airship/treasuremap
Code Issues Proposed changes

Updates for Rook 1.6.2 and Ceph 15.2.11

Browse Source 
This PS is to update the Rook yaml files for version v1.6.2. Additionally, the version of Ceph is upgraded to v15.2.11 and Ceph-CSI is upgraded to v3.3.1.

v1.6 provides a few features the storage team wants:

* The operator supports upgrading multiple OSDs in parallel
* LVM no longer used to provision OSDs by default
* Monitor failover can be disabled if needed
* Operator support for Ceph Pacific (v16)
* Ceph 15.2.11 by default
* CephClient CRD standardized to controller-runtime library (kubebuilder)

https://github.com/kubernetes-sigs/controller-runtime

* Pod Disruption Budgets enabled by default.

https://github.com/rook/rook/blob/master/design/ceph/ceph-managed-disruptionbudgets.md

More notes:

* There are many indentation changes in common.yaml
* There is now a variable in operator.yaml for enabling host networking for the CSI pods. Default is to use host network.

* CSI image updates:

ROOK_CSI_CEPH_IMAGE: "quay.io/cephcsi/cephcsi:v3.3.1"
ROOK_CSI_SNAPSHOTTER_IMAGE: "k8s.gcr.io/sig-storage/csi-snapshotter:v4.0.0"

* There is a very large update to crds.yaml largely due to the controller-runtime being employed.

* Ceph 15.2.11 needed for CVE-2021-20288

Change-Id: I5cf0cf63bfcf4b0ea1d242d6eae2f53adda7be5e
This commit is contained in:
Frank Ritchie 2021-05-11 13:56:37 -04:00 committed by Alexey
parent 2946a13806
commit e7130f4301
12 changed files with 8179 additions and 1824 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
manifests/function/rook-cluster/cephfs/base/filesystem.yaml
View File

@ -27,7 +27,6 @@ spec:
preserveFilesystemOnDelete: true
# The metadata service (mds) configuration
metadataServer:
# The affinity rules to apply to the mds deployment
placement:
# nodeAffinity:
@ -72,4 +71,3 @@ spec:
# A key/value list of labels
labels:
# key: value

2
manifests/function/rook-cluster/pools/base/pool.yaml
View File

@ -58,7 +58,7 @@ spec:
# quota in bytes and/or objects, default value is 0 (unlimited)
# see https://docs.ceph.com/en/latest/rados/operations/pools/#set-pool-quotas
# quotas:
# maxSize: "10Gi" # valid suffixes include K, M, G, T, P, Ki, Mi, Gi, Ti, Pi
# maxSize: "10Gi" # valid suffixes include k, M, G, T, P, E, Ki, Mi, Gi, Ti, Pi, Ei
# maxObjects: 1000000000 # 1 billion objects
# A key/value list of annotations
annotations:

2
manifests/function/rook-cluster/pools/data/data-pool.yaml
View File

@ -8,5 +8,5 @@ spec:
replicated:
size: 2
quotas:
maxSize: "10Gi" # valid suffixes include K, M, G, T, P, Ki, Mi, Gi, Ti, Pi
maxSize: "10Gi" # valid suffixes include k, M, G, T, P, E, Ki, Mi, Gi, Ti, Pi, Ei
maxObjects: 1000000000 # 1 billion objects

3
manifests/function/rook-cluster/pools/rbd/rbd-pool.yaml
View File

@ -8,6 +8,5 @@ spec:
replicated:
size: 3
quotas:
maxSize: "0" # valid suffixes include K, M, G, T, P, Ki, Mi, Gi, Ti, Pi, eg: "10Gi"
# "0" means no quotas. Since rook 1.5.9 you must use string as a value's type
maxSize: "0" # e.g. "10Gi" - valid suffixes include k, M, G, T, P, E, Ki, Mi, Gi, Ti, Pi, Ei
maxObjects: 0 # 1000000000 = billion objects, 0 means no quotas

