forked from rook/rook
Blaine Gardner
086fa8231c
Configure the Ceph rgw daemon completely from the operator a la the recent changes to the Ceph mon, mgr, and mds operators. Create the rgw deployment or daemonset first, and then create the keyring secret for the object store with its owner reference as the corresponding deployment or daemonset. When the replication controller is deleted, the secret is also deleted. The RGW's mime.types file is now stored in a configmap with a different file created for each object store. This is primarily just a means to get the mime.types file into the rgw pod, but the added benefit is that the administrator can modify the configmap, which could reduce susceptibility to file type execution vulnerabilities (worst case). Signed-off-by: Blaine Gardner <blaine.gardner@suse.com>
1.7 KiB
1.7 KiB
Major Themes
Action Required
Notable Features
Minio Object Stores
- Now have an additional label named `objectstore` with the name of the Object Store, to allow better selection for Services.
- Use `Readiness` and `Liveness` probes.
- Updated automatically on Object Store CRD changes.
Ceph
- A
CephNFSCRD will start NFS daemon(s) for exporting CephFS volumes or RGW buckets. See the NFS documentation. - Selinux labeling for mounts can now be toggled with the ROOK_ENABLE_SELINUX_RELABELING environment variable.
- Recursive chown for mounts can now be toggled with the ROOK_ENABLE_FSGROUP environment variable.
- Added the dashboard
portconfiguration setting. - Added the dashboard
sslconfiguration setting. - Added Ceph CSI driver deployments on Kubernetes 1.13 and above.
Breaking Changes
- Rook no longer supports Kubernetes
1.8and1.9. - Rook no longer supports running more than one monitor on the same node when
hostNetworkandallowMultiplePerNodearetrue. - Rook Operator switches from Extensions v1beta1 to use Apps v1 API for DaemonSet and Deployment.
Ceph
- Rook will no longer create a directory-based osd in the
dataDirHostPathif no directories or devices are specified or if there are no disks on the host. - Containers in
mon,mgr,mds, andrgwpods have been removed and/or changed names. - Config paths in
mon,mgr,mdsandrgwcontainers are now always under/etc/cephor/var/lib/cephand as close to Ceph's default path as possible regardless of thedataDirHostPathsetting.