Know your app: Add metrics to Java with Micrometer | DevNation Tech Talk
1 like4,201 views
The document discusses collecting metrics in Java applications using Micrometer, an open-source metrics library. It covers key concepts such as observable systems, time series data, metric types (e.g., counters, timers, gauges), and provides examples of how to implement these metrics effectively. It emphasizes the importance of metrics for system monitoring and performance analysis rather than debugging.
![© 2020 IBM Corporation
Attacks : Hits and Misses
• AC == Armor class
Attacker rolls a d20...
20 – critical hit (HIT)
1 – critical miss (MISS)
• DC == Dif
fi
culty class
Defender rolls a d20...
: oneRound:
Troll(LARGE GIANT){AC:15,HP:84(8d10+40),STR:18(+4),DEX:13(+1),CON:20(+5),INT:7(-2),WIS:9(-1),C
Pit Fiend(LARGE FIEND){AC:19,HP:300(24d10+168),STR:26(+8),DEX:14(+2),CON:24(+7),INT:22(+6),W
: attack: miss: Troll(36) -> Pit Fiend(100)
: attack: miss: Troll(36) -> Pit Fiend(100)
: attack: hit> Troll(36) -> Pit Fiend(97) for 9 damage using Claws[7hit,11(2d6+4)|slashing]
: attack: hit> Pit Fiend(97) -> Troll(10) for 22 damage using Bite[14hit,22(4d6+8)|piercing]
: attack: MISS: Pit Fiend(97) -> Troll(10)
: attack: HIT> Pit Fiend(97) -> Troll(0) for 34 damage using Mace[14hit,15(2d6+8)|bludgeoning]
: oneRound: survivors
Pit Fiend(LARGE FIEND){AC:19,HP:300(24d10+168),STR:26(+8),DEX:14(+2),CON:24(+7),INT:22(+6),W](https://image.slidesharecdn.com/2021-micrometer-devnation6-10-21-210610173000/85/Know-your-app-Add-metrics-to-Java-with-Micrometer-DevNation-Tech-Talk-19-320.jpg)
Know your app: Add metrics to Java with Micrometer | DevNation Tech Talk
- 1. Erin Schnabel @ebullientworks Know Your (Java) App Collect metrics with Micrometer
- 2. What makes a system (or service) observable? Workload routing System health Service-centric problem determination Context + relationships for end–to-end analysis Statistics & trends Analytics Service is ready Service is not a zombie How many times was method x called? Method x was called What happened when method x was called Health Checks Distributed Trace Metrics Log Entries
- 3. Metrics: Time series data How does data change over time? • Time is the x-axis • Data collected at regular intervals • Each measured value appended to the end of a series • Used for trend analysis, anomaly detection, alerting Not for debugging, diagnostics, or root cause analysis
- 4. Metrics: Dimensions Filtered/Selective aggregation http_server_requests_seconds_count{ env="test", method="GET", outcome="SUCCESS", status="200", uri="/example/prime/{number}",} 5.0 Cardinality: Each unique combination (name & all tag/values) is its own time series
- 5. What is Micrometer? https://micrometer.io -- Open Source metrics library "Screw gauge" micrometer
- 6. Developer API • Obtain a MeterRegistry • Create a Meter (name + tag=value combination) Gauge Counter Timer DistributionSummary FunctionCounter FunctionTimer, LongTaskTimer
- 7. Counter: increase only → .increment() 1. registry.counter("example.prime.number", "type", "prime"); 2. Counter.builder("counter") .baseUnit("beans") / / optional .description("a description") / / optional .tags("region", "test") / / optional .register(registry); 3. @Counted(value = "metric.all", extraTags = { "region", "test" }) * *
- 8. Timer → .record(duration) 1. registry.timer("my.timer", "region", "test"); 2. Timer.builder("my.timer") .description("description ") / / optional .tags("region", "test") / / optional .register(registry); 3. @Timed(value = "call", extraTags = {"region", "test"}) * *
- 9. Working with Timers 1. timer.record(() - > noReturnValue()); 2. timer.recordCallable(() - > returnValue()); 3. Runnable r = timer.wrap(() - > noReturnValue()); 4. Callable c = timer.wrap(() - > returnValue()); 5. Sample s = Timer.start(registry); doStuff; s.stop(timer);
