Dashboard shows no live instances for multiple TiKV hosts

Translated
1

Note:
This topic has been translated from a Chinese forum by GPT and might contain errors.

Original topic: dashboard 多台tikv主机显示没有存活实例异常

| username: 像风一样的男子

【TiDB Environment】
Production
【TiDB Version】
4.0.9
【Encountered Issue】
Today, I suddenly found that the TiKV in the dashboard host list became unable to retrieve information, but querying all nodes on the tiup cluster display server showed normal status.

【Issue Phenomenon and Impact】

image
image1185×670 37.8 KB

image
image992×116 9.61 KB

image
image1170×670 33 KB

2
| username: xfworld | Original post link

Check the status of the PD leader to see if it is normal.
Look at the logs to see if there are any errors.

3
| username: 像风一样的男子 | Original post link

pd is normal, and there are no errors in the logs.

4
| username: xfworld | Original post link

How about switching the service instance of the dashboard?
But some historical data will be lost.

image
image815×573 29.5 KB

5
| username: tidb狂热爱好者 | Original post link

I feel that his machine’s memory is too small, only 8GB. The SQL on the dashboard consumes a lot of memory, causing the instance to display abnormally.