124
manifests/function/rook-operator/upstream/common.yaml
View File

@ -9,7 +9,6 @@
#
# Most of the sections are prefixed with a 'OLM' keyword which is used to build our CSV for an OLM (Operator Life Cycle manager)
###################################################################################################################
kind: ClusterRoleBinding
apiVersion: rbac.authorization.k8s.io/v1
metadata:
@ -63,7 +62,7 @@ metadata:
operator: rook
storage-backend: ceph
rules:
- apiGroups:
- apiGroups:
- ""
- apps
- extensions
@ -94,7 +93,7 @@ metadata:
operator: rook
storage-backend: ceph
rules:
- apiGroups:
- apiGroups:
- ""
resources:
- pods
@ -108,7 +107,7 @@ rules:
- create
- update
- delete
- apiGroups:
- apiGroups:
- apps
- extensions
resources:
@ -122,6 +121,12 @@ rules:
- create
- update
- delete
- apiGroups:
- batch
resources:
- cronjobs
verbs:
- delete
---
# The cluster role for managing the Rook CRDs
apiVersion: rbac.authorization.k8s.io/v1
@ -132,7 +137,7 @@ metadata:
operator: rook
storage-backend: ceph
rules:
- apiGroups:
- apiGroups:
- ""
resources:
# Pod access is needed for fencing
@ -145,7 +150,7 @@ rules:
- get
- list
- watch
- apiGroups:
- apiGroups:
- ""
resources:
- events
@ -161,7 +166,7 @@ rules:
- create
- update
- delete
- apiGroups:
- apiGroups:
- storage.k8s.io
resources:
- storageclasses
@ -169,10 +174,11 @@ rules:
- get
- list
- watch
- apiGroups:
- apiGroups:
- batch
resources:
- jobs
- cronjobs
verbs:
- get
- list
@ -180,19 +186,19 @@ rules:
- create
- update
- delete
- apiGroups:
- apiGroups:
- ceph.rook.io
resources:
- "*"
verbs:
- "*"
- apiGroups:
- apiGroups:
- rook.io
resources:
- "*"
verbs:
- "*"
- apiGroups:
- apiGroups:
- policy
- apps
- extensions
@ -204,7 +210,7 @@ rules:
- replicasets
verbs:
- "*"
- apiGroups:
- apiGroups:
- healthchecking.openshift.io
resources:
- machinedisruptionbudgets
@ -215,7 +221,7 @@ rules:
- create
- update
- delete
- apiGroups:
- apiGroups:
- machine.openshift.io
resources:
- machines
@ -226,7 +232,7 @@ rules:
- create
- update
- delete
- apiGroups:
- apiGroups:
- storage.k8s.io
resources:
- csidrivers
@ -235,7 +241,7 @@ rules:
- delete
- get
- update
- apiGroups:
- apiGroups:
- k8s.cni.cncf.io
resources:
- network-attachment-definitions
@ -251,7 +257,7 @@ metadata:
operator: rook
storage-backend: ceph
rules:
- apiGroups:
- apiGroups:
- ""
resources:
- configmaps
@ -261,7 +267,7 @@ rules:
- get
- list
- watch
- apiGroups:
- apiGroups:
- ""
resources:
- events
@ -280,14 +286,14 @@ metadata:
operator: rook
storage-backend: ceph
rules:
- apiGroups:
- apiGroups:
- ""
verbs:
- "*"
resources:
- secrets
- configmaps
- apiGroups:
- apiGroups:
- storage.k8s.io
resources:
- storageclasses
@ -295,7 +301,7 @@ rules:
- get
- list
- watch
- apiGroups:
- apiGroups:
- "objectbucket.io"
verbs:
- "*"
@ -333,7 +339,7 @@ roleRef:
kind: Role
name: rook-ceph-system
subjects:
- kind: ServiceAccount
- kind: ServiceAccount
name: rook-ceph-system
namespace: rook-ceph # namespace:operator
---
@ -350,7 +356,7 @@ roleRef:
