--- title: Charmhub | Deploy microceph using Charmhub - The Open Operator Collection description: Deploy the latest version of microceph on any cloud. url: https://charmhub.io/microceph/configurations --- # microceph [OpenStack Charmers](https://charmhub.io/publisher/openstack-charmers "View all packages from OpenStack Charmers") * [OpenStack Charmers](https://charmhub.io/publisher/openstack-charmers "View all packages from OpenStack Charmers") Platform: 26.04 24.04 22.04 squid/stable 331 ``` juju deploy microceph --channel squid/stable ``` [Learn to deploy on juju >](https://juju.is/docs/juju/manage-applications) * [ceph-cluster-network](https://charmhub.io/microceph/configurations#ceph-cluster-network) * [ceph-public-network](https://charmhub.io/microceph/configurations#ceph-public-network) * [db-devices](https://charmhub.io/microceph/configurations#db-devices) * [db-size](https://charmhub.io/microceph/configurations#db-size) * [default-pool-size](https://charmhub.io/microceph/configurations#default-pool-size) * [device-add-flags](https://charmhub.io/microceph/configurations#device-add-flags) * [enable-perf-metrics](https://charmhub.io/microceph/configurations#enable-perf-metrics) * [enable-rgw](https://charmhub.io/microceph/configurations#enable-rgw) * [namespace-projects](https://charmhub.io/microceph/configurations#namespace-projects) * [nfs-use-dedicated-binding](https://charmhub.io/microceph/configurations#nfs-use-dedicated-binding) * [osd-devices](https://charmhub.io/microceph/configurations#osd-devices) * [rbd-stats-pools](https://charmhub.io/microceph/configurations#rbd-stats-pools) * [region](https://charmhub.io/microceph/configurations#region) * [site-name](https://charmhub.io/microceph/configurations#site-name) * [snap-channel](https://charmhub.io/microceph/configurations#snap-channel) * [wait-to-adopt](https://charmhub.io/microceph/configurations#wait-to-adopt) * [wal-devices](https://charmhub.io/microceph/configurations#wal-devices) * [wal-size](https://charmhub.io/microceph/configurations#wal-size) [Learn about configurations >](https://juju.is/docs/juju/configuration#heading--application-configuration) * ceph-cluster-network | string The IP address and netmask of the cluster (back-side) network (e.g., 192.168.0.0/24 or fd00:b::/64). If multiple networks are to be used, a comma-delimited or space-delimited list of subnets in CIDR notation can be provided. Every unit must have an address on one of the configured subnets. Left unset, the cluster network falls back to a single subnet the leader unit resides on within the cluster space. Clearing the option after it has been set reverts the cluster network to the subnet the then-current leader resides on within the cluster space, which can differ from the subnet chosen at bootstrap if leadership has moved. Modifying this configuration option can cause momentary service disruption. * ceph-public-network | string The IP address and netmask of the public (front-side) network (e.g., 192.168.0.0/24 or fd00:a::/64). If multiple networks are to be used, a comma-delimited or space-delimited list of subnets in CIDR notation can be provided. Every unit must have an address on one of the configured subnets. This configuration option only applies at bootstrap and if changed later the unit blocks, as microceph cannot apply a new public network to a running cluster. Set it at deploy time with juju deploy --config ...; to recover a blocked unit, revert the option to the value shown in the unit's status message. Left unset, the public network falls back to a single subnet the leader unit resides on within the public space. * db-devices | string DSL expression to match block devices used as RocksDB carrier devices for config-driven enrollment triggered by osd-devices. This option is ignored unless osd-devices is also set. * db-size | string Requested RocksDB partition size for devices selected by db-devices. This option is ignored unless osd-devices is also set. * default-pool-size | int Default: 3 The default replication factor for pools. Note that changing this value only sets the default value; it doesn't change the replication factor for existing pools. * device-add-flags | string Comma-separated list of flags controlling config-driven device addition with the osd-devices flow. Supported flags: wipe:osd - Wipe non-pristine OSD devices before enrollment encrypt:osd - Encrypt OSD devices wipe:wal - Wipe WAL carrier devices before use encrypt:wal - Encrypt WAL partitions/devices wipe:db - Wipe DB carrier devices before use encrypt:db - Encrypt DB partitions/devices Example: wipe:osd,encrypt:osd,wipe:wal,encrypt:db Note: WAL/DB config and these flags only apply to config-driven provisioning initiated by osd-devices. They are not in effect when using the add-osd action, which currently supports only device-level wipe and encryption parameters. * enable-perf-metrics | boolean Include Ceph performance metrics in the prometheus endpoint. * enable-rgw | string Supported values for this option are "\*" - Enable RGW on all storage nodes "" - Disable RGW on all storage nodes * namespace-projects | boolean Enable project namespacing. If project namespacing is enabled, keystone projects will be implicitly added to a matching project in radosgw, in addition to updating the catalog URL to allow radosgw to support publicly-readable containers and temporary URLS. This namespacing also allows multiple projects to create buckets with the same names, as the bucket names are namespaced into the project namespaces in the RADOS gateway. This configuration option will not be enabled on a charm upgrade, and cannot be toggled on in an existing installation as it will remove tenant access to existing buckets. * nfs-use-dedicated-binding | boolean Controls which network the NFS (Ganesha) service binds to when serving shares (e.g. to Manila clients). When false (the default), NFS binds to the Ceph public network. When true, NFS binds to the network space the "nfs" extra-binding is bound to (set with "juju bind microceph nfs="); left unbound, that binding resolves to the model's default space. Warning: enabling this option without binding the "nfs" extra-binding to a space resolves to the model's default space, which in a typical deployment is the management network -- not the public network. This can expose NFS on an unintended, more restricted network. Always run "juju bind microceph nfs=" before enabling this option, and verify the published nfs-address before pointing clients (e.g. Manila) at it. Note: this option is applied when a host is added to an NFS cluster, not retroactively. Changing it does not rebind NFS services that are already running -- those keep the address they were created with. To rebind an existing service, remove and re-add the ceph-nfs relation, which recreates the cluster on the current address. * osd-devices | string DSL expression to match block devices for config-driven OSD enrollment. The DSL uses functional syntax with predicates and variables. Supported predicates: and(), or(), not(), eq(), ne(), gt(), ge(), lt(), le(), in(), re() Supported variables: @type, @vendor, @model, @size, @devnode, @host Examples: eq(@type,'nvme') and(eq(@type,'ssd'), ge(@size,100GiB)) or(eq(@type,'ssd'), eq(@type,'nvme')) When set, the charm evaluates this expression against available block devices on each unit and enrolls matching devices as OSDs. This is additive to action-based and Juju storage-based OSDs. * rbd-stats-pools | string Enable RBD IO statistics for specified rbd pools. * region | string Default: RegionOne Name of the OpenStack region * site-name | string The local site name. This should be unique amongst the connected remote microceph clusters. * snap-channel | string Default: squid/stable The snap-channel option determines the MicroCeph snap that is being installed via charm. Change this option to perform an upgrade of MicroCeph. See https://charmhub.io/microceph for details on the upgrade process. * wait-to-adopt | boolean Whether charm microceph should wait for an external ceph cluster to yield admin credentials using which microceph will adopt the external cluster rather than bootstrapping a new ceph cluster on microceph bootstrap. * wal-devices | string DSL expression to match block devices used as WAL carrier devices for config-driven enrollment triggered by osd-devices. This option is ignored unless osd-devices is also set. * wal-size | string Requested WAL partition size for devices selected by wal-devices. This option is ignored unless osd-devices is also set.