finished and decremented (so the directory is genuinely idle).
The end-of-push cleanup now also runs when receive-pack returns a
protocol-error response (200 with ERR pkt-line), which the previous
code's early-return skipped — a probe push that failed git-level
validation after `ensure_repo_initialised` had created the directory
used to leak an empty `.git` dir, which now gets removed in the same
critical section.
A small `path_state(&locks, path)` helper centralises the
get-or-insert-Arc dance the three sites used to duplicate.
Docs updated:
* docs/explanation/grasp-06-contributor-pr-submission.md — replace
the "lock held for the entire push pipeline" paragraph with the
new mutex + `in_flight` discipline; update the three-sites
enumeration to describe the `in_flight == 0` guard.
* docs/explanation/architecture.md — same shape, one-paragraph.
* docs/how-to/enable-grasp-06.md — call out that the mutex is held
only briefly so concurrent pushes don't serialise.
Explanation
Understanding-oriented documentation - Concepts, design decisions, and the "why" behind ngit-grasp.
What Is Explanation?
Explanation documentation helps you understand concepts and design decisions, providing context and discussing alternatives.
Characteristics:
- ✅ Understanding-oriented (clarify concepts)
- ✅ Theoretical (ideas and design)
- ✅ Discuss alternatives
- ✅ Provide context and background
- ✅ Answer "why" questions
Not explanation:
- ❌ Step-by-step lessons (those are Tutorials)
- ❌ Problem-solving recipes (those are How-To)
- ❌ Technical specifications (those are Reference)
Available Explanation Documentation
Architecture Overview
Understand the system design and component interaction
Topics:
- Overall architecture
- Component responsibilities
- Data flows
- Technology choices
- Design patterns
Read when: You want to understand how ngit-grasp works as a system
Inline Authorization
Why we validate pushes inline instead of using Git hooks
Topics:
- The authorization problem
- Git hooks approach
- Inline approach
- Comparison and trade-offs
- Implementation details
Read when: You want to understand the core architectural decision
Design Decisions
Key architectural choices and their rationale
Topics:
- Inline authorization vs hooks
- Technology stack choices
- Storage design
- API design
- Performance considerations
Read when: You want to know why things are the way they are
Comparison with ngit-relay
How ngit-grasp differs from the reference implementation
Topics:
- Architecture comparison
- Component differences
- Trade-offs
- Migration path
- Compatibility
Read when: You're familiar with ngit-relay and want to understand differences
Purgatory Design
In-memory holding area for events awaiting git data
Topics:
- The "which arrives first?" problem
- Separate storage for state vs PR events
- Late binding for state events
- Bidirectional waiting for PR events
- Authorization during push
Read when: You want to understand how ngit-grasp handles out-of-order event/git data arrival
GRASP-02 Proactive Sync
Relay-to-relay synchronization for repository discovery
Topics:
- Negentropy-based event sync
- Repository announcement discovery
- Relay management and reconnection
- Layer 2 filtering
- Bootstrap and dynamic relay discovery
Read when: You want to understand how ngit-grasp discovers and syncs repositories across relays
GRASP-02 Purgatory Git Data Fetching
Proactive git data fetching from remote servers
Topics:
- Identifier-based batching
- Exponential backoff with fresh start
- Domain throttling (5 concurrent, 30/min)
- Debounced delays (3min user, 500ms sync)
- 30-minute expiry
- Mock-based testability
Read when: You want to understand how purgatory automatically fetches missing git data
Unified Git Data Sync
Shared processing for git push and purgatory sync paths
Topics:
- Why unify push and sync processing
- OID syncing to owner repos
- Ref alignment logic
- Event release from purgatory
- WebSocket notification
Read when: You want to understand how git data is processed consistently regardless of arrival method
Monitoring Overview
Prometheus metrics and observability
Topics:
- Metrics philosophy
- Connection tracking
- Git operation metrics
- Nostr event metrics
- Privacy considerations
Read when: You want to understand how to monitor ngit-grasp in production
Defensive Measures & Rate Limiting
Protection against abuse, spam, and denial-of-service attacks
Topics:
- Connection and subscription management
- Event publishing rate limits
- Content filtering (blacklists/whitelists)
- Event validation plugin system (WritePolicy/QueryPolicy)
- Relay health management (naughty list, exponential backoff)
- Privacy-preserving IP tracking
- Future enhancements (per-IP rate limiting)
Read when: You want to understand how ngit-grasp protects against abuse and what defensive features are available
GRASP-05 Archive Mode
Read-only mirroring of repositories
Topics:
- Archive whitelist configuration
- Archive-all mode
- Read-only mode defaults
- Use cases for backup/mirror relays
Read when: You want to understand how to run an archive/backup relay
Deletion Requests
Handling repository and event deletion
Topics:
- Deletion request architecture
- Delete disrespector concept
- Preventing left-pad scenarios
- Archival policies
Read when: You want to understand how ngit-grasp handles deletion events (planned feature)
Planned Explanation Documentation
GRASP Protocol Design
Status: 🔜 Planned
Topics:
- Why Nostr for Git?
- Authorization model
- Trust and verification
- Decentralization benefits
Storage Architecture
Status: 🔜 Planned
Topics:
- Why separate Git and Nostr storage?
- Indexing strategy
- Performance considerations
- Scaling approach
Testing Philosophy
Status: 🔜 Planned
Topics:
- Why test isolation?
- Integration vs unit tests
- Compliance testing approach
- Test-driven development
Performance Considerations
Status: 🔜 Planned
Topics:
- Async architecture
- Caching strategy
- Database choices
- Bottlenecks and solutions
How to Use Explanation Documentation
- Read to understand - Not to accomplish a task
- Follow your curiosity - Read what interests you
- Connect concepts - Link ideas together
- Question and explore - Think critically
Not sure if this is what you need?
- Want to learn by doing? → Tutorials
- Need to solve a problem? → How-To Guides
- Looking for technical details? → Reference
Contributing Explanation Documentation
When writing explanation:
DO:
- ✅ Discuss concepts and ideas
- ✅ Provide context and background
- ✅ Explain alternatives
- ✅ Use analogies and examples
- ✅ Connect to broader context
- ✅ Answer "why" questions
DON'T:
- ❌ Provide step-by-step instructions (link to Tutorials/How-To)
- ❌ List technical details (link to Reference)
- ❌ Assume you must be comprehensive
- ❌ Avoid opinions (explanation can be opinionated)
Template:
# Explanation: [Topic]
**Purpose:** [What concept/decision this explains]
**Audience:** [Who wants to understand this]
---
## The Problem/Question
[What are we trying to understand?]
---
## Background
[Context and history]
---
## Our Approach
[How we address it]
### Why This Works
[Explanation of benefits]
### Trade-offs
[What we gain and lose]
---
## Alternatives Considered
### [Alternative 1]
**Pros:**
- [Benefits]
**Cons:**
- [Drawbacks]
**Why we didn't choose it:**
[Reasoning]
---
## Conclusion
[Summary of understanding]
---
## Related Documentation
- [Links to relevant docs]
See Diátaxis: Explanation for detailed guidance.
Part of the ngit-grasp documentation using the Diátaxis framework.