Why monitoring is an analytics problem
1 like261 views
SignalFx uses an analytics approach to monitoring microservices. They instrument each microservice to collect metrics and log exceptions. All data is sent to an analytics service for analysis and detection of issues. For example, when a code push caused unexpected timeouts in an upstream service, SignalFx was able to quickly identify the outlier service through exception analysis, narrow down the problem to a specific exception, and find the root cause in just 15 minutes. Their analytics approach helps reduce mean time to resolution compared to traditional monitoring.
![Experience
[2013 - ] SignalFx - Founder, CTO, Software Engineer
Microservices; Monitoring using Analytics
[2008 - 2012] Facebook - Software Engineer, Software Architect
Hyperscale SOA; Monitoring using Nagios, Ganglia, and in-house
Analytics
[2004 - 2008] Opsware - Chief Architect, Software Engineer
Monolithic Architecture; Monitoring using Ganglia, Nagios, Splunk
[2000 - 2004] Loudcloud - Software Engineer
LAMP, Application Server; Monitoring using SNMP, Ganglia, NetCool
[1998 - 2000] Marimba - Software Engineer
Client / Server; Monitoring using SNMP, FreshWater Software
[ … ]](https://image.slidesharecdn.com/whymonitoringisananalyticsproblem-150521185753-lva1-app6892/85/Why-monitoring-is-an-analytics-problem-5-320.jpg)
![Code Push Side Effects
java.io.InvalidObjectException: enum constant MURMUR128_MITZ_64 does
not exist in class com.google.common.hash.BloomFilterStrategies
at java.io.ObjectInputStream.readEnum(ObjectInputStream.java:1743) ~[na:
1.7.0_79]
at java.io.ObjectInputStream.readObject0(ObjectInputStream.java:1347)
~[na:1.7.0_79]
at java.io.ObjectInputStream.defaultReadFields(ObjectInputStream.java:
1990) ~[na:1.7.0_79]
…
Looking at Analytic’s logs revealed
source of the problem.](https://image.slidesharecdn.com/whymonitoringisananalyticsproblem-150521185753-lva1-app6892/85/Why-monitoring-is-an-analytics-problem-24-320.jpg)
Why monitoring is an analytics problem
- 1. SignalFx Microservices and Devs in Charge: Why Monitoring is an Analytics Problem
- 2. SignalFx Microservices and Devs in Charge: Why Monitoring is an Analytics Problem Phillip Liu phillip@signalfx.com @SignalFx - signalfx.com
- 3. Agenda • My background • Microservices, a review • Analytics approach to monitoring • Code push side effects, an example • Summary
- 5. Experience [2013 - ] SignalFx - Founder, CTO, Software Engineer Microservices; Monitoring using Analytics [2008 - 2012] Facebook - Software Engineer, Software Architect Hyperscale SOA; Monitoring using Nagios, Ganglia, and in-house Analytics [2004 - 2008] Opsware - Chief Architect, Software Engineer Monolithic Architecture; Monitoring using Ganglia, Nagios, Splunk [2000 - 2004] Loudcloud - Software Engineer LAMP, Application Server; Monitoring using SNMP, Ganglia, NetCool [1998 - 2000] Marimba - Software Engineer Client / Server; Monitoring using SNMP, FreshWater Software [ … ]
- 7. A Microservices Definition Loosely coupled service oriented architecture with bounded context. Adrian Cockcroft
- 8. SignalFx’s Microservices More than 15 internal services. Spanning hundreds of instances. Across 3 AZs. Have dependencies on tens of external services.
- 9. Monitoring Challenges • High iteration rate leads to shortened test cycles • Integration test combinations are intractable • Catch problems during rolling deployments • Identify upstream/downstream side effects • e.g. backpressure • Identify brownouts before the customer • etc.
- 10. SignalFx Analytics Approach to Monitoring
- 11. Measure
- 12. Store
- 13. Analyze
- 14. Detect
- 16. Monitoring at SignalFx •We use SignalFx to monitor SignalFx •CollectD for OS and Docker metrics on all VMs •Yammer metrics for all Java app servers •Custom logger to count exception types •All metrics are sent to an analytics service •Each service deploy a their cadence •Push lab, then canary in prod, then rest of tier
- 17. Code Push Side Effects
- 18. Code Push Side Effects Push canary instance and Metadata API dashboard shows healthy tier.
- 19. Code Push Side Effects However, upstream UI dashboard showed unusual # of timeouts.
- 20. Code Push Side Effects In search of root cause. Always safe to start by looking at exception counts. Can’t derive much from all the noise.
- 21. Code Push Side Effects Sum the # of exceptions to create a single signal.
- 22. Code Push Side Effects Compare sum with time-shifted sum from a day ago.
- 23. Code Push Side Effects Look at an outlier host - an Analytics service host.
- 24. Code Push Side Effects java.io.InvalidObjectException: enum constant MURMUR128_MITZ_64 does not exist in class com.google.common.hash.BloomFilterStrategies at java.io.ObjectInputStream.readEnum(ObjectInputStream.java:1743) ~[na: 1.7.0_79] at java.io.ObjectInputStream.readObject0(ObjectInputStream.java:1347) ~[na:1.7.0_79] at java.io.ObjectInputStream.defaultReadFields(ObjectInputStream.java: 1990) ~[na:1.7.0_79] … Looking at Analytic’s logs revealed source of the problem.
- 25. Code Push Side Effects • Analytics across multiple microservices reduced time to identify problem. From push to resolution was ~15min • Service instrumentation helped narrowed down root cause • Discovery allowed us to create a detector using analytics to notify similar problems in the future
- 26. Other Examples • A customer started dropping data because they reverted to an unsupported API • Compare tsdb write throughput of two different write strategies • Create per-service capacity reports • Identify memory usage patterns across our Analytics service • Create a detector for every previously uncaught error conditions - postmortem output
- 27. SignalFx Summary
- 28. • Measure and Store as much metrics and events as possible • Use data analytics techniques to • Identify problems • Chase down root cause • Create analytics based detectors to notify you of recurrence
- 29. SignalFx Thank You! Phillip Liu phillip@signalfx.com WE’RE HIRING jobs@signalfx.com @SignalFx - signalfx.com