Both these two cephfs tests are using 1-mds-2-client-micro.yaml:
roles:
- [mon.a, mon.b, mon.c, mgr.x, mgr.y, mds.a, mds.b, mds.c, osd.0, osd.1, osd.2, osd.3]
- [client.0]
- [client.1]
So all the mon/mgr/mds/osd daemons are running in the first single
node. It makes no sense to upgrade the same node twice.
Signed-off-by: Xiubo Li <xiubli@redhat.com>
- mds_pre_upgrade:
- print: "**** done mds pre-upgrade sequence"
- install.upgrade:
+ # upgrade the single cluster node, which is running all the mon/mds/osd/mgr daemons
mon.a:
branch: quincy
- mon.b:
- branch: quincy
-- print: "**** done install.upgrade both hosts"
+- print: "**** done install.upgrade the host"
- ceph.restart:
daemons: [mon.*, mgr.*]
mon-health-to-clog: false
- mds_pre_upgrade:
- print: "**** done mds pre-upgrade sequence"
- install.upgrade:
+ # upgrade the single cluster node, which is running all the mon/mds/osd/mgr daemons
mon.a:
branch: quincy
- mon.b:
- branch: quincy
-- print: "**** done install.upgrade both hosts"
+- print: "**** done install.upgrade the host"
- ceph.restart:
daemons: [mon.*, mgr.*]
mon-health-to-clog: false