Spring i/o: Deep and Brief Dive Into Causality

Slides from my presentation at Spring i/o 2024:
https://2024.springio.net/sessions/navigating-the-complexities-of-service-to-service-invocations-deep-and-brief-dive-into-causality/

1. Navigating the Complexity of
Service to Service Invocation

Deep and Brief Dive Into
causality

3. Cause -> Effect

4. Spring i/0
-How did a talk Influence your Perspective?

-Which new Tool will you try out next week?

-Did any networking interaction lead to a
significant connection?

5. Causality in tech
High traffic -> increase latency SLO’s
Alerts
Business Problem

8. @nele_lea


@[email protected]
@nlea
Nele Lea Uhlemann

11. Call Graphs Traces Workflows

12. Call Graphs
Service A
Micrometer
Service B
Micrometer

14. Call Graph with a
Service Mesh
Service A
Envoy
Service B
Envoy

15. DEMO

19. Trace 1
Span 1.1
Span 1.2
Span 1.3
Span 1.4

20. Service A Service B Service C
Request
header
Request
header
Trace 1
Span 1.1 Span 1.2 Span 2.1 Span 2.2
Trace 2

21. Service A
Service B
Service C
Request
header
Request
header
Trace 1
Span 1.1 Span 1.2 Span 1.3 Span 1.4
Context
Propagation

22. Service A
Service B Service C
Service Invocation Besides HTTP
Message Broker

23. TRACES
Service A
Auto instrumentation
Manual
instrumentation

24. OpenTelemetry
Application
Instrument
Telemetry data
Query-,
alerting-,
visualization
Platform
Observibility
data backend

25. DEMO

26. Ice Cream Recommender
Recommen-
dation
Service
Location
Service
Weather
Service
AI Service
Kafka

27. Ice Cream Recommender
Recommen-
dation
Service
Location
Service
Weather
Service
AI Service
Kafka

30. Traces

34. Traces
Metrics
Logs
< Exemplars
Tags >
Trace ID
https://spring.io/blog/2022/10/12/observability-with-spring-boot-3

35. Where is my Ice Cream
Recommendation?

36. Workflows
User A
Service A
Activity BActivity A
Event

37. Ice Cream Recommender
Recommen-
dation
Service
Location
Service
Weather
Service
AI Service
Kafka

38. workflow
Recommendation
Workflow
Application
Location
Service
Get Weather
Activity
REcommender
Service
Request Geo
Data Activity
Request Recommendation
Activity
Geo data
received
Recommendation
received
Kafka

39. DEMO
Service A
+
Workflow
Service B

41. public class IceCreamRecommendationWorkflow extends Workflow {
@Override
public WorkflowStub create() {
return ctx-> {
Logger logger = ctx.getLogger();
String instanceId = ctx.getInstanceId();
logger.info("Instance ID: " + instanceId);

RecommendationPayload recommendationPayload = ctx.getInput(RecommendationPayload .class);
logger.info("Received recommendation request for: " +recommendationPayload .toString());
recommendationPayload .setWorkflowID(instanceId);

//Send Location name to Location Service
Location location = new Location();
location.setName(recommendationPayload .getLocation());
location.setWorkflowID(instanceId);
ctx.callActivity(RequestLocationDetailsActivity .class.getName(), location).await();
System.out.println("Workflow waits for geo Info" + instanceId);

//Wait for the geo information
Geodata geodata = null;
try {
geodata = ctx.waitForExternalEvent( "geoInformationRecieved" , Duration.ofDays(1), Geodata.class).await();
System.out.println("Workflow received geo Info " + instanceId);
} catch (TaskCanceledException e ) {
Recommendation norecommendation = new Recommendation();
norecommendation.setContent("Sorry we can't provide and recommendation for your location" );
ctx.complete(norecommendation);
}
//More things

ctx.complete(recommendation);

};
}
}

43. Same Same but Different

44. -Network Traffic
-Service to Service Invocation
-Based On traces
-Context propagation
-Outer Edges
-Observability
-Auto or Manual
Instrumentation
-Customize Span definitions
-Inside the application
Context
-Observability
-Application Design/ Build
-State Management
-Rollbacks, Timeouts etc
-Business Case

45. @nele_lea


@[email protected]
@nlea