[Bug]: HotRod run failed if using the old env variable

[asatt]

What happened?

The HotRod can fail if we try to run it in Kubernetes with Jaeger Agent (with a Service name jaeger-agent) or with one of the old parameters like JAEGER_AGENT_PORT (that have OTEL_EXPORTER_* alternatives).

During a short investigation, I found the issue added after PR #5437.
After removing expvar the method

jaegerclientenv2otel.MapJaegerToOtelEnvVars(logger)

here https://github.com/jaegertracing/jaeger/blob/main/examples/hotrod/cmd/root.go#L59
now calling with a non-initialized logger variable.

As a result, the logger in jaegerclientenv2otel.MapJaegerToOtelEnvVars will fail
https://github.com/jaegertracing/jaeger/blob/main/internal/jaegerclientenv2otel/envvars.go#L57
with nil pointer.

Steps to reproduce

1. Add in ENV the variable like JAEGER_AGENT_HOST=<value> or JAEGER_AGENT_PORT=<value> (in Kubernetes deployment or for local run)
2. Run HotRod

Expected behavior

It should run as usual, without exceptions, and print in logs

Replacing deprecated Jaeger SDK env var %s with OpenTelemetry env var %s

Relevant log output

panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x1 addr=0x0 pc=0xa6d5c5]
goroutine 1 [running]:
go.uber.org/zap.(*Logger).clone(...)
 go.uber.org/[email protected]/logger.go:318
go.uber.org/zap.(*Logger).Sugar(...)
 go.uber.org/[email protected]/logger.go:147
github.com/jaegertracing/jaeger/internal/jaegerclientenv2otel.MapJaegerToOtelEnvVars(0x0)
 github.com/jaegertracing/jaeger/internal/jaegerclientenv2otel/envvars.go:57 +0x1a5
github.com/jaegertracing/jaeger/examples/hotrod/cmd.onInitialize()
 github.com/jaegertracing/jaeger/examples/hotrod/cmd/root.go:59 +0x29
github.com/spf13/cobra.(*Command).preRun(...)
 github.com/spf13/[email protected]/command.go:1019
github.com/spf13/cobra.(*Command).execute(0x1221740, {0x128eb40, 0x0, 0x0})
 github.com/spf13/[email protected]/command.go:929 +0x5a2
github.com/spf13/cobra.(*Command).ExecuteC(0x1221180)
 github.com/spf13/[email protected]/command.go:1117 +0x3ff
github.com/spf13/cobra.(*Command).Execute(...)
 github.com/spf13/[email protected]/command.go:1041
github.com/jaegertracing/jaeger/examples/hotrod/cmd.Execute()
 github.com/jaegertracing/jaeger/examples/hotrod/cmd/root.go:46 +0x26
main.main()
 ./main.go:11 +0xf

Screenshot

No response

Additional context

I'm using the HotRod with the ENV OTEL_EXPORTER_OTLP_ENDPOINT. By mistake, I deployed the deprecated Jaeger Agent (as a DaemonSet) with a Kubernetes Service jaeger-agent name. So Kubernetes injected the ENV JAEGER_AGENT_PORT inside the HotRod pod and the HotRod started crashing with the error.

Jaeger backend version

v1.58.x

SDK

No response

Pipeline

No response

Stogage backend

Cassandra

Operating system

Linux

Deployment model

Kubernetes, binaries

Deployment configs

No response

[yurishkuro]

Thanks for reporting. Fix #5760

[Manoramsharma]

I was just about to make those changes 😄