cockroachdb / cockroachdb/cockroach
Provide better observability into protected timestamps (PTS) impacting workload performance
Open
A-observability-inf
C-enhancement
C-escalation-improvement
O-support
P-3
T-observability
- Dominant language
- Go
- Stars
- 32.5k
- Forks
- 4.1k
- PR merge metrics
- PR metrics pending
Description
This issue tracks providing better observability into protected timestamps that may be impacting workload performance and costs.
- How much Non-live data is due to PTS for a particular table/index?
- What is causing PTS to be accumulated? For instance, which job(s) or schedule(s)?
- How much PTS is accumulated due to a particular job(s) or schedule(s)? At what time?
An example user flow:
1. User deletes from, truncates, or drops a table
2. Data is still not GC'd due to a 'paused' job on that table.
3. Example symptoms: Query latency increases (garbage data), Storage capacity alerts
Jira issue: CRDB-24183
Contributor guide
Assessment
This issue has not been assessed yet.