SQLServer plugin shows wrong latency values about telegraf HOT 4 CLOSED

Hello! I recommend posting this question in our Community Slack or Community Forums, we have a lot of talented community members there who could help answer your question more quickly. You can also learn more about Telegraf by enrolling at InfluxDB University for free!

Heads up, this issue will be automatically closed after 7 days of inactivity. Thank you!

from telegraf.

@Trovalo any idea what is going on here? Is there a chance that those are micro- or nanoseconds?

from telegraf.

The number is fine as it is, but it's not an average... it's the total "latency" since the beginning of time (server start).
Here are the MsDocs, reference to io_stall_read_ms.

Going back to Telegraf, if you want the average latency, divide read_latency_ms by reads, and you will have the average latency per read (since the beginning of time... not so useful), do the same with the difference between samples (as those are incremental counters) and you will have the actual latency per time interval

something like this (on InfluxDB)

SELECT
	 non_negative_difference(last("read_bytes")) AS "read_bytes"
	,non_negative_difference(last("read_latency_ms")) AS "read_latency_ms"
	,non_negative_difference(last("reads")) AS "reads"
	,.....
FROM ___."sqlserver_database_io"
WHERE time >= now()-5m
GROUP BY
	time(1m)
	,*

from telegraf.

Thanks for the quick explanation Trovalo. I'm closing this as it isn't an issue.

from telegraf.