alombarte
commented
6 months ago Hi @tsemczyszyn, Thanks for testing the feature the same day we release it! I love it!
It is clear that adding the new module is causing the crash you are having, and we need to figure out what is causing this behavior. Jaeger was one of the components we used during testing with no problems, and I am interested in knowing this particular scenario.
Aside from Lua, are there any other custom plugins in your system? Can you report a minimal configuration example that leads to this crash without revealing anything sensitive? You can for instance use the echo or the debug endpoints to not reveal your backends. Something minimal that we can use to have a look, because otherwise this might be pretty hard to find.
Here's an example of a (working) minimal configuration:
{
"version": 3,
"$schema": "https://www.krakend.io/schema/krakend.json",
"host": [
"http://localhost:8080"
],
"debug_endpoint": true,
"echo_endpoint": true,
"endpoints": [
{
"endpoint": "/test",
"backend": [
{
"url_pattern": "/__debug/test"
}
]
}
],
"extra_config": {
"telemetry/opentelemetry": {
"service_name": "krakend",
"trace_sample_rate": 1,
"metric_reporting_period": 1,
"exporters": {
"otlp": [
{
"name": "jaeger",
"host": "jaeger",
"port": 4317,
"use_http": false,
"disable_metrics": true
}
]
}
}
}
}Testable with a docker compose:
version: "3"
services:
jaeger:
image: jaegertracing/all-in-one:1.54
environment:
COLLECTOR_ZIPKIN_HOST_PORT: ":9411"
ports:
- "55778:5778" # serve configs
- "56686:16686" # serve frontend UI
- "55317:4317" # otlp grpc: we remap this to be able to run other envs
- "55318:4318" # otlp http: we reamp this to be able to run other envs
krakend:
image: devopsfaith/krakend:watch
volumes:
- "./:/etc/krakend"
ports:
- "8080:8080"
command: ["run", "-c", "krakend.json"]Do you think that you can provide something similar that reproduces the crash?
Thank you very much for the report
tsemczyszyn
kpacha
Environment info:
Describe the bug I have a working configuration of krakend running in a testing environment I'm experimenting with. I wanted to try out the opentelemetry stuff added in 2.6 but when I add in the opentelemetry configuration the gateway crashes when I hit a specific endpoint. If I remove the opentelemetry config and repeat the same request everything works.
Your configuration file:
Here's the opentelemetry section I added:
Commands used How did you start the software?
Expected behavior The request should be routed normally without the gateway crashing
Logs
Additional context The endpoint in question uses modifier/lua-backend to call some scripts, but so do many other endpoints I'm using and those don't crash. I don't want to reveal my complete configuration out of concern there may be sensitive information. Please let me know what other information I could supply to help.