MariaDB is mentioned as supported in requirements, so adding to README too. Also raises a discussion earlier: https://github.com/k3s-io/k3s/issues/10686
Signed-off-by: Robert Silén <robert.silen@iki.fi>
* Fix: Add $SUDO prefix to transactional-update commands in install script
This commit addresses an issue in the k3s install script where the transactional-update commands did not include the $SUDO variable. This omission caused the script to fail if sudo was not used. Adding the $SUDO prefix ensures that the script runs correctly in all environments.
Signed-off-by: ChrisBeaker <christian.siebigteroth@suse.com>
Signed-off-by: Derek Nola <derek.nola@suse.com>
Signed-off-by: Vitor Savian <vitor.savian@suse.com>
Add recursive search and deletion of unmounted/mounted dirs in killall
Signed-off-by: Vitor Savian <vitor.savian@suse.com>
Only clean the server and agent directory if it is uninstall
Signed-off-by: Vitor Savian <vitor.savian@suse.com>
Add uninstall test to check mount points
Signed-off-by: Vitor Savian <vitor.savian@suse.com>
Add uninstall test in CI
Signed-off-by: Vitor Savian <vitor.savian@suse.com>
FindString would silently skip parsing dropins if the main config file
didn't exist. If a custom config file path was passed it would raise an
error, but if we were parsing the default config file and it didn't
exist it would just silently fail to load the dropins.
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
This was only used to pass the bundled strongswan path through to the flannel ipsec backend, and is no longer needed. Ref: #719
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
* Allow pprof to run on server with `--disable-agent`
* Allow supervisor metrics to run on server with `--disable-agent`
Signed-off-by: Derek Nola <derek.nola@suse.com>
Fixes an issue where running etcd-snapshot commands on a node that has a server address set in the config will manage snapshots on that server, instead of on the local node as intended.
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
This should give us more detail on how long dials take before failing, so that we can perhaps better tune the retry loop in the future.
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
I should have caught `[]string{cfg.NodeIP}[0]` and `[]string{envInfo.NodeIP.String()}[0]` in code review...
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
We shouldn't be replacing the configured server address on agents. Doing
so breaks the agent's ability to fall back to the fixed registration
endpoint when all servers are down, since we replaced it with the first
discovered apiserver address. The fixed registration endpoint will be
restored as default when the service is restarted, but this is not the
correct behavior. This should have only been done on etcd-only nodes
that start up using their local supervisor, but need to switch to a
control-plane node as soon as one is available.
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
* For E2E upgrade test, determine the upgrade channel
Signed-off-by: Derek Nola <derek.nola@suse.com>
* Fix typos
Signed-off-by: Derek Nola <derek.nola@suse.com>
---------
Signed-off-by: Derek Nola <derek.nola@suse.com>
* Move snapshot structs and functions into pkg/etcd/snapshot
* Move s3 client code and functions into pkg/etcd/s3
* Refactor pkg/etcd to track snapshot and s3 moves
* Add support for reading s3 client config from secret
* Add minio client cache, since S3 client configuration can now be
changed at runtime by modifying the secret, and don't want to have to
create a new minio client every time we read config.
* Add tests for pkg/etcd/s3
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
Derek Nola