ceph/ceph-ansible

Error in ceph-ansible Cluster after OS Upgrade from Bionic to Focal

rjhaikal opened this issue · 1 comments

rjhaikal commented

Bug Report

What happened:

I have a Ceph cluster version Octopus to be used with OpenStack version Ussuri, and the current Operating System is Ubuntu Bionic. I have a need to upgrade OpenStack from Ussuri to Victoria, but there are required version differences as shown in the image below.

image

I tried it in my lab when I had a Ceph cluster and upgraded the OS, for example, from Bionic to Focal. After the OS upgrade was completed, the Ceph cluster was destroyed and not running.

What are the correct steps if I have an existing Ceph Cluster and want to perform an OS Upgrade on that Ceph Cluster?

What you expected to happen:

Successfully upgrading the OS on the Ceph Cluster from Ubuntu Bionic to Focal without destroying the Ceph Cluster itself.

How to reproduce it (minimal and precise):

Install ceph-ansible version Octopus on Ubuntu Bionic, then upgrade to Ubuntu Focal.

Share your group_vars files, inventory and full ceph-ansibe log

group_vars/all.yml

dummy:
ceph_origin: repository
ceph_repository: community
ceph_stable_release: octopus

#Interface Public Network Ceph
monitor_interface: ens7

osd_objectstore: bluestore
public_network: 10.24.14.0/24
cluster_network: 10.24.15.0/24

dashboard_enabled: True
dashboard_admin_user: admin
dashboard_admin_password: $password
prometheus_container_image: "docker.io/prom/prometheus:v2.23.0"
ntp_daemon_type: timesyncd
grafana_plugins: {}            # temporary
grafana_admin_user: admin      # temporary
grafana_admin_password: admin  # temporary

group_vars/osds.yml

dummy:
devices:
  - /dev/sda
  - /dev/sdb
  - /dev/sdc
  - /dev/sdd
  - /dev/sde

osd_auto_discovery: false

hosts

[mons]
controller0[1:3]

[osds]
controller0[1:3]
compute0[1:3]

[mgrs]
controller0[1:3]

[grafana-server]
controller0[1:3]

Environment:

  • OS (e.g. from /etc/os-release): Ubuntu 18.04 Bionic
  • Kernel (e.g. uname -a): 4.15.0-210-generic
  • Docker version if applicable (e.g. docker version): 20.10.21
  • Ansible version (e.g. ansible-playbook --version): 2.9.27
  • ceph-ansible version (e.g. git head or tag or stable branch): stable-5.0
  • Ceph version (e.g. ceph -v): ceph version 15.2.17 (8a82819d84cf884bd39c17e3236e0632ac146dc4) octopus
rjhaikal commented

this case solve by update ceph repo to focal, reinstall ceph packages and reboot nodes.

thanks.