We're excited to share our October updates!
Kubecon | CloudNativeCon
- We'll be on the floor next week, if you'd like to connect let us know
- We're going for the first time! If you'll be there, let us know, we'd love to meet in person and we're planning an event, stay tuned!
End-to-end testing webinar
- Join us on Thursday, November 16th at 12p ET for our webinar where we'll demonstrate how to run end-to-end tests with Ephemeral Preview Environments
Build time/speed insights
This is a fun one! Speed Insights let you dive into where you spend time during a build in Coherence
- For each service, Insights show the breakdown of time between build/test/deploy
- For the pipeline, Insights show the time spent:
- Waiting for infra changes or in the coherence queue
- Waiting within the cloud systems for workers to process the job
- Running the pipeline steps for build/test/deploy
- If your build stage takes a while, we also added a GPT-powered dockerfile advisor that will suggest improvements that might help speed it up
GPT yaml config
During onboarding, we now offer the option to generate a coherence.yml file for your app, alongside referencing the examples in our docs. Our hope is that this makes it even easier to get your app deployed on Coherence the first time, and lowers the time to seeing it working in your own cloud account
- As with many AI features this is still in beta, not 100% accurate and will continue to improve - but please give it a try for your next project and let us know how it goes!
- Works best for simple vs complex use cases today
Build speed improvements
Continued work on build speed, we've made builds up to 40% faster for AWS apps by fine-tuning the deploy stage - with no changes required for our customers to see the impact on their app
- Excited to keep delivering continued wins in this area and there’s a lot more work on speed in the pipeline for us
GCP kubernetes deployment support
We have shipped the completed v1 support for using Google Kubernetes Engine, where you can use GKE instead of cloud run for your API, web, and frontend servers. This includes a default helm chart that we provide, but you can customize your own helm chart for each service as well, for example to add service mesh sidecars and talk to other services in your cluster not managed by Coherence. Further documentation coming soon.
There’s more to come on customizability and advanced use-case support - please get in touch if you want to give this a try or are interested in the AWS version of this feature down the line.
PR labels for feature auto-create
This is a highly-requested feature where if you enable auto-create of features for PRs in github, you can now limit that to PRs with a label applied.
This means that you can turn on the setting, add a label, e.g. “preview-required”, and then within github add that label to a PR and see the Coherence comment appear with the preview link and deploy status, without needing to go to the Coherence UI, and without needing to get a preview for every feature.
Notable Bug fixes:
AWS Tokyo region support, PR comment race condition with multiple apps in a monorepo, continued rollout of the GCP database over VPC connections, improvements to the GPT-powered error log debugger for failed pipelines, fixes for un-archiving environments, and better github integration by storing the username when you auth to github directly in your profile during signup.