upgrade ceph nodes from nautilus to octopus

As a prerequisite to upgrade to proxmox from 6.0 to 7.0 (prerequisite to upgrade the proxmox nodes to bullseye), the 3 ceph servers need to be upgraded to the octopus

Migration guide: https://pve.proxmox.com/wiki/Ceph_Nautilus_to_Octopus

The 3 nodes to upgrade:

• beaubourg
• branly
• hypervisor3

---

Migrated from T3787 (view on Phabricator)

mentioned in issue #3771 (closed)

marked this issue as related to #3771 (closed)

added Component upgrades priority:Normal state:wip labels

We (all sysadm \o/) followed the guide one step at a time. Cluster is in HEALTH_OK [1]

• [1]

root@branly:~# ceph status
  cluster:
    id:     82307cfd-401a-4b58-8d29-7cbcd98f26fd
    health: HEALTH_OK

  services:
    mon: 3 daemons, quorum beaubourg,hypervisor3,branly (age 18m)
    mgr: branly(active, since 37m), standbys: beaubourg, hypervisor3
    mds: proxmox-cephfs:1 {0=branly=up:active} 2 up:standby
    osd: 14 osds: 14 up (since 18m), 14 in (since 4w)

  data:
    pools:   4 pools, 593 pgs
    objects: 734.71k objects, 2.7 TiB
    usage:   8.0 TiB used, 8.4 TiB / 16 TiB avail
    pgs:     593 active+clean

  io:
    client:   5.1 MiB/s rd, 77 MiB/s wr, 220 op/s rd, 709 op/s wr

assigned to @ardumont

removed state:wip label

closed

I had forgotten to set a maintenance tag about this in grafana. Fixed now.

The ceph packages need to be also updated on the proxmox nodes even if they are not in the ceph cluster (from the output of pve6to7)

# CHECKING HYPER-CONVERGED CEPH STATUS

INFO: hyper-converged ceph setup detected!
INFO: getting Ceph status/health information..
PASS: Ceph health reported as 'HEALTH_OK'.
INFO: getting Ceph daemon versions..
PASS: single running version detected for daemon type monitor.
PASS: single running version detected for daemon type manager.
PASS: single running version detected for daemon type MDS.
PASS: single running version detected for daemon type OSD.
PASS: single running overall version detected for all Ceph daemon types.
WARN: 'noout' flag not set - recommended to prevent rebalancing during cluster-wide upgrades.
INFO: checking Ceph config..
FAIL: local Ceph version too low, at least Octopus required..

After the packages upgrade:

# CHECKING HYPER-CONVERGED CEPH STATUS

INFO: hyper-converged ceph setup detected!
INFO: getting Ceph status/health information..
PASS: Ceph health reported as 'HEALTH_OK'.
INFO: getting Ceph daemon versions..
PASS: single running version detected for daemon type monitor.
PASS: single running version detected for daemon type manager.
PASS: single running version detected for daemon type MDS.
PASS: single running version detected for daemon type OSD.
PASS: single running overall version detected for all Ceph daemon types.
WARN: 'noout' flag not set - recommended to prevent rebalancing during cluster-wide upgrades.
INFO: checking Ceph config..

same ^ for pompidou