cephadm: add some coarse-grained locking

Take per-cluster 'deploy' lock for

 bootstrap
 deploy
 adopt
 rm-daemon
 rm-cluster

No locking for

 version
 pull
 run (this is never actually used anyway)
 shell
 enter
 ceph-volume
 unit (this is a clean pass-through to systemctl)
 logs
 ls
 check-host

The only one I'm not sure about is ceph-volume.  I don't want the c-v
list/inventory commands to block, but perhaps the deploy ones should...
but perhaps not, since there's no reason we can't deploy multiple OSDs
on different devices at the same time.

Signed-off-by: Sage Weil <sage@redhat.com>

diff --git a/src/cephadm/cephadm b/src/cephadm/cephadm
index b8961e543bcf10a9064a421d407fbf133c878041..c8fbfd2a5d24b7437ad420885cc0d2874e47e678 100755 (executable)
--- a/src/cephadm/cephadm
+++ b/src/cephadm/cephadm
@@ -1381,6 +1381,9 @@ def command_bootstrap():
     mgr_id = args.mgr_id or generate_service_id()
     logging.info('Cluster fsid: %s' % fsid)
 
+    l = FileLock(fsid)
+    l.acquire()
+
     # config
     cp = read_config(args.config)
     if args.mon_ip:
@@ -1692,6 +1695,9 @@ def command_deploy():
     # type: () -> None
     (daemon_type, daemon_id) = args.name.split('.', 1)
 
+    l = FileLock(args.fsid)
+    l.acquire()
+
     supported_daemons = list(Ceph.daemons)
     supported_daemons.extend(Monitoring.components)
 
@@ -2008,6 +2014,7 @@ def list_daemons(detail=True, legacy_dir=None):
 
 def command_adopt():
     # type: () -> None
+
     (daemon_type, daemon_id) = args.name.split('.', 1)
     (uid, gid) = extract_uid_gid()
     if args.style == 'legacy':
@@ -2018,6 +2025,9 @@ def command_adopt():
         if not fsid:
             raise Error('could not detect legacy fsid; set fsid in ceph.conf')
 
+        l = FileLock(fsid)
+        l.acquire()
+
         data_dir_src = ('/var/lib/ceph/%s/%s-%s' %
                         (daemon_type, args.cluster, daemon_id))
         data_dir_src = os.path.abspath(args.legacy_dir + data_dir_src)
@@ -2087,6 +2097,10 @@ def command_adopt():
 
 def command_rm_daemon():
     # type: () -> None
+
+    l = FileLock(args.fsid)
+    l.acquire()
+
     (daemon_type, daemon_id) = args.name.split('.', 1)
     if daemon_type in ['mon', 'osd'] and not args.force:
         raise Error('must pass --force to proceed: '
@@ -2109,6 +2123,9 @@ def command_rm_cluster():
         raise Error('must pass --force to proceed: '
                       'this command may destroy precious data!')
 
+    l = FileLock(args.fsid)
+    l.acquire()
+
     # stop + disable individual daemon units
     for d in list_daemons(detail=False):
         if d['fsid'] != args.fsid: