canonical / juju-lint

Run checks against a juju model
GNU General Public License v3.0
0 stars 4 forks source link

ceph-mon must not be deployed on top of an osd node #124

Closed zxhdaze closed 7 months ago

zxhdaze commented 7 months ago

mons and osds should live on separate nodes. Alert if the mon lxd is deployed on top of a baremetal storage node.


Imported from Launchpad using lp2gh.

zxhdaze commented 7 months ago

(by elmo) To clarify - it's OK for ceph-mon's to cohabitate with ceph-osd in a hyperconverged environment. Perhaps the alert should be if a ceph-mon lxd is the only container on a ceph-osd host?

zxhdaze commented 7 months ago

(by phausman) Andrea, could you please clarify why "mons and osds should live on separate nodes"?

For a disaggregated architecture, with distinct Storage and Control nodes, I understand that ceph-mons should live on Control nodes, while ceph-osds are placed -- as usual -- on Storage nodes. Do I get it right?