Pitfalls of Continuous Deployment

on YouTube ↗01h 12m

" summary

Adopts frequent incremental deploys with a CI pipeline based on Jenkins + Vagrant + Fabric + Sentry, enforces feature switches (Gargoyle) and early-adopter flags, and keeps fast rollbacks by symlinking current/previous releases so a revert is literal and quick (roll back in <10s). Tests run on CI (builds ~20 minutes today), flaky Selenium and third-party API tests cause false positives, so strategies include mocking, a local PyPI (G-Shot), distributed test runner (mule/Celery) and metrics via Graphite/statsd with Nagios alerts.

AI-written summary · about

" chapters (11)

1. 0:00Continuous deployment philosophy
2. 2:26Pros and cons
3. 9:27Feature switches and rollouts
4. 14:44Jenkins for integration testing
5. 23:13Fixing false positives
6. 29:28Fabric deployments and rollbacks
7. 33:38Scaling deployment challenges
8. 38:10Database schema changes
9. 42:44Reporting and monitoring
10. 55:51Key takeaways
11. 1:00:55Q&A

" related talks

• 30mImplementing Continuous Deployment with Jenkins & GerrittR. Tyler Croy · Jenkins User Conference San Francisco 2012
• 36mContinuous Deployment with RailsSebastian von Conrad · RubyConf AU 2014
• 39mRemove your Testing Bottleneck with CI/CD PipelinesBen Cooney · Testing Talks Conference 2019 Melbourne
• 01h 01mA team's transition to Continuous DeliveryAshley Noble · NDC Sydney 2017
• 51mBetter software, faster: principles of Continuous Delivery and DevOpsBert Jan Schrijver · Devoxx Belgium 2019
• 24mContinuous Deployment at ScalePremshree Pillai · PHPConf.Asia 2016
• 59mDevOps for the DiscouragedJames Wickett · AppSec California 2015
• 28mClimbing the Mountain of CI/CD: A Tester’s TaleMichele Campbell · UtahJS Conf 2018