tikv / tikv/pd

For some Gauge metrics items, `max` should be used

Open
#7,506 0 comments 0 reactions 0 assignees View on GitHub
type/enhancement
Dominant language
Go
Stars
1.2k
Forks
783
Avg merge
5d 21h
Merged PRs (30d)
36

Description

## Enhancement Task
There are one tikv. I killed PD and restarted PD, and put a store.
I see

![image](https://github.com/tikv/pd/assets/23399268/01fd7655-8c75-4b32-910a-a935e0fa4c4e)
![image](https://github.com/tikv/pd/assets/23399268/52c9f9e9-1f06-4657-8142-618eedbec10c)

Now we don't choose a PD instance in grafana.
When PD killed or panic, the prometheus metrics will not be reset, so `sum` maybe wrong for `Gauge`

Contributor guide

Open the contributing guide

Research direction

Start by identifying the Grafana panels and Prometheus queries for the affected Gauge metrics, then reproduce the PD restart scenario described in the issue. Determine which Gauge items should use max rather than sum, and verify that the dashboards show correct values after a PD instance is killed and restarted.

Written by the indexing model from the issue text.

Assessment

Tech stack
go, grafana, prometheus
Domain
observability
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.