Downloaded 28 times
Uploaded byBhakti Mehta
Expect the unexpected: Anticipate and prepare for failures in microservices based architectures
Bhakti Mehta discusses strategies for building resilient microservices architectures. Mehta covers challenges like cascading failures and latency that can occur at scale. Techniques like circuit breakers, timeouts, retries, and bulkheading are presented to isolate failures and prevent them from spreading. Logging and metrics are also important for monitoring systems and identifying issues after deployment. The talk emphasizes anticipating failures through approaches like load testing and designing systems to automatically recover from failures.
More Related Content
![Top 50 java ee 7 best practices [con5669]](https://cdn.slidesharecdn.com/ss_thumbnails/top50javaee7bestpracticescon5669-160927041126-thumbnail.jpg?width=640&height=640&fit=bounds)
Similar to Expect the unexpected: Anticipate and prepare for failures in microservices based architectures
![ANPARA THERMAL POWER STATION[1] sangam.pdf](https://cdn.slidesharecdn.com/ss_thumbnails/anparathermalpowerstation1sangam-251121115219-9261cde4-thumbnail.jpg?width=640&height=640&fit=bounds)
Expect the unexpected: Anticipate and prepare for failures in microservices based architectures
- 1. Expect the unexpected: Anticipateand prepare for failures in microservices based architectures Bhakti Mehta @bhakti_mehta
- 2. Introduction • Senior SoftwareEngineer at Blue Jeans Network • Worked at Sun Microsystems/Oracle for 13 years • Committer to numerous open source projects including GlassFish Application Server
- 3.
- 4.
- 5. Blue Jeans Network •Video conferencing in the cloud • 4000+ customers • Millions of users
- 6. What you willlearn • Monoliths v/s microservices • Challenges at scale • Preventing Cascading failures • Resilience planning at various stages • Dealing with latencies in response • Real world examples
- 7.
- 8.
- 9. Microservices Billing Provisioning accounts Notification Meeting Amicro service based application puts each element of functionality in a separate service
- 10.
- 11. Microservices • Advantages – Simplicity –Isolation of problems – Scale up and scale down – Easy deployment – Clear separation of concerns – Heterogeneity and polyglotism
- 12. Microservices • Disadvantages – Nota free lunch! – Distributed systems prone to failures – Eventual consistency – More effort in terms of deployments, release managements – Challenges in testing the various services evolving independently, regression tests etc
- 13.
- 14. Resilient system • Processestransactions, even when there are transient impulses, persistent stresses • Functions even when there are component failures disrupting normal processing • Accepts failures will happen • Designs for crumple zones
- 15. Kinds of failures •Challenges at scale • Integration point failures – Network errors – Semantic errors. – Slow responses – Outright hang – GC issues
- 18. Anticipate failures atscale • Anticipate growth • Design for next order of magnitude • Design for 10x plan to rewrite for 100x
- 19. Resiliency planning Stage1 • When developing code – Avoiding Cascading failures • Circuit breaker • Timeouts • Retry • Bulkhead • Cache optimizations – Avoid malicious clients • Rate limiting
- 20. Resiliency planning Stage2 • Planning for dealing with failures before deploy – load test – a/b test – longevity
- 21. Resiliency planning Stage3 • Watching out for failures after deploy – health check – metrics
- 23. Cascading failures Caused byChain reactions For example One node in a load balance group fails Others need to pick up work Eventually performance can degenerate
- 24. Cascading failures withaggregation
- 25. Cascading failure withaggregation
- 27. Timeouts • Clients mayprefer a response – failure – success – job queued for later All aggregation requests to microservices should have reasonable timeouts set
- 28. Types of Timeouts •Connection timeout – Max time before connection can be established or Error • Socket timeout – Max time of inactivity between two packets once connection is established
- 29. Timeouts pattern • Timeouts+ Retries go together • Transient failures can be remedied with fast retries • However problems in network can last for a while so probability of retries failing
- 30. Timeouts in code InJAX-RS Client client = ClientBuilder.newClient(); client.property(ClientProperties.CONNECT_TIMEOUT, 5000); client.property(ClientProperties.READ_TIMEOUT, 5000)
- 31. Retry pattern • Retryfor failures in case of network failures, timeouts or server errors • Helps transient network errors such as dropped connections or server fail over
- 32. Retry pattern • Ifone of the services is slow or malfunctioning and other services keep retrying then the problem becomes worse • Solution – Exponential backup – Circuit breaker pattern
- 33. Circuit breaker pattern Circuitbreaker A circuit breaker is an electrical device used in an electrical panel that monitors and controls the amount of amperes (amps) being sent through
- 34. Circuit breaker pattern •Safety device • If a power surge occurs in the electrical wiring, the breaker will trip. • Flips from “On” to “Off” and shuts electrical power from that breaker
- 35. Circuit breaker • NetflixHystrix follows circuit breaker pattern • If a service’s error rate exceeds a threshold it will trip the circuit breaker and block the requests for a specific period of time
- 36.
