Most OpenTelemetry automatic instrumentation metrics don't work on native #48365
Description
Describe the bug
We use external libraries from OTel to create these metrics and it turns out most of them don't work with native mode.
This this will require a lot of work to fix and should be made in the scope of a new issue.
Most likely we will need to either contribute fixes to upstream or instrument ourselves.
This the current compatibility matrix:
| Metric Name | Description | Type | Available on JVM? | Available on Native? | MP 2.0? |
|---|---|---|---|---|---|
| http.server.request.duration | Duration of HTTP server requests | HISTOGRAM | Y | Y | Y |
| jvm.memory.committed | Measure of memory committed | LONG_SUM | Y | No data produced | Y |
| jvm.memory.used | Measure of memory used | LONG_SUM | Y | No data produced | Y |
| jvm.memory.limit | Measure of max obtainable memory | LONG_SUM | Y | Not present | Y |
| jvm.memory.used_after_last_gc | Measure of memory used, as measured after the most recent garbage collection event on this pool. | LONG_SUM | Y | No data produced | Y |
| jvm.gc.duration | Duration of JVM garbage collection actions | HISTOGRAM | Y | Not present | Y |
| jvm.class.count | Number of classes currently loaded. | LONG_SUM | Y | No data produced | Y |
| jvm.class.loaded | Number of classes loaded since JVM start. | LONG_SUM | Y | No data produced | Y |
| jvm.class.unloaded | Number of classes unloaded since JVM start. | LONG_SUM | Y | No data produced | Y |
| jvm.cpu.count | Number of processors available to the Java virtual machine. | LONG_SUM | Y | Y | N |
| jvm.cpu.limit | LONG_SUM | Y | No data produced | N | |
| jvm.cpu.time | CPU time used by the process as reported by the JVM. | DOUBLE_SUM | Y | Not present | N |
| jvm.system.cpu.utilization | CPU time used by the process as reported by the JVM. | DOUBLE_SUM | Not present | No data produced | N |
| jvm.cpu.recent_utilization | Recent CPU utilization for the process as reported by the JVM. | DOUBLE_GAUGE | Y | No data produced | N |
| jvm.cpu.longlock | Long lock times | HISTOGRAM | Y | Y | N |
| jvm.cpu.context_switch | DOUBLE_SUM | Y | No data produced | N | |
| jvm.network.io | Network read/write bytes. | HISTOGRAM | Y | Not present | N |
| jvm.network.time | Network read/write duration. | HISTOGRAM | Y | Not present | N |
| jvm.thread.count | Number of executing platform threads. | LONG_SUM | Y | Y | Y |
Expected behavior
Most metrics, especially the MP 2.0 related ones, should be available on native mode.
Actual behavior
See table in the description.
How to Reproduce?
No response
Output of uname -a or ver
No response
Output of java -version
No response
Quarkus version or git rev
No response
Build tool (ie. output of mvnw --version or gradlew --version)
No response
Additional information
No response