You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Expanding Helm charts bundle to support and simplify deployment of Selenium Grid server to Kubernetes cluster with the following design.
Individual chart
Each component (including the standalone image) will have an individual chart to deploy independently.
Component Individual chart will be imported as sub-chart in Grid umbrella chart
The chart selenium-grid will be an umbrella chart with chart dependencies, including Hub (or isolated components) and Nodes. This chart creates the resources needed to deploy Selenium Grid to the Kubernetes cluster only.
A new umbrella chart, selenium-grid-solution with chart dependencies, including selenium-grid, and other tools chart (e.g. KEDA, ingress-nginx, prometheus-stack, etc.). This chart deploys to cluster the Selenium Grid server with other solutions such as autoscaling, observability, monitoring, etc.
Chart selenium-grid-solution also provides specific configuration files based on different environments or deployment scenarios.
Usage example
Organizing Helm charts to manage and deploy across environments without duplicating charts.
The text was updated successfully, but these errors were encountered:
I see your proposal plan to publish node chart separately. Does it mean it may open doors to have flexibility to deploy different versions of browser nodes at same time in same namespace (two separate deployments of node chart in same namespace).
I am just checking if one can deploy chrome version 128 and 129 pods at the same time in same grid.
If someone takes the pain to build images themselves using backward compatibility feature then it may be rule out mismatch of selenium server on hub/router and nodes.
I am also not sure how autoscaling logic will react to it. It may be very complex or may be very simple but triggering few thoughts here.
Feature and motivation
Upgrade chart major release from v0.x.x to v1.x.x
Expanding Helm charts bundle to support and simplify deployment of Selenium Grid server to Kubernetes cluster with the following design.
selenium-grid
will be an umbrella chart with chart dependencies, including Hub (or isolated components) and Nodes. This chart creates the resources needed to deploy Selenium Grid to the Kubernetes cluster only.selenium-grid-solution
with chart dependencies, includingselenium-grid
, and other tools chart (e.g. KEDA, ingress-nginx, prometheus-stack, etc.). This chart deploys to cluster the Selenium Grid server with other solutions such as autoscaling, observability, monitoring, etc.selenium-grid-solution
also provides specific configuration files based on different environments or deployment scenarios.Usage example
Organizing Helm charts to manage and deploy across environments without duplicating charts.
The text was updated successfully, but these errors were encountered: