]
Jeff Mesnil resolved WFLY-11394.
--------------------------------
Resolution: Won't Fix
I agree, the unit of the metrics is handled differently by the MP Metrics endpoints.
The JSON format returns the "raw" value and leaves to the client to find the
unit by looking at the metrics metadata with the OPTIONS HTTP method.
Prometheus format returns the "base" value (eg seconds for time units) and
appends the unit to the metric name.
Wrong units of jvm.uptime metric - milliseconds vs. seconds
-----------------------------------------------------------
Key: WFLY-11394
URL:
https://issues.jboss.org/browse/WFLY-11394
Project: WildFly
Issue Type: Bug
Components: MP Metrics
Reporter: Rostislav Svoboda
Assignee: Jeff Mesnil
Priority: Major
Wrong units of jvm.uptime metric - milliseconds vs. seconds
{code}
curl -H "Accept: application/json" -X OPTIONS
http://localhost:9990/metrics/base/jvm.uptime
{
"jvm.uptime": {
"unit": "milliseconds",
"type": "gauge",
"description": "Displays the start time of the Java virtual
machine in milliseconds. This attribute displays the approximate time when the Java
virtual machine started.",
"displayName": "JVM Uptime",
"tags": ""
}
{code}
This metric reports results in seconds, not in milliseconds as description says.
{code}
curl
http://localhost:9990/metrics/base/jvm.uptime
# HELP base:jvm_uptime_seconds Displays the start time of the Java virtual machine in
milliseconds. This attribute displays the approximate time when the Java virtual machine
started.
# TYPE base:jvm_uptime_seconds gauge
base:jvm_uptime_seconds 2011.875
{code}