kind: ClusterRole
name: rook-ceph-global
subjects:
- kind: ServiceAccount
- kind: ServiceAccount
name: rook-ceph-system
namespace: rook-ceph # namespace:operator
# OLM: END OPERATOR ROLEBINDING
@ -399,19 +405,19 @@ metadata:
name: rook-ceph-osd
namespace: rook-ceph # namespace:cluster
rules:
- apiGroups: [""]
- apiGroups: [""]
resources: ["configmaps"]
verbs: [ "get", "list", "watch", "create", "update", "delete" ]
- apiGroups: ["ceph.rook.io"]
verbs: ["get", "list", "watch", "create", "update", "delete"]
- apiGroups: ["ceph.rook.io"]
resources: ["cephclusters", "cephclusters/finalizers"]
verbs: [ "get", "list", "create", "update", "delete" ]
verbs: ["get", "list", "create", "update", "delete"]
---
kind: ClusterRole
apiVersion: rbac.authorization.k8s.io/v1
metadata:
name: rook-ceph-osd
rules:
- apiGroups:
- apiGroups:
- ""
resources:
- nodes
@ -425,7 +431,7 @@ apiVersion: rbac.authorization.k8s.io/v1
metadata:
name: rook-ceph-mgr-system
rules:
- apiGroups:
- apiGroups:
- ""
resources:
- configmaps
@ -441,7 +447,7 @@ metadata:
name: rook-ceph-mgr
namespace: rook-ceph # namespace:cluster
rules:
- apiGroups:
- apiGroups:
- ""
resources:
- pods
@ -451,8 +457,10 @@ rules:
- get
- list
- watch
- create
- update
- delete
- apiGroups:
- apiGroups:
- batch
resources:
- jobs
@ -463,7 +471,7 @@ rules:
- create
- update
- delete
- apiGroups:
- apiGroups:
- ceph.rook.io
resources:
- "*"
@ -478,7 +486,7 @@ metadata:
name: rook-ceph-cmd-reporter
namespace: rook-ceph # namespace:cluster
rules:
- apiGroups:
- apiGroups:
- ""
resources:
- pods
@ -504,7 +512,7 @@ roleRef:
kind: ClusterRole
name: rook-ceph-cluster-mgmt
subjects:
- kind: ServiceAccount
- kind: ServiceAccount
name: rook-ceph-system
namespace: rook-ceph # namespace:operator
---
@ -519,7 +527,7 @@ roleRef:
kind: Role
name: rook-ceph-osd
subjects:
- kind: ServiceAccount
- kind: ServiceAccount
name: rook-ceph-osd
namespace: rook-ceph # namespace:cluster
---
@ -534,7 +542,7 @@ roleRef:
kind: Role
name: rook-ceph-mgr
subjects:
- kind: ServiceAccount
- kind: ServiceAccount
name: rook-ceph-mgr
namespace: rook-ceph # namespace:cluster
---
@ -549,7 +557,7 @@ roleRef:
kind: ClusterRole
name: rook-ceph-mgr-system
subjects:
- kind: ServiceAccount
- kind: ServiceAccount
name: rook-ceph-mgr
namespace: rook-ceph # namespace:cluster
---
@ -563,7 +571,7 @@ roleRef:
kind: ClusterRole
name: rook-ceph-mgr-cluster
subjects:
- kind: ServiceAccount
- kind: ServiceAccount
name: rook-ceph-mgr
namespace: rook-ceph # namespace:cluster
@ -578,7 +586,7 @@ roleRef:
kind: ClusterRole
name: rook-ceph-osd
subjects:
- kind: ServiceAccount
- kind: ServiceAccount
name: rook-ceph-osd
namespace: rook-ceph # namespace:cluster
@ -595,7 +603,7 @@ roleRef:
kind: Role
name: rook-ceph-cmd-reporter
subjects:
- kind: ServiceAccount
- kind: ServiceAccount
name: rook-ceph-cmd-reporter
namespace: rook-ceph # namespace:cluster
# OLM: END CMD REPORTER ROLEBINDING
@ -613,8 +621,8 @@ metadata:
# need to be renamed with a value that will match before others.
name: 00-rook-privileged
