Note:
This topic has been translated from a Chinese forum by GPT and might contain errors.
Original topic: TiKV_approximate_region_size 告警但是没有具体的tikv实例说明
When the TiKV_approximate_region_size alarm is triggered, but the specific alarm information does not show the specific TiKV instance, which is not conducive to analysis, does anyone have any methods to display the specific TiKV instance information in this alarm strategy?
It is estimated that it is not possible to directly obtain information about the TiKV instance. If the splitting speed is slower than the writing speed, it should be easy to see through the hotspot TOP on the dashboard.
Transform the expression to “by (le, instance)” and take a look first.
Okay, I’ll give it a try.
Sometimes, it may also be due to region merge causing the region to become too large, leading to a split.
This situation is very rare, usually, empty regions are merged to reduce resource consumption.
Just pay attention to the hotspots.