These examples show how to perform advanced configuration tasks on your Rook storage cluster.
Most of the examples make use of the
ceph client command. A quick way to use the Ceph client suite is from a Rook Toolbox container.
The Kubernetes based examples assume Rook OSD pods are in the
rook-ceph namespace. If you run them in a different namespace, modify
kubectl -n rook-ceph [...] to fit your situation.
Using alternate namespaces¶
If you wish to deploy the Rook Operator and/or Ceph clusters to namespaces other than the default
rook-ceph, the manifests are commented to allow for easy
sed replacements. Change
ROOK_CLUSTER_NAMESPACE to tailor the manifests for additional Ceph clusters. You can choose to also change
ROOK_OPERATOR_NAMESPACE to create a new Rook Operator for each Ceph cluster (don't forget to set
ROOK_CURRENT_NAMESPACE_ONLY), or you can leave it at the same value for every Ceph cluster if you only wish to have one Operator manage all Ceph clusters.
This will help you manage namespaces more easily, but you should still make sure the resources are configured to your liking.
Deploying a second cluster¶
If you wish to create a new CephCluster in a different namespace than
rook-ceph while using a single operator to manage both clusters execute the following:
This will create all the necessary RBACs as well as the new namespace. The script assumes that
common.yaml was already created. When you create the second CephCluster CR, use the same
NAMESPACE and the operator will configure the second cluster.
All Rook logs can be collected in a Kubernetes environment with the following command:
This gets the logs for every container in every Rook pod and then compresses them into a
.gz archive for easy sharing. Note that instead of
gzip, you could instead pipe to
less or to a single text file.
Keeping track of OSDs and their underlying storage devices can be difficult. The following scripts will clear things up quickly.
The output should look something like this.
Separate Storage Groups¶
It is deprecated to manually need to set this, the
deviceClass property can be used on Pool structures in
CephObjectStore CRD objects.
By default Kook Storage Distribution/Ceph puts all storage under one replication rule in the CRUSH Map which provides the maximum amount of storage capacity for a cluster. If you would like to use different storage endpoints for different purposes, you'll have to create separate storage groups.
In the following example we will separate SSD drives from spindle-based drives, a common practice for those looking to target certain workloads onto faster (database) or slower (file archive) storage.
Placement Group Sizing¶
Since Ceph Nautilus (v14.x), you can use the Ceph MGR
pg_autoscaler module to auto scale the PGs as needed. If you want to enable this feature, please refer to Default PG and PGP counts.
The general rules for deciding how many PGs your pool(s) should contain is:
- Fewer than 5 OSDs set
- Between 5 and 10 OSDs set
- Between 10 and 50 OSDs set
If you have more than 50 OSDs, you need to understand the tradeoffs and how to calculate the pg_num value by yourself. For calculating pg_num yourself please make use of the pgcalc tool.
Setting PG Count¶
Be sure to read the placement group sizing section before changing the number of PGs.
The advised method for controlling Ceph configuration is to manually use the Ceph CLI or the Ceph dashboard because this offers the most flexibility. It is highly recommended that this only be used when absolutely necessary and that the
config be reset to an empty string if/when the configurations are no longer necessary. Configurations in the config file will make the Ceph cluster less configurable from the CLI and dashboard and may make future tuning or debugging difficult.
Setting configs via Ceph's CLI requires that at least one mon be available for the configs to be set, and setting configs via dashboard requires at least one mgr to be available. Ceph also has a number of very advanced settings that cannot be modified easily via the CLI or dashboard. In order to set configurations before monitors are available or to set advanced configuration settings, the
rook-config-override ConfigMap exists, and the
config field can be set with the contents of a
ceph.conf file. The contents will be propagated to all mon, mgr, OSD, MDS, and RGW daemons as an
Rook performs no validation on the config, so the validity of the settings is the user's responsibility.
rook-config-override ConfigMap is created before the cluster is started, the Ceph daemons will automatically pick up the settings. If you add the settings to the ConfigMap after the cluster has been initialized, each daemon will need to be restarted where you want the settings applied:
- mons: ensure all three mons are online and healthy before restarting each mon pod, one at a time.
- mgrs: the pods are stateless and can be restarted as needed, but note that this will disrupt the Ceph dashboard during restart.
- OSDs: restart your the pods by deleting them, one at a time, and running
ceph -sbetween each restart to ensure the cluster goes back to "active/clean" state.
- RGW: the pods are stateless and can be restarted as needed.
- MDS: the pods are stateless and can be restarted as needed.
After the pod restart, the new settings should be in effect. Note that if the ConfigMap in the Ceph cluster's namespace is created before the cluster is created, the daemons will pick up the settings at first launch.
To automate the restart of the Ceph daemon pods, you will need to trigger an update to the pod specs. The simplest way to trigger the update is to add annotations or labels to the CephCluster CR for the daemons you want to restart. The operator will then proceed with a rolling update, similar to any other update to the cluster.
In this example we will set the default pool
size to two, and tell OSD daemons not to change the weight of OSDs on startup.
Modify Ceph settings carefully. You are leaving the sandbox tested by Rook. Changing the settings could result in unhealthy daemons or even data loss if used incorrectly.
