Issue with the paths for monitoring using the _internal database

Hi everyone,

I have got an influxDB version 1.8 running on AWS EC2, and I would like to monitor it using the metrics from the internal database (that I query using Grafana).

For example I want to monitor the cache memory so the field memBytes of the measurement tsm1_cache. When I query it I get multiple values for the same time that correspond to several paths I believe.
The query: SELECT memBytes, path FROM _internal.monitor.tsm1_cache WHERE $timeFilter LIMIT 20

The result:

2022-03-20 16:05:10 0 /influxdb/data/smartservices/autogen/1900
2022-03-20 16:05:10 344.00 /influxdb/data/smartservices/autogen/1898
2022-03-20 16:05:10 0 /influxdb/data/smartservices/one_year/1528
2022-03-20 16:05:10 0 /influxdb/data/smartservices/one_year/1273
2022-03-20 16:05:10 0 /influxdb/data/smartservices/autogen/1896
2022-03-20 16:05:10 0 /influxdb/data/smartservices/autogen/1894
2022-03-20 16:05:10 0 /influxdb/data/smartservices/one_year/1513
2022-03-20 16:05:10 0 /influxdb/data/smartservices/autogen/1892
2022-03-20 16:05:10 0 /influxdb/data/smartservices/autogen/1890
2022-03-20 16:05:10 0 /influxdb/data/smartservices/one_year/1498
2022-03-20 16:05:10 0 /influxdb/data/smartservices/one_year/1258
2022-03-20 16:05:10 21.73 Mil /influxdb/data/_internal/monitor/1944
2022-03-20 16:05:10 0 /influxdb/data/_internal/monitor/1936
2022-03-20 16:05:10 0 /influxdb/data/_internal/monitor/1931
2022-03-20 16:05:10 57.01 Mil /influxdb/data/smartservices/autogen/1945
2022-03-20 16:05:10 0 /influxdb/data/_internal/monitor/1929

What am I supposed to do with that? I just want a metric that gives the total memory used by the database cache.

Thanks in advance,

Hello @Quentin_Na,
I think you could do:
sum(memBytes)

Thanks for the answer !
So it means that those metrics for a same timestamp are not redundant. Cache memory is not counted twice. Can you confirm that ?