The state of your Kubernetes cluster is kept in the etcd datastore. This page shows how to backup and restore the etcd included in Charmed Kubernetes.
Backing up application specific data, normally stored in a persistent volume, is not currently supported by native Kubernetes. Various third party solutions are available - please refer to their own documentation for details.
Creating an etcd snapshot
etcd is a distributed key/value store. To create a snapshot, all that is required is to run the
snapshot action on one of the units running etcd:
juju run etcd/0 snapshot keys-version=v3
The console will wait to return the result of running the action, which in this case includes the path and filename of the generated snapshot file. For example:
unit-etcd-0: id: 3d6a505e-07d7-4697-8471-60156f87b1b4 results: copy: cmd: juju scp etcd/0:/home/ubuntu/etcd-snapshots/etcd-snapshot-2018-09-26-18.04.02.tar.gz . snapshot: path: /home/ubuntu/etcd-snapshots/etcd-snapshot-2018-09-26-18.04.02.tar.gz sha256: e85ae4d49b6a889de2063031379ab320cc8f09b6e328cdff2fb9179fc641eee9 size: 68K version: |- etcdctl version: 3.2.10 API version: 2 status: completed timing: completed: 2018-09-26 18:04:04 +0000 UTC enqueued: 2018-09-26 18:04:04 +0000 UTC started: 2018-09-26 18:04:03 +0000 UTC unit: etcd/0
This path/filename relates to the unit where the action was run. As we will likely want to use the snapshot on a different unit, we should fetch the snapshot to the local machine. The command to perform this is also helpfully supplied in the
copy section of the output (see above):
juju scp etcd/0:/home/ubuntu/etcd-snapshots/etcd-snapshot-2018-09-26-18.04.02.tar.gz .
It is also wise to check the sha256 checksum of the file you have copied against the value in the previous output:
Restoring a snapshot
As restoring only works when there is a single unit of etcd, it is usual to deploy a new instance of the application first.
juju deploy etcd new-etcd --series=focal --config channel=3.4/stable
--series option is included here to illustrate how to specify which series the new unit should be running on.
--config option is required to specify the same channel of etcd as the original unit.
Next we upload and identify the snapshot file to this new unit:
juju attach-resource new-etcd snapshot=./etcd-snapshot-2018-09-26-18.04.02.tar.gz
Then run the restore action:
juju run new-etcd/0 restore
Once the restore action has finished, you should see output confirming that the operation is
completed. The new etcd application will need to be connected to the rest of the deployment:
juju integrate new-etcd [calico|flannel|$cni] juju integrate new-etcd kubernetes-control-plane
To restore the cluster capabilities of etcd, you can now add more units:
juju add-unit new-etcd -n 2
Once the deployment has settled and all
new-etcd units report ready, verify the cluster health with:
juju run new-etcd/0 health
which should return something similar to:
unit-new-etcd-0: id: 27fe2081-6513-4968-869d-6c2c092210a1 results: result-map: message: |- member 3c149609bfcf7692 is healthy: got healthy result from https://172.31.18.7:2379 cluster is healthy status: completed timing: completed: 2018-10-26 15:16:33 +0000 UTC enqueued: 2018-10-26 15:16:32 +0000 UTC started: 2018-10-26 15:16:33 +0000 UTC unit: new-etcd/0