feat: Added option to customize kubelet arguments

[lukapetrovic-git]

Description

Added option to customize kubelet arguments (https://docs.rke2.io/reference/server_config#agentflags).

I used the variable rke2_server_options somewhat successfully too, but it becomes very messy very fast since it requires specifying all kubelet arguments on a single line like so:

rke2_server_options:
  - "kubelet-arg: ['--system-reserved=cpu=100m,memory=100Mi,pid=100000','--kube-reserved=cpu=80m,memory=1400Mi,pid=100000','--eviction-hard=memory.available<100Mi,nodefs.available<10%,nodefs.inodesFree<5%,imagefs.available<15%','--enforce-node-allocatable=pods', '--pod-max-pids=32768']"

I would constantly break it trying to add new/adjust existing arguments so i created a new variable so that arguments can be passed like below:

rke_kubelet_arg:
  - "--system-reserved=cpu=100m,memory=100Mi,pid=100000"
  - "--kube-reserved=cpu=80m,memory=1400Mi,pid=100000"
  - "--eviction-hard=memory.available<100Mi,nodefs.available<10%,nodefs.inodesFree<5%,imagefs.available<15%"
  - "--enforce-node-allocatable=pods"
  - "--pod-max-pids=32768"

Type of change

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update
• Small minor change not affecting the Ansible Role code (Github Actions Workflow, Documentation, etc.)

How Has This Been Tested?

I have been successfully using this variable on all my environments both dev/test and production in the last month or so.

[MonolithProjects]

LGTM. Thanks

[lukapetrovic-git]

@MonolithProjects I just noticed i named the variable rke_kubelet_arg instead of rke2_kubelet_arg and pushed another commit just to stay consistent across all vars.

[MonolithProjects]

LGTM

[MonolithProjects]

@MonolithProjects I just noticed i named the variable rke_kubelet_arg instead of rke2_kubelet_arg and pushed another commit just to stay consistent across all vars.

Thanks. I didn't notice that too.

[albundy83]

Hello,
just to harmonize things, won't it be possible to use the same format than kube_controller_manager and kube_scheduler_arg:

# rke2_kube_controller_manager_arg:
#   - "bind-address=0.0.0.0"

# rke2_kube_scheduler_arg:
#   - "bind-address=0.0.0.0"

Like this (I have removed -- from parameter)

# (Optional) Customize default kubelet arguments
# rke2_kubelet_arg:
#   - "system-reserved=cpu=100m,memory=100Mi"

instead of:

# (Optional) Customize default kubelet arguments
# rke2_kubelet_arg:
#   - "--system-reserved=cpu=100m,memory=100Mi"