- K3s has some absolute paths set in various places which cannot be
changed. Using a symbolic link was the easiest fix. This is nice for
running K3S on SBC that must boot to SD but K3S data should be stored
on a faster drive.
- Other changes are for making the site playbook replayable without
resetting the cluster. Ideally you can rerun it to check existing
nodes or to add new ones.
Signed-off-by: Derek Nola <derek.nola@suse.com>
- Reboot playbook updated for HA testing
- Small adjustment made after testing HA with playbook
- Fix reset playbook failing in some cases
Signed-off-by: Derek Nola <derek.nola@suse.com>
- Inital server starts first so it can grab the token for other servers and copy the config to localhost
- More cleanup of agent/server naming scheme
Signed-off-by: Derek Nola <derek.nola@suse.com>
- HA option for multiple server nodes using embedded etcd
- Switch to yaml inventory file for easier editing and combining vars
- Update to full ansible module names
- Change master/node names to server/agent
- Cleanup small linting errors
- Add reboot playbook which staggers reboot to keep HA cluster up
- Move playbooks to playbook directory
Signed-off-by: Derek Nola <derek.nola@suse.com>
When k3s is installed on an OS with default high ulimits, performance
issues can be observed. This was discovered on CoreOS where the default
value is 1073741816. Symptoms include very slow file operations such
as installing a Rook/Ceph cluster will take ~6 hours instead of ~10 minutes.
A google search for 'container LimitNOFILE' will show that most major
projects set this already, including the (unused) containerd systemd unit
found in this repository at /vendor/github.com/containerd/containerd/containerd.service
k3OS is not affected becuasse the default there is already 1048576.
See description in coreos/fedora-coreos-tracker#329
Sometimes https://127.0.0.1:6443 can be written in the conf, the regexp does
not account for that.
Signed-off-by: Julien DOCHE <julien.doche@gmail.com>