opct adm e2e-dedicated taint-node
opct adm e2e-dedicated taint-node manages the setup of the dedicated compute node in the cluster.
The dedicated node is allocated to be used by the services to launch and manage
the test environment (Aggregator server, workflow steps, etc). It is created
to prevent the scheduler from using that node by default, to avoid e2e disruption
workloads impacting the environment (drain the node, evict services, etc),
as well as making sure we have the required resources to host the conformance jobs.
The following configuration is added:
- Create the node label
node-role.kubernetes.io/tests - Set
NoScheduletaints on the node
When allowing the command to select the node automatically, it avoids nodes hosting monitoring workloads, including Prometheus, preventing oversizing and disruptions in the test environment.
Options
Administrative commands.
Usage:
opct adm [flags]
opct adm [command]
Available Commands:
baseline Administrative commands to manipulate baseline results.
cleaner Utility to apply pre-defined patches to existing result archive.
generate Generate administrative commands
parse-etcd-logs Parse ETCD logs.
parse-metrics Process the metrics collected by OPCT and create a HTML report graph.
setup-node Setup the node for the validation process.
Flags:
-h, --help help for adm
Global Flags:
--kubeconfig string kubeconfig for target OpenShift cluster
--log-level string logging level (default "info")
Use "opct adm [command] --help" for more information about a command.
Summary
The opct adm e2e-dedicated taint-node command sets up a node for the validation process by applying the required label and taint configuration to NoSchedule.
Examples
# Select a node randomly and set it up
opct adm e2e-dedicated taint-node
# Specify a node to set up
opct adm e2e-dedicated taint-node --node <node-name>
# Run the command with confirmation
opct adm e2e-dedicated taint-node -y
Flags
-h, --help: Help for setup-node--node string: Node to set required label and taints-y, --yes: Confirm the setup without prompting
Global Flags
--kubeconfig string: Kubeconfig for target OpenShift cluster--log-level string: Logging level (default "info")