Download to read offline
![if [ "$1" = "deploy" ] && [ $# -eq 4 ]; then
STAGE=$2
REGION=$3
PROFILE=$4
npm install
AWS_PROFILE=$PROFILE 'node_modules/.bin/sls' deploy -s $STAGE -r $REGION
elif [ "$1" = "int-test" ] && [ $# -eq 4 ]; then
STAGE=$2
REGION=$3
PROFILE=$4
npm install
AWS_PROFILE=$PROFILE npm run int-$STAGE
elif [ "$1" = "acceptance-test" ] && [ $# -eq 4 ]; then
STAGE=$2
REGION=$3
PROFILE=$4
npm install
AWS_PROFILE=$PROFILE npm run acceptance-$STAGE
else
usage
exit 1
fi](https://image.slidesharecdn.com/serverlessinprodlondonjscommunity-180206225121/75/Serverless-in-production-an-experience-report-London-js-community-94-2048.jpg)
Uploaded byYan Cui
Serverless in production, an experience report (London js community)
The document discusses key considerations and best practices for deploying serverless applications in production, highlighting the advantages of AWS Lambda over traditional EC2 deployments, including cost savings and faster release cycles. It emphasizes the importance of effective testing, monitoring, and configuration management to ensure system reliability and efficiency. It also covers various tools and techniques for improving development workflows, minimizing downtime, and managing configuration and security in serverless architectures.
More Related Content
Similar to Serverless in production, an experience report (London js community)
Serverless in production, an experience report (London js community)
- 1. in production an experiencereportan experience report what you should know before you go to production ServerlessServerless
- 2.
- 3. Yan Cui Server Architect PrincipalEngineer Lead Developer Senior Developer http://theburningmonk.com @theburningmonk Senior Developer
- 4. Yan Cui Server Architect PrincipalEngineer Lead Developer Senior Developer http://theburningmonk.com @theburningmonk Senior Developer
- 5. Diana Ionita Senior Developer SeniorDeveloper Senior Developer
- 8.
- 9. hidden complexities anddependencies low utilisation to leave room for traffic spikes EC2 scaling is slow, so scale earlier lots of cost for unused resources up to 30 mins for deployment deployment required downtime
- 10. - Dan North “leadtime to someone saying thank you is the only reputation metric that matters.”
- 12. “what would good looklike for us?”
- 13. be small be fast havezero downtime have no lock-step DEPLOYMENTS SHOULD...
- 14. FEATURES SHOULD... be deployableindependently be loosely-coupled
- 15. WE WANT TO... minimisecost for unused resources minimise ops effort reduce tech mess deliver visible improvements faster
- 16.
- 17. 170 Lambda functionsin prod 1.2 GB deployment packages in prod 95% cost saving vs EC2 15x no. of prod releases per month
- 18.
- 19. 1st function inprod! time is a good fit
- 20. ? time is a goodfit 1st function in prod!
- 21.
- 22. Practices ToolsPrinciples what isgood? how to make it good? with what?
- 23.
- 24. 170 functions WOOF! ? ? time isa good fit 1st function in prod!
- 25.
- 26.
- 27.
- 28. Legacy Monolith AmazonKinesis Amazon Lambda Amazon CloudSearch
- 29. Legacy Monolith AmazonKinesis Amazon Lambda Amazon CloudSearchAmazon API Gateway Amazon Lambda
- 30.
- 31. Legacy Monolith AmazonKinesis Amazon Lambda Google BigQuery
- 32. Legacy Monolith AmazonKinesis Amazon Lambda Google BigQuery 1 developer, 2 days design production (his 1st serverless project)
- 33. Legacy Monolith AmazonKinesis Amazon Lambda Google BigQuery “nothing ever got done this fast at Skype!” - Chris Twamley
- 34. - Dan North “leadtime to someone saying thank you is the only reputation metric that matters.”
- 35.
- 42.
- 43.
- 44.
- 45.
- 46.
- 47.
- 48.
- 49.
- 50.
- 51.
- 52.
- 53.
- 54.
- 55.
- 56.
- 57.
- 58.
- 59. Level of Testing 1.Unit doour objects do the right thing? are they easy to work with?
- 61. Level of Testing 1.Unit 2.Integration doesour code work against code we can’t change?
- 62.
- 63. handler test by invoking thehandler
- 64. Level of Testing 1.Unit 2.Integration 3.Acceptance doesthe whole system work?
- 65.
- 66. “…We find thattests that mock external libraries often need to be complex to get the code into the right state for the functionality we need to exercise. The mess in such tests is telling us that the design isn’t right but, instead of fixing the problem by improving the code, we have to carry the extra complexity in both code and test…” Don’t Mock Types You Can’t Change
- 67. “…The second riskis that we have to be sure that the behaviour we stub or mock matches what the external library will actually do… Even if we get it right once, we have to make sure that the tests remain valid when we upgrade the libraries…” Don’t Mock Types You Can’t Change
- 68. Don’t Mock TypesYou Can’t Change Services
- 69. Paul Johnston The serverlessapproach to testing is different and may actually be easier. http://bit.ly/2t5viwK
- 70.
- 71.
- 72. LambdaAPI Gateway DynamoDB UnitTests Mock/Stub
- 73. is our requestcorrect? is the request mapping set up correctly?is the API resources configured correctly? are we assuming the correct schema? LambdaAPI Gateway DynamoDB is Lambda proxy configured correctly? is IAM policy set up correctly? is the table created? what unit tests will not tell you…
- 75. most Lambda functionsare simple have single purpose, the risk of shipping broken software has largely shifted to how they integrate with external services observation
- 77. But it slowsdown my feedback loop… IT’S NOT ABOUT YOU!
- 78. …if a servicecan’t provide you with a relatively easy way to test the interface in reality, then you should consider using another one. Paul Johnston
- 79. “…Wherever possible, anacceptance test should exercise the system end-to- end without directly calling its internal code. An end-to-end test interacts with the system only from the outside: through its interface…” Testing End-to-End
- 80. Legacy Monolith AmazonKinesis Amazon Lambda Amazon CloudSearchAmazon API Gateway Amazon Lambda
- 81. Legacy Monolith AmazonKinesis Amazon Lambda Amazon CloudSearchAmazon API Gateway Amazon Lambda Test Input
- 82. Legacy Monolith AmazonKinesis Amazon Lambda Amazon CloudSearchAmazon API Gateway Amazon Lambda Test Input Validate
- 83. integration tests exercise system’sIntegration with its external dependencies my code
- 84. acceptance tests exercise systemEnd-to-End from the outside my code
- 85. integration tests differfrom acceptance tests only in HOW the Lambda functions are invoked observation
- 89. CI + CDPIPELINE
- 90. “the earlier youconsider CI + CD, the more time you save in the long run” - me
- 91. “…We prefer tohave the end-to-end tests exercise both the system and the process by which it’s built and deployed… This sounds like a lot of effort (it is), but has to be done anyway repeatedly during the software’s lifetime…” Testing End-to-End
- 92. “deployment scripts that onlylive on the CI box is a disaster waiting to happen” - me
- 93. Jenkins build configdeploys and tests unit + integration tests deploy acceptance tests
- 94. if [ "$1"= "deploy" ] && [ $# -eq 4 ]; then STAGE=$2 REGION=$3 PROFILE=$4 npm install AWS_PROFILE=$PROFILE 'node_modules/.bin/sls' deploy -s $STAGE -r $REGION elif [ "$1" = "int-test" ] && [ $# -eq 4 ]; then STAGE=$2 REGION=$3 PROFILE=$4 npm install AWS_PROFILE=$PROFILE npm run int-$STAGE elif [ "$1" = "acceptance-test" ] && [ $# -eq 4 ]; then STAGE=$2 REGION=$3 PROFILE=$4 npm install AWS_PROFILE=$PROFILE npm run acceptance-$STAGE else usage exit 1 fi
- 95. build.sh allows repeatablebuilds on both local & CI
- 97.
- 98.
- 100. 2016-07-12T12:24:37.571Z 994f18f9-482b-11e6-8668-53e4eab441ae GOT isoff air, what do I do now?
- 101. 2016-07-12T12:24:37.571Z 994f18f9-482b-11e6-8668-53e4eab441ae GOT isoff air, what do I do now? UTC Timestamp API Gateway Request Id your log message
- 102.
- 103. Yan Logs are noteasily searchable in CloudWatch Logs.
- 104.
- 105.
- 106. CENTRALISE LOGS MAKE THEMEASILY SEARCHABLE
- 107. + + the elkstack
- 108.
- 109. CloudWatch Logs AWSLambda ELK stack
- 110.
- 111.
- 112.
- 116.
- 117.
- 119. “my followers didn’t receivemy new post!” - a user
- 120.
- 121. correlation IDs* * eg.request-id, user-id, yubl-id, etc.
- 122.
- 123.
- 124.
- 125.
- 126.
- 127.
- 128. traces do notspan over API Gateway
- 129.
- 130. “where do Iinstall monitoring agents?”
- 131.
- 132. • invocation Count •error Count • latency • throttling • granular to the minute • support custom metrics
- 133. • same metricsas CW • better dashboard • support custom metrics https://www.datadoghq.com/blog/monitoring-lambda-functions-datadog/
- 137.
- 138.
- 139. my code internet internet pressbutton something happens
- 140. “how do Ibatch up and send metrics in the background?”
- 141.
- 142. console.log(“hydrating yubls fromdb…”); console.log(“fetching user info from user-api”); console.log(“MONITORING|1489795335|27.4|latency|user-api-latency”); console.log(“MONITORING|1489795335|8|count|yubls-served”); timestamp metric value metric type metric namemetrics logs
- 143. CloudWatch Logs AWSLambda ELK stack logs metrics CloudWatch
- 144.
- 145.
- 146.
- 147.
- 148.
- 150. “you really don'twant your monitoring system to fail at the same time as the system it monitors” - Yan
- 151.
- 152. easily and quicklypropagate config changes
- 154. me Environment variables makeit hard to share configurations across functions.
- 155. me Environment variables makeit hard to implement fine-grained access to sensitive info.
- 156.
- 157.
- 161.
- 162. sensitive data shouldbe encrypted in-flight, and at rest (credentials, connection string, etc.)
- 163.
- 164. SSM Parameter Store HTTPS role-basedaccess encrypted in-flight
- 165.
- 166.
- 167. HTTPS role-based access SSM ParameterStore encrypted in-flight
- 168.
- 169. fetch & cacheat Cold Start
- 170. invalidate at interval+ signal
- 171.
- 172.
- 173. max 75 GBtotal deployment package size* * limit is per AWS region
- 174.
- 175.
- 176.
- 177. install Serverless frameworkas dev dependency at project level dev dependencies are excluded since 1.16.0
- 178.
- 179.
- 180.
- 181. Amazon X-Ray 1st invocation 2ndinvocation cold start
- 182.
- 183.
- 184.
- 185.
- 186.
- 187.
- 188. what about typesafety?
- 190. complexity ceiling ofa Node.js app complexity
- 191. complexity ceiling ofa Node.js app complexity referential transparency immutability as default type inference option types union types …
- 192. for managing complexity complexityceiling of a Node.js app complexity referential transparency immutability as default type inference option types union types …
- 193. complexity ceiling ofa Node.js app complexity complexity ceiling of a Node.js Lambda function
- 194. if you canlimit the complexity of your solution, maybe you won’t need the tools for managing that complexity. me
- 195.
- 196. CloudWatch Event AWSLambda
- 197. CloudWatch Event AWSLambda ping ping ping ping
- 198. CloudWatch Event AWSLambda ping ping ping ping
- 199. CloudWatch Event AWSLambda ping ping ping ping HEALTH CHECKS?
- 200.
- 201.
- 202. max 5 minsexecution time
- 203.
- 204.
- 205. “AWS Lambda pollsyour stream and invokes your Lambda function. Therefore, if a Lambda function fails, AWS Lambda attempts to process the erring batch of records until the time the data expires…” http://docs.aws.amazon.com/lambda/latest/dg/retries-on-errors.html
- 206. should function failon partial/any failures?
- 207. SNS Kinesis SQS after 3 attempts shareprocessing logic events are processed in chronological order failed events are retried out of sequence
- 208.
- 209.
- 210.
- 211. “Each shard cansupport up to 5 transactions per second for reads, up to a maximum total data read rate of 2 MB per second.” http://docs.aws.amazon.com/streams/latest/dev/service-sizes-and-limits.html
- 212. “If your streamhas 100 active shards, there will be 100 Lambda functions running concurrently. Then, each Lambda function processes events on a shard in the order that they arrive.” http://docs.aws.amazon.com/lambda/latest/dg/concurrent-executions.html
- 213. when no. ofprocessors goes up…
- 214. ReadProvisionedThroughputExceeded can have toomany Kinesis read operations…
- 215. ReadRecords.IteratorAge unpredictable spikes inread ‘latency’…
- 216.
- 217.
- 218.
- 219. for subsystems thatdon’t have to be realtime, or are task- based (ie. order doesn’t matter), consider other triggers such as S3 or SNS.me
- 220.
- 221. API Gateway andKinesis Authentication & authorisation (IAM, Cognito) Testing Running & Debugging functions locally Log aggregation Monitoring & Alerting X-Ray Correlation IDs CI/CD Performance and Cost optimisation Error Handling Configuration management VPC Security Leading practices (API Gateway, Kinesis, Lambda) http://bit.ly/2AA5zzk
- 222. API Gateway andKinesis Authentication & authorisation (IAM, Cognito) Testing Running & Debugging functions locally Log aggregation Monitoring & Alerting X-Ray Correlation IDs CI/CD Performance and Cost optimisation Error Handling Configuration management VPC Security Leading practices (API Gateway, Kinesis, Lambda) http://bit.ly/2AA5zzk get 40% off with: ytcui