Files
explorer-monorepo/docs/specs/ccip/ccip-observability.md

111 lines
2.3 KiB
Markdown

# CCIP Observability Dashboard Specification
## Overview
This document specifies the observability dashboards and analytics for CCIP message lifecycle visualization and monitoring.
## Dashboard Components
### Message Lifecycle Visualization
**Components**:
- Timeline view showing source → CCIP → destination flow
- Status indicators at each stage
- Time metrics (send time, delivery time, execution time)
- Transaction links (click to view source/destination transactions)
**Visualization**:
```
[Source Chain] → [CCIP DON] → [Destination Chain]
✓ ✓ ✓
Sent Delivered Executed
00:00 00:02 00:05
```
### Status Aggregation
**Metrics**:
- Total messages by status
- Success rate (% executed successfully)
- Average execution time
- Failed message count
- Expired message count
**Time Series**:
- Messages per hour/day
- Success rate over time
- Average execution time over time
### Failure Analysis
**Failure Categories**:
- Execution failures (revert on destination)
- Timeouts (expired messages)
- Delivery failures (CCIP DON issues)
- Invalid messages
**Analysis**:
- Failure rate by category
- Common failure reasons
- Failure patterns (time, chain pairs, etc.)
### Performance Metrics
**Metrics**:
- Average delivery time (source → CCIP)
- Average execution time (CCIP → destination)
- Total end-to-end time
- P50, P95, P99 latencies
**Charts**:
- Latency distribution
- Latency trends over time
- Latency by chain pair
## Cross-Chain Analytics
### Chain Pair Analysis
**Metrics per Chain Pair**:
- Message volume
- Success rate
- Average latency
- Popular routes
**Visualization**:
- Network graph showing chain connections
- Edge weights showing message volume
- Color coding for success rates
### Token Flow Analysis
**Tracking**:
- Token transfers via CCIP
- Volume by token
- Volume by chain pair
- Cumulative volume over time
## Real-Time Monitoring
### Live Message Stream
**Features**:
- Real-time updates of new messages
- Status change notifications
- Alert on failures
- Filter by chain, status, etc.
### Alerts
**Alert Conditions**:
- High failure rate (> 5%)
- Message timeout (> 30 minutes)
- Unusual message volume spike
- Chain connectivity issues
## References
- CCIP Tracking: See `ccip-tracking.md`
- Observability: See `../observability/`