feat: add Adaline Trace integration for OpenTelemetry #135
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Adds native support for exporting OpenTelemetry traces to Adaline Trace,
the hosted observability platform for LLM applications.
Changes:
- Add AdalineTraceExporter class (SpanExporter implementation)
- Add AdalineTracer wrapper class with helper methods
- Export classes from @adaline/gateway package
- Add comprehensive but concise documentation to README
Features:
- Scoped tracing (only Gateway operations, not entire app)
- Automatic span creation for all LLM calls
- Built-in trace() helper for grouping workflows
- Session IDs and tags for categorization
- Configurable service name, version, and debug mode
Usage:
```typescript
import { Gateway, AdalineTracer } from "@adaline/gateway";
const adalineTracer = new AdalineTracer({
apiKey: process.env.ADALINE_API_KEY,
projectId: process.env.ADALINE_PROJECT_ID,
});
const gateway = new Gateway({
telemetry: { tracer: adalineTracer.tracer }
});
```
All Gateway operations are automatically traced with full observability:
model names, cache hits, latency, time-to-first-token, and span hierarchies.
Contributor
Greptile OverviewGreptile SummaryAdds native OpenTelemetry integration for exporting traces to Adaline Trace, a hosted observability platform for LLM applications. Key Changes:
Implementation Quality:
Issue Found:
Confidence Score: 4/5
Important Files ChangedFile Analysis
Sequence DiagramsequenceDiagram
participant User
participant AdalineTracer
participant Gateway
participant NodeTracerProvider
participant BatchSpanProcessor
participant AdalineTraceExporter
participant AdalineAPI as Adaline Trace API
User->>AdalineTracer: new AdalineTracer(options)
AdalineTracer->>AdalineTraceExporter: new AdalineTraceExporter(options)
AdalineTracer->>NodeTracerProvider: new NodeTracerProvider(resource)
AdalineTracer->>BatchSpanProcessor: new BatchSpanProcessor(exporter)
AdalineTracer->>NodeTracerProvider: addSpanProcessor(processor)
AdalineTracer->>NodeTracerProvider: getTracer('adaline-gateway')
NodeTracerProvider-->>AdalineTracer: tracer
User->>Gateway: new Gateway({ telemetry: { tracer } })
Gateway->>Gateway: TelemetryManager.setTracer(tracer)
User->>Gateway: completeChat(request)
Gateway->>Gateway: tracer.startActiveSpan('complete-chat')
Gateway->>Gateway: Execute LLM call
Gateway->>Gateway: span.end()
Gateway->>BatchSpanProcessor: Queue span
BatchSpanProcessor->>AdalineTraceExporter: export(spans)
AdalineTraceExporter->>AdalineTraceExporter: groupSpansByTrace(spans)
AdalineTraceExporter->>AdalineTraceExporter: sendTrace(traceId, spans)
AdalineTraceExporter->>AdalineTraceExporter: convertSpan(span)
AdalineTraceExporter->>AdalineAPI: POST /v2/api/logs/trace
AdalineAPI-->>AdalineTraceExporter: 200 OK
AdalineTraceExporter-->>BatchSpanProcessor: ExportResultCode.SUCCESS
User->>AdalineTracer: shutdown()
AdalineTracer->>NodeTracerProvider: shutdown()
NodeTracerProvider->>BatchSpanProcessor: shutdown()
BatchSpanProcessor->>AdalineTraceExporter: forceFlush()
|
![greptile-apps[bot]](https://avatars.githubusercontent.com/in/867647?s=60&v=4){kind=small}
Comment on lines
+164
to
+171
| const response = await fetch(this.endpoint, { | ||
| method: 'POST', | ||
| headers: { | ||
| 'Authorization': `Bearer ${this.apiKey}`, | ||
| 'Content-Type': 'application/json', | ||
| }, | ||
| body: JSON.stringify(trace), | ||
| }); |
Contributor
There was a problem hiding this comment.
logic: timeout option is not applied to the fetch call
Suggested change
| const response = await fetch(this.endpoint, { | |
| method: 'POST', | |
| headers: { | |
| 'Authorization': `Bearer ${this.apiKey}`, | |
| 'Content-Type': 'application/json', | |
| }, | |
| body: JSON.stringify(trace), | |
| }); | |
| const controller = new AbortController(); | |
| const timeoutId = setTimeout(() => controller.abort(), this.timeout); | |
| try { | |
| const response = await fetch(this.endpoint, { | |
| method: 'POST', | |
| headers: { | |
| 'Authorization': `Bearer ${this.apiKey}`, | |
| 'Content-Type': 'application/json', | |
| }, | |
| body: JSON.stringify(trace), | |
| signal: controller.signal, | |
| }); | |
| clearTimeout(timeoutId); |
Prompt To Fix With AI
This is a comment left during a code review.
Path: core/gateway/src/plugins/telemetry/adaline-trace.exporter.ts
Line: 164:171
Comment:
**logic:** `timeout` option is not applied to the fetch call
```suggestion
const controller = new AbortController();
const timeoutId = setTimeout(() => controller.abort(), this.timeout);
try {
const response = await fetch(this.endpoint, {
method: 'POST',
headers: {
'Authorization': `Bearer ${this.apiKey}`,
'Content-Type': 'application/json',
},
body: JSON.stringify(trace),
signal: controller.signal,
});
clearTimeout(timeoutId);
```
How can I resolve this? If you propose a fix, please make it concise.
Author
|
A question for this is how we should generally handle the 1MB limit with Adaline trace. I've ran into this issue before. |
Contributor
|
is this tested? The package doesnt builds with this changes for me! related logs |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Adds native support for exporting OpenTelemetry traces to Adaline Trace, the hosted observability platform for LLM applications.
Changes:
Features:
Usage:
All Gateway operations are automatically traced with full observability: model names, cache hits, latency, time-to-first-token, and span hierarchies.