- 10. Gauge: increase/decrease, observed value 1. List<String> list = registry.gauge("list.size", Tags.of( . . . ), new ArrayList < > (), List : : size); 2. Gauge.builder("jvm.threads.peak", threadBean, ThreadMXBean : : getPeakThreadCount) .tags("region", "test") .description("The peak live thread count . . . ") .baseUnit(BaseUnits.THREADS) .register(registry);
- 11. FunctionCounter: gauge-like, counter function 1. registry.more().counter( . . . ) 2. FunctionCounter.builder("hibernate.transactions", statistics, s - > s.getTransactionCount() - s.getSuccessfulTransactionCount()) .tags("entityManagerFactory", sessionFactoryName) .tags("result", "failure") .description("The number of transactions we know to have failed") .register(registry);
- 12. FunctionTimer: gauge-like, counter & duration function 1. registry.more().timer( . . . ); 2. FunctionTimer.builder("cache.gets.latency", cache, c - > c.getLocalMapStats().getGetOperationCount(), c - > c.getLocalMapStats().getTotalGetLatency(), TimeUnit.NANOSECONDS) .tags("name", cache.getName()) .description("Cache gets") .register(registry);
- 13. LongTaskTimer (active count, longest active task) 1. registry.more().longTaskTimer("long.task", "region", "test"); 2. LongTaskTimer.builder("long.task") .description("a description") / / optional .tags("region", "test") / / optional .register(registry); 3. @Timed(value = "long.task", extraTags = {"extra", "tag"}, longTask = true) * *
- 14. Distribution Summary → .record(value), not time 1. registry.summary("name", "region", "test") 2. DistributionSummary.builder("response.size") .description("a description") / / optional .baseUnit("bytes") / / optional .tags("region", "test") / / optional .register(registry);
- 15. Library Maintainer API • Provide a MeterBinder (optional dependency) • Register library-speci fi c metrics • Quarkus / Spring discover MeterBinder implementations binder.bindTo(registry)
- 16. Configuration and Filtering • Late-binding con fi guration provided by MeterFilter • Sourced from user code or shared library • Accept/Deny fi lters: fi ne-grained control over gathered metrics • Transform: rename metrics, add/ignore/replace/rename tags • Customize Timer / Histogram behavior • Percentiles, quantiles, Service Level Agreement(SLA) boundaries
- 18. Rolling dice Micrometer: Dice.setMonitor((k, v) -> { registry.counter("dice.rolls", "die", k, "face", label(v)) .increment(); });
- 19. © 2020 IBM Corporation Attacks : Hits and Misses • AC == Armor class Attacker rolls a d20... 20 – critical hit (HIT) 1 – critical miss (MISS) • DC == Dif fi culty class Defender rolls a d20... : oneRound: Troll(LARGE GIANT){AC:15,HP:84(8d10+40),STR:18(+4),DEX:13(+1),CON:20(+5),INT:7(-2),WIS:9(-1),C Pit Fiend(LARGE FIEND){AC:19,HP:300(24d10+168),STR:26(+8),DEX:14(+2),CON:24(+7),INT:22(+6),W : attack: miss: Troll(36) -> Pit Fiend(100) : attack: miss: Troll(36) -> Pit Fiend(100) : attack: hit> Troll(36) -> Pit Fiend(97) for 9 damage using Claws[7hit,11(2d6+4)|slashing] : attack: hit> Pit Fiend(97) -> Troll(10) for 22 damage using Bite[14hit,22(4d6+8)|piercing] : attack: MISS: Pit Fiend(97) -> Troll(10) : attack: HIT> Pit Fiend(97) -> Troll(0) for 34 damage using Mace[14hit,15(2d6+8)|bludgeoning] : oneRound: survivors Pit Fiend(LARGE FIEND){AC:19,HP:300(24d10+168),STR:26(+8),DEX:14(+2),CON:24(+7),INT:22(+6),W
- 20. Micrometer: Timer & Distribution Summary • Records value registry.summary("round.attacks", "hitOrMiss", event.hitOrMiss(), "attackType", event.getAttackType(), "damageType", event.getType()) .record((double) event.getDamageAmount()); • Default metrics: count, sum, max • Optional: cumulative histogram buckets, pre-computed quantile
- 21. So many bugs... There is often no practical way to write tests for all permutations... This ended up being a problem with the original conversion from HTML to JSON. Who wrote this code, anyway?
- 22. Questions?