annotations:
seccomp.security.alpha.kubernetes.io/allowedProfileNames: 'runtime/default'
seccomp.security.alpha.kubernetes.io/defaultProfileName: 'runtime/default'
seccomp.security.alpha.kubernetes.io/allowedProfileNames: "runtime/default"
seccomp.security.alpha.kubernetes.io/defaultProfileName: "runtime/default"
spec:
privileged: true
allowedCapabilities:
@ -682,7 +690,7 @@ spec:
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
name: 'psp:rook'
name: "psp:rook"
rules:
- apiGroups:
- policy
@ -700,7 +708,7 @@ metadata:
roleRef:
apiGroup: rbac.authorization.k8s.io
kind: ClusterRole
name: 'psp:rook'
name: "psp:rook"
subjects:
- kind: ServiceAccount
name: rook-ceph-system
@ -716,7 +724,7 @@ roleRef:
kind: ClusterRole
name: psp:rook
subjects:
- kind: ServiceAccount
- kind: ServiceAccount
name: default
namespace: rook-ceph # namespace:cluster
---
@ -730,7 +738,7 @@ roleRef:
kind: ClusterRole
name: psp:rook
subjects:
- kind: ServiceAccount
- kind: ServiceAccount
name: rook-ceph-osd
namespace: rook-ceph # namespace:cluster
---
@ -744,7 +752,7 @@ roleRef:
kind: ClusterRole
name: psp:rook
subjects:
- kind: ServiceAccount
- kind: ServiceAccount
name: rook-ceph-mgr
namespace: rook-ceph # namespace:cluster
---
@ -758,7 +766,7 @@ roleRef:
kind: ClusterRole
name: psp:rook
subjects:
- kind: ServiceAccount
- kind: ServiceAccount
name: rook-ceph-cmd-reporter
namespace: rook-ceph # namespace:cluster
# OLM: END CLUSTER POD SECURITY POLICY BINDINGS
@ -893,7 +901,7 @@ metadata:
roleRef:
apiGroup: rbac.authorization.k8s.io
kind: ClusterRole
name: 'psp:rook'
name: "psp:rook"
subjects:
- kind: ServiceAccount
name: rook-csi-cephfs-plugin-sa
@ -906,7 +914,7 @@ metadata:
roleRef:
apiGroup: rbac.authorization.k8s.io
kind: ClusterRole
name: 'psp:rook'
name: "psp:rook"
subjects:
- kind: ServiceAccount
name: rook-csi-cephfs-provisioner-sa
@ -1065,7 +1073,19 @@ rules:
verbs: ["update", "patch"]
- apiGroups: [""]
resources: ["configmaps"]
verbs: [ "get"]
verbs: ["get"]
- apiGroups: ["replication.storage.openshift.io"]
resources: ["volumereplications", "volumereplicationclasses"]
verbs: ["create", "delete", "get", "list", "patch", "update", "watch"]
- apiGroups: ["replication.storage.openshift.io"]
resources: ["volumereplications/finalizers"]
verbs: ["update"]
- apiGroups: ["replication.storage.openshift.io"]
resources: ["volumereplications/status"]
verbs: ["get", "patch", "update"]
- apiGroups: ["replication.storage.openshift.io"]
resources: ["volumereplicationclasses/status"]
verbs: ["get"]
# OLM: END CSI RBD CLUSTER ROLE
# OLM: BEGIN CSI RBD CLUSTER ROLEBINDING
---
@ -1076,7 +1096,7 @@ metadata:
roleRef:
apiGroup: rbac.authorization.k8s.io
kind: ClusterRole
name: 'psp:rook'
name: "psp:rook"
subjects:
- kind: ServiceAccount
name: rook-csi-rbd-plugin-sa
@ -1089,7 +1109,7 @@ metadata:
roleRef:
apiGroup: rbac.authorization.k8s.io
kind: ClusterRole
name: 'psp:rook'
name: "psp:rook"
subjects:
- kind: ServiceAccount
name: rook-csi-rbd-provisioner-sa

