Grafana memory leak?

[atb2215]

Running several dashboards with IotDB version 1.1.1-SNAPSHOT (Build: c9b4dc2-dev) and Grafana v9.5.2.
There seems to be a memory leak. Running "show queries" gives hundreds of apparently unresolved queries
+------------------------+-----------------------+----------+-----------+
| Time| QueryId|DataNodeId|ElapsedTime|
|2023-06-12T01:48:39.177Z|20230612_014839_00000_1| 1| 12055.599|
|2023-06-12T01:48:39.178Z|20230612_014839_00001_1| 1| 12055.598|
|2023-06-12T01:48:39.186Z|20230612_014839_00002_1| 1| 12055.59 |
....
Running 'KILL ALL QUERIES' and 'CLEAR CACHE' seems to have no impact and query_timeout_threshold is set to 10000ms.

Is this a bug or am I misunderstanding how "KILL QUERY" is intended to work?

[JackieTien97]

Can you give the full result set of show queries, what's exactly the running queries?
KILL ALL QUERIES will kill the running queries, but you should know that

1. It's an async opreation and you can take it as a kill signal which means that the actual killing operation will run in the background.
2. V1.1.2 IoTDB has fixed some memory dead lock bug, you can upgrade to it.

[atb2215]

Thanks @JackieTien97 , upgrading to 1.1.2 seems to have fixed it!