Install MicroCeph with Juju¶
This guide shows how to perform a general install of MicroCeph using Juju.
Prerequisites¶
A snapd-compatible host to run the Juju client
A machine cloud (substrate), e.g. MAAS, LXD or AWS at your disposal. See the Juju documentation for the full list of supported clouds and more information on how to set them up.
Important: MicroCeph requires a machine cloud and will not work with a Kubernetes cluster.
Disks on each node to add as object storage daemons (OSDs) to the Ceph cluster
Deploy MicroCeph¶
Deploy three units of the MicroCeph charm to three machines.
juju deploy -n 3 microceph --channel latest/edge --to 0,1,2
The output to the juju status command should look similar to this:
Model Controller Cloud/Region Version SLA Timestamp
microceph sunbeam-default sunbeam/default 3.2-beta3 unsupported 03:40:02Z
App Version Status Scale Charm Channel Rev Exposed Message
microceph active 3 microceph latest/edge 3 no
Unit Workload Agent Machine Public address Ports Message
microceph/0* active idle 0 10.5.0.106
microceph/1 active idle 1 10.5.1.191
microceph/2 active idle 2 10.5.1.81
Machine State Address Inst id Base AZ Message
0 started 10.5.0.106 manual:10.5.0.106 ubuntu@22.04 Manually provisioned machine
1 started 10.5.1.191 manual:10.5.1.191 ubuntu@22.04 Manually provisioned machine
2 started 10.5.1.81 manual:10.5.1.81 ubuntu@22.04 Manually provisioned machine
Verify your deployment¶
Verify the state of the Ceph cluster by running the ceph status command on one of the nodes:
juju ssh microceph/leader sudo microceph.ceph status
Sample output:
cluster:
id: edd914f5-fdf8-4b56-bdd7-95d6c5e10d81
health: HEALTH_WARN
OSD count 0 < osd_pool_default_size 3
services:
mon: 3 daemons, quorum microceph2,microceph3,microceph4 (age 57s)
mgr: microceph2(active, since 74s), standbys: microceph3, microceph4
osd: 0 osds: 0 up, 0 in
data:
pools: 0 pools, 0 pgs
objects: 0 objects, 0 B
usage: 0 B used, 0 B / 0 B avail
pgs:
The next step is to add disks to the microceph nodes.