8883
manifests/function/rook-operator/upstream/crds.yaml
View File

File diff suppressed because it is too large Load Diff

37
manifests/function/rook-operator/upstream/operator.yaml
View File

@ -29,6 +29,11 @@ data:
ROOK_CSI_ENABLE_RBD: "true"
ROOK_CSI_ENABLE_GRPC_METRICS: "false"
# Set to true to enable host networking for CSI CephFS and RBD nodeplugins. This may be necessary
# in some network configurations where the SDN does not provide access to an external cluster or
# there is significant drop in read/write performance.
# CSI_ENABLE_HOST_NETWORK: "true"
# Set logging level for csi containers.
# Supported values from 0 to 5. 0 for general useful logs, 5 for trace level verbosity.
# CSI_LOG_LEVEL: "0"
@ -64,11 +69,11 @@ data:
# The default version of CSI supported by Rook will be started. To change the version
# of the CSI driver to something other than what is officially supported, change
# these images to the desired release of the CSI driver.
ROOK_CSI_CEPH_IMAGE: "quay.io/cephcsi/cephcsi:v3.2.1"
ROOK_CSI_CEPH_IMAGE: "quay.io/cephcsi/cephcsi:v3.3.1"
ROOK_CSI_REGISTRAR_IMAGE: "k8s.gcr.io/sig-storage/csi-node-driver-registrar:v2.0.1"
ROOK_CSI_RESIZER_IMAGE: "k8s.gcr.io/sig-storage/csi-resizer:v1.0.1"
ROOK_CSI_PROVISIONER_IMAGE: "k8s.gcr.io/sig-storage/csi-provisioner:v2.0.4"
ROOK_CSI_SNAPSHOTTER_IMAGE: "k8s.gcr.io/sig-storage/csi-snapshotter:v3.0.2"
ROOK_CSI_SNAPSHOTTER_IMAGE: "k8s.gcr.io/sig-storage/csi-snapshotter:v4.0.0"
ROOK_CSI_ATTACHER_IMAGE: "k8s.gcr.io/sig-storage/csi-attacher:v3.0.2"
# (Optional) set user created priorityclassName for csi plugin pods.
@ -274,6 +279,16 @@ data:
# Whether the OBC provisioner should watch on the operator namespace or not, if not the namespace of the cluster will be used
ROOK_OBC_WATCH_OPERATOR_NAMESPACE: "true"
# Whether to enable the flex driver. By default it is enabled and is fully supported, but will be deprecated in some future release
# in favor of the CSI driver.
ROOK_ENABLE_FLEX_DRIVER: "false"
# Whether to start the discovery daemon to watch for raw storage devices on nodes in the cluster.
# This daemon does not need to run if you are only going to create your OSDs based on StorageClassDeviceSets with PVCs.
ROOK_ENABLE_DISCOVERY_DAEMON: "false"
# Enable volume replication controller
CSI_ENABLE_VOLUME_REPLICATION: "false"
# CSI_VOLUME_REPLICATION_IMAGE: "quay.io/csiaddons/volumereplication-operator:v0.1.0"
# (Optional) Admission controller NodeAffinity.
# ADMISSION_CONTROLLER_NODE_AFFINITY: "role=storage-node; storage=rook, ceph"
# (Optional) Admission controller tolerations list. Put here list of taints you want to tolerate in YAML format.
@ -308,7 +323,7 @@ spec:
serviceAccountName: rook-ceph-system
containers:
- name: rook-ceph-operator
image: rook/ceph:v1.5.9
image: rook/ceph:v1.6.2
args: ["ceph", "operator"]
volumeMounts:
- mountPath: /var/lib/rook
@ -386,12 +401,6 @@ spec:
# (Optional) Discover Agent Pod Labels.
# - name: DISCOVER_AGENT_POD_LABELS
# value: "key1=value1,key2=value2"
# Allow rook to create multiple file systems. Note: This is considered
# an experimental feature in Ceph as described at
# http://docs.ceph.com/docs/master/cephfs/experimental-features/#multiple-filesystems-within-a-ceph-cluster
# which might cause mons to crash as seen in https://github.com/rook/rook/issues/1027
- name: ROOK_ALLOW_MULTIPLE_FILESYSTEMS
value: "false"
# The logging level for the operator: INFO | DEBUG
- name: ROOK_LOG_LEVEL
@ -430,16 +439,6 @@ spec:
- name: DISCOVER_DAEMON_UDEV_BLACKLIST
value: "(?i)dm-[0-9]+,(?i)rbd[0-9]+,(?i)nbd[0-9]+"
# Whether to enable the flex driver. By default it is enabled and is fully supported, but will be deprecated in some future release
# in favor of the CSI driver.
- name: ROOK_ENABLE_FLEX_DRIVER
value: "false"
# Whether to start the discovery daemon to watch for raw storage devices on nodes in the cluster.
# This daemon does not need to run if you are only going to create your OSDs based on StorageClassDeviceSets with PVCs.
- name: ROOK_ENABLE_DISCOVERY_DAEMON
value: "false"
# Time to wait until the node controller will move Rook pods to other
# nodes after detecting an unreachable node.
# Pods affected by this setting are:

12
manifests/function/treasuremap-base-catalogues/versions-treasuremap.yaml
View File

