SST Missing Investigation

关于

This skill automates the complex process of debugging missing SST files within TiDB environments. It systematically correlates TiFlash FATAL logs with TiKV compaction and read records, maps SST IDs to specific shards and regions, and integrates s3clean removal timestamps. By constructing a pod-scoped timeline and gathering evidence with exact file-and-line references, it enables engineers to quickly determine if an incident was triggered by HA chaos, improper compaction, or cleanup overlaps, significantly reducing the mean time to resolution for storage layer errors.

主要功能

• Cross-pod log correlation between TiKV, TiFlash, and s3clean
• HA chaos window and TiKV restart gap analysis
• 0 GitHub stars
• Automated timeline generation for compaction and deletion events
• SST ID to shard/region mapping and metadata extraction
• Structured evidence collection with exact file and line references

使用场景

• Debugging NoSuchKey FATAL errors in TiFlash proxy logs
• Investigating storage gaps after TiKV node restarts or HA events
• Tracing the lifecycle of an SST file from creation to S3 deletion