Joe Ferris
Joe Ferris
This adds a Dockerfile that can serve as a reference point for Suspenders-like applications which will be deployed to OCI platforms like Docker, Kubernetes, and ECS.
This adds an autoscaling option to the Helm chart using the addon-resizer configuration provided from the Kustomize manifests. Setting addonResizer.enabled to true will add the sidecar. https://github.com/kubernetes-sigs/metrics-server/blob/v0.5.0/manifests/autoscale/patch.yaml
Several users are registering custom webkit drivers: ``` ruby Capybara.register_driver :quiet_webkit do |app| Capybara::Webkit::Driver.new(app, stderr: QuietWebkit.new) end ``` When passing a hash, the globally configured options are not applied, which...
- Remove old, non-standard driver methods - Remove non-global config methods - Remove access to browser instance - Remove support commands for removed methods - Remove support for Qt 4
We can't distinguish between ClickFailed errors which will be retried, and those which have actually failed for good. This means we end up writing hundreds of screenshot files during a...
The Ruby core team has [stopped supporting Ruby 1.9](https://www.ruby-lang.org/en/news/2015/02/23/support-for-ruby-1-9-3-has-ended/), so I think it makes sense for us to do so as well.
When we perform blue/green upgrades to infrastructure, metrics are sent twice: once from the blue cluster, and once from the green cluster. This results in odd artifacts in Grafana as...
https://registry.terraform.io/providers/hashicorp/aws/latest/docs/resources/eks_node_group#capacity_type This will let us set up staging clusters using Spot.
https://github.com/aws/aws-node-termination-handler This will allow workloads to gracefully shutdown when an instance is shut down for any reason.
We deploy Fluent Bit and we're sending logs to Cloudwatch in the AWS platform. Cloudwatch is hard to search and expensive for data retention. It's also only viewable from workload...