[DRBD-announce] LINSTOR Operator 1.7.0-rc.3 released

[Moritz Wanzenböck]

Dear LINSTOR on Kubernetes Users,

We just released a new release candidate for the upcoming Operator release:
1.7.0-rc.3

There are quite a few new features, which we want to ensure are ready for
release, hence the extended release candidate phase. We did not announce
the previous RCs, as they were mostly used for internal testing.

The most exciting feature is certainly the option to run LINSTOR without an
additional database for the Controller. LINSTOR 1.16.0 added experimental
support for using the Kubernetes API directly to store its internal state.
The current plan is to support both Etcd and Kubernetes API as datastore,
with the eventual goal of removing Etcd support once we are happy with the
stability of this new backend. To use it, simply disable Etcd deployment
(--set etcd.enabled=false) and enable the "k8s" datastore (--set
operator.controller.dbConnectionURL=k8s). You can read a bit more about the
details in the upstream documentation. [1]

We also improved support for the LINSTOR backup feature: in order to
securely store any access tokens for remote locations, LINSTOR needs to be
configured with a master passphrase. If no passphrase is defined, the Helm
chart will create one for you. Currently, the backup feature is only
available using the LINSTOR CLI, take a look at the linstor remote ... and
linstor backup ... commands. There are some improvements to the CSI driver
in the pipeline which provide a more natural integration with Kubernetes.

Lastly, if a node is shut down and removed from the cluster, the Operator
will now trigger eviction of that node, i.e. it will try to recreate the
lost resources on a new node.

The chart is available from charts.linstor.io. For pre-releases helm
requires setting the exact version:

  helm install linstor linstor/linstor --version 1.7.0-rc.3 ...

Please help by testing this release. Please also be aware of the known
issues listed below. Source is available, as always: [2], [3]

Best regards,
Moritz

Known Issues
------------

There is one known issue with the current release candidate, related to the
automatic master passphrase: If a LINSTOR controller restarts and tries
connecting to a satellite, that satellite might refuse the connection, and
you will get a "node not authorized" error in the error reports. The main
symptoms of this is that any commands depending on a satellite hang
indefinitely. The workaround is to restart the satellites (kubectl rollout
restart daemonset/linstor-ns-node).

Changes since 1.6.0
-------------------

Added

- Support LINSTORs `k8s` backend by adding the necessary RBAC resources.
- Automatically create a LINSTOR passphrase when none is configured.
- Automatic eviction and deletion of offline satellites if the Kubernetes
node object was also deleted.

Changed
- DRBD version 9.1.4
- LINSTOR 1.16.0
- LINSTOR CSI 0.16.1
- Updated CSI sidecars

[1]:
https://github.com/piraeusdatastore/piraeus-operator/blob/master/doc/k8s-backend.md
[2]: https://github.com/piraeusdatastore/piraeus-operator
[3]: https://github.com/LINBIT/linstor-operator-builder
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linbit.com/pipermail/drbd-announce/attachments/20211117/e2e11063/attachment.htm>