When the Rook Operator creates a cluster, a placeholder ConfigMap is created that will allow you to override Ceph configuration settings. When the daemon pods are started, the settings specified in this ConfigMap will be merged with the default settings generated by Rook.
The default override settings are blank. Cutting out the extraneous properties, we would see the following defaults after creating a cluster:
To apply your desired configuration, you will need to update this ConfigMap. The next time the daemon pod(s) start, they will use the updated configs.
Modify the settings and save. Each line you add should be indented from the
config property as such:
It is highly recommended to use the default setting that comes with CephCSI and this can only be used when absolutely necessary. The
ceph.conf should be reset back to default values if/when the configurations are no longer necessary.
csi-ceph-conf-override ConfigMap is created before the cluster is started, the CephCSI pods will automatically pick up the settings. If you add the settings to the ConfigMap after the cluster has been initialized, you can restart the Rook operator pod and wait for Rook to recreate CSI pods to take immediate effect.
After the CSI pods are restarted, the new settings should be in effect.
In this Example we will set the
false to skip rbd pool validation.
Modify Ceph settings carefully to avoid modifying the default configuration. Changing the settings could result in unexpected results if used incorrectly.
Restart the Rook operator pod and wait for CSI pods to be recreated.
OSD CRUSH Settings¶
A useful view of the CRUSH Map is generated with the following command:
In this section we will be tweaking some of the values seen in the output.
The CRUSH weight controls the ratio of data that should be distributed to each OSD. This also means a higher or lower amount of disk I/O operations for an OSD with higher/lower weight, respectively.
By default OSDs get a weight relative to their storage capacity, which maximizes overall cluster capacity by filling all drives at the same rate, even if drive sizes vary. This should work for most use-cases, but the following situations could warrant weight changes:
- Your cluster has some relatively slow OSDs or nodes. Lowering their weight can reduce the impact of this bottleneck.
- You're using bluestore drives provisioned with Rook v0.3.1 or older. In this case you may notice OSD weights did not get set relative to their storage capacity. Changing the weight can fix this and maximize cluster capacity.
This example sets the weight of osd.0 which is 600GiB
OSD Primary Affinity¶
When pools are set with a size setting greater than one, data is replicated between nodes and OSDs. For every chunk of data a Primary OSD is selected to be used for reading that data to be sent to clients. You can control how likely it is for an OSD to become a Primary using the Primary Affinity setting. This is similar to the OSD weight setting, except it only affects reads on the storage device, not capacity or writes.
In this example we will ensure that
osd.0 is only selected as Primary if all other OSDs holding data replicas are unavailable:
OSD Dedicated Network¶
It is possible to configure ceph to leverage a dedicated network for the OSDs to communicate across. A useful overview is the CEPH Networks section of the Ceph documentation. If you declare a cluster network, OSDs will route heartbeat, object replication and recovery traffic over the cluster network. This may improve performance compared to using a single network, especially when slower network technologies are used, with the tradeoff of additional expense and subtle failure modes.
Two changes are necessary to the configuration to enable this capability:
Use hostNetwork in the cluster configuration¶
hostNetwork setting in the Ceph Cluster CRD configuration. For example,
Changing this setting is not supported in a running Rook cluster. Host networking should be configured when the cluster is first created.
Define the subnets to use for public and private OSD networks¶
rook-config-override configmap to define the custom network configuration:
In the editor, add a custom configuration to instruct ceph which subnet is the public network and which subnet is the private network. For example:
After applying the updated rook-config-override configmap, it will be necessary to restart the OSDs by deleting the OSD pods in order to apply the change. Restart the OSD pods by deleting them, one at a time, and running ceph -s between each restart to ensure the cluster goes back to "active/clean" state.
Phantom OSD Removal¶
If you have OSDs in which are not showing any disks, you can remove those "Phantom OSDs" by following the instructions below. To check for "Phantom OSDs", you can run (example output):
node2.example.com in the output has no disks, so it is most likely a "Phantom OSD".
Now to remove it, use the ID in the first column of the output and replace
<ID> with it. In the example output above the ID would be
-7. The commands are:
To recheck that the Phantom OSD was removed, re-run the following command and check if the OSD with the ID doesn't show up anymore:
Auto Expansion of OSDs¶
Prerequisites for Auto Expansion of OSDs¶
1) A PVC-based cluster deployed in dynamic provisioning environment with a
2) Create the Rook Toolbox.
Prometheus Operator and [Prometheus ../Monitoring/ceph-monitoring.mdnitoring.md#prometheus-instances) are Prerequisites that are created by the auto-grow-storage script.
To scale OSDs Vertically¶
Run the following script to auto-grow the size of OSDs on a PVC-based Rook cluster whenever the OSDs have reached the storage near-full threshold.
growth-rate percentage represents the percent increase you want in the OSD capacity and maxSize represent the maximum disk size.
For example, if you need to increase the size of OSD by 30% and max disk size is 1Ti
To scale OSDs Horizontally¶
Run the following script to auto-grow the number of OSDs on a PVC-based Rook cluster whenever the OSDs have reached the storage near-full threshold.
Count of OSD represents the number of OSDs you need to add and maxCount represents the number of disks a storage cluster will support.
For example, if you need to increase the number of OSDs by 3 and maxCount is 10