- 37. Bulkhead • Avoiding chainreactions by isolating failures • Helps prevent cascading failures
- 38. Bulkhead • An exampleof bulkhead could be isolating the database dependencies per service • Similarly other infrastructure components can be isolated such as cache infrastructure
- 39. Rate Limiting • Restrictingthe number of requests that can be made by a client • Client can be identified based on the access token used • Additionally clients can be identified based on IP address
- 40. Rate Limiting • WithJAX-RS Rate limiting can be implemented as a filter • This filter can check the access count for a client and if within limit accept the request • Else throw a 429 Error • Code at https://github.com/bhakti- mehta/samples/tree/master/ratelimiting
- 41. Cache optimizations • Storesresponse information related to requests in a temporary storage for a specific period of time • Ensures that server is not burdened processing those requests in future when responses can be fulfilled from the cache
- 42. Cache optimizations Getting fromfirst level cache Getting from second level cache Getting from the DB
- 43. Dealing with latenciesin response • Have a timeout for the aggregation service • Dispatch requests in parallel and collect responses • Associate a priority with all the responses collected
- 44. Handling partial failuresbest practices • One service calls another which can be slow or unavailable • Never block indefinitely waiting for the service • Try to return partial results • Provide a caching layer and return cached data
- 45. Asynchronous Patterns • Patternto deal with long running jobs • Some resources may take longer time to provide results • Not needing client to wait for the response
- 46. Reactive programming model •Use reactive programming such as CompletableFuture in Java 8, ListenableFuture • Rx Java
- 47. Asynchronous API • Reactivepatterns • Message Passing – Akka actor model • Message queues – Communication between services via shared message queues – Websockets
- 48. Logging • Complex distributedsystems introduce many points of failure • Logging helps link events/transactions between various components that make an application or a business service • ELK stack • Splunk, syslog • Loggly • LogEntries
- 49. Logging best practices •Include detailed, consistent pattern across service logs • Obfuscate sensitive data • Identify caller or initiator as part of logs • Do not log payloads by default
- 50. Best practices whendesigning APIs for mobile clients – Avoid chattiness – Use aggregator pattern
- 51. Resilience planning Stage2 • Before deploy – Load testing – Longevity testing – Capacity planning
- 52. Load testing • Ensurethat you test for load on APIs – Jmeter • Plan for longevity testing
- 53. Capacity Planning • Anticipategrowth • Design for handling exponential growth
- 54. Resilience planning Stage3 • After deploy – Health check – Metrics – Phased rollout of features
- 56. Health Check • Memory •CPU • Threads • Error rate • If any of the checks exceed a threshold send alert
- 58. Metrics • Response times,throughput – Identify slow running DB queries • GC rate and pause duration – Garbage collection can cause slow responses • Monitor unusual activity • Third party library metrics – For example Couchbase hits – atop
- 59. Rollout of newfeatures • Phasing rollout of new features • Have a way to turn features off if not behaving as expected • Alerts and more alerts!
- 60. Real time examples •Netflix's Simian Army induces failures of services and even datacenters during the working day to test both the application's resilience and monitoring. • Latency Monkey to simulate slow running requests • Wiremock to mock services • Saboteur to create deliberate network mayhem
- 61. Takeaway • Inevitability offailures – Expect systems will fail – Failure prevention
- 63. References • https://commons.wikimedia.org/wiki/File:Bulkhead_PSF.png • https://en.wikipedia.org/wiki/Circuit_breaker#/media/File:Four_1_pole_circuit_breakers_fitted_in_a_met er_box.jpg •https://www.flickr.com/photos/skynoir/ Beer in hand: skynoir/Flickr/Creative Commons License
Editor's Notes
- #8 The service will have a caching layer and a database layer