@ -108,12 +108,12 @@ spec:
rook-operator:
rook-ceph-operator:
rook-ceph-operator:
image: rook/ceph:v1.5.9
image: rook/ceph:v1.6.2
rook-ceph-operator-config:
ceph_daemon:
image: ceph/ceph:v15.2.10
image: ceph/ceph:v15.2.11
rook_csi_ceph_image:
image: quay.io/cephcsi/cephcsi:v3.2.1
image: quay.io/cephcsi/cephcsi:v3.3.1
rook_csi_registrar_image:
image: k8s.gcr.io/sig-storage/csi-node-driver-registrar:v2.0.1
rook_csi_resizer_image:
@ -121,15 +121,15 @@ spec:
rook_csi_provisioner_image:
image: k8s.gcr.io/sig-storage/csi-provisioner:v2.0.4
rook_csi_snapshotter_image:
image: k8s.gcr.io/sig-storage/csi-snapshotter:v3.0.2
image: k8s.gcr.io/sig-storage/csi-snapshotter:v4.0.0
rook_csi_attacher_image:
image: k8s.gcr.io/sig-storage/csi-attacher:v3.0.2
storage-rook:
ceph:
ceph-version:
image: ceph/ceph:v15.2.10
image: ceph/ceph:v15.2.11
rook-ceph-tools:
image: rook/ceph:v1.5.9
image: rook/ceph:v1.6.2
image_components:
# image_components are organized by

3
manifests/type/airship-core/target/workload/storage/rook/ceph-conf.yaml
View File

@ -16,6 +16,7 @@ data:
mon_warn_on_pool_no_redundancy = true
# # You can add other default configuration sections
# # to create fully customized ceph.conf
# [mon]
[mon]
auth_allow_insecure_global_id_reclaim = false
# [osd]
# [rgw]

21
manifests/type/airship-core/target/workload/storage/rook/cephcluster.yaml
View File

@ -6,12 +6,16 @@ metadata:
spec:
dataDirHostPath: /var/lib/rook
cephVersion:
#see: https://tracker.ceph.com/issues/48797
image: ceph/ceph:v15.2.10
image: ceph/ceph:v15.2.11
#allowUnsupported: true
mon:
count: 3
allowMultiplePerNode: false
mgr:
count: 1
modules:
- name: pg_autoscaler
enabled: true
dashboard:
enabled: true
# If you are going to use the dashboard together with ingress-controller,
@ -57,4 +61,17 @@ spec:
# deviceFilter: "^/dev/sd[c-h]"
# Also you can configure each device and/or each node. Please refer to the official rook
# documentation for the branch 1.5.x
# The section for configuring management of daemon disruptions during upgrade or fencing.
disruptionManagement:
# If true, the operator will create and manage PodDisruptionBudgets for OSD, Mon, RGW, and MDS daemons. OSD PDBs are managed dynamically
# via the strategy outlined in the [design](https://github.com/rook/rook/blob/master/design/ceph/ceph-managed-disruptionbudgets.md). The operator will
# block eviction of OSDs by default and unblock them safely when drains are detected.
managePodBudgets: true
# A duration in minutes that determines how long an entire failureDomain like `region/zone/host` will be held in `noout` (in addition to the
# default DOWN/OUT interval) when it is draining. This is only relevant when `managePodBudgets` is `true`. The default value is `30` minutes.
osdMaintenanceTimeout: 30
# A duration in minutes that the operator will wait for the placement groups to become healthy (active+clean) after a drain was completed and OSDs came back up.
# Operator will continue with the next drain if the timeout exceeds. It only works if `managePodBudgets` is `true`.
# No values or 0 means that the operator will wait until the placement groups are healthy before unblocking the next drain.
pgHealthCheckTimeout: 0
---

2
manifests/type/airship-core/target/workload/storage/rook/rook-ceph-tools.yaml
View File

@ -19,7 +19,7 @@ spec:
dnsPolicy: ClusterFirstWithHostNet
containers:
- name: rook-ceph-tools
image: rook/ceph:v1.5.9
image: rook/ceph:v1.6.2
command: ["/tini"]
args: ["-g", "--", "/usr/local/bin/toolbox.sh"]
imagePullPolicy: IfNotPresent