ceph-disk: protect deactivate with activate lock

When ceph-disk prepares the disk, it triggers udev events and each of
them ceph-disk activate. If systemctl stop ceph-osd@2 happens while
there still are ceph-disk activate in flight, the systemctl stop may be
cancelled by the systemctl enable issued by one of the pending ceph-disk
activate.

This only matters in a test environment where disks are destroyed
shortly after they are activated.

Signed-off-by: Loic Dachary <loic@dachary.org>
(cherry picked from commit 6395bf856b4d4511f0758174ef915ebcafbe3777)

Conflicts:

        src/ceph-disk: ceph-disk deactivate does not exist in ceph-disk
            on infernalis. But the same feature is implemented in
            ceph-test-disk.py for test purposes and has the same
            problem. The patch is adapted to ceph-test-disk.py.

diff --git a/qa/workunits/ceph-disk/ceph-disk-test.py b/qa/workunits/ceph-disk/ceph-disk-test.py
index 522f727610e95e12d251adcb215518e2b2c6bf44..a2bbd018d30af7f22b8632edbe79bee5a388765f 100644 (file)
--- a/qa/workunits/ceph-disk/ceph-disk-test.py
+++ b/qa/workunits/ceph-disk/ceph-disk-test.py
@@ -30,6 +30,7 @@
 #  sudo /usr/sbin/ceph-disk -v trigger /dev/vdb1 # activates if vdb1 is data
 #
 import argparse
+import fcntl
 import json
 import logging
 import configobj
@@ -44,6 +45,21 @@ import uuid
 
 LOG = logging.getLogger('CephDisk')
 
+class filelock(object):
+    def __init__(self, fn):
+        self.fn = fn
+        self.fd = None
+
+    def acquire(self):
+        assert not self.fd
+        self.fd = file(self.fn, 'w')
+        fcntl.lockf(self.fd, fcntl.LOCK_EX)
+
+    def release(self):
+        assert self.fd
+        fcntl.lockf(self.fd, fcntl.LOCK_UN)
+        self.fd = None
+
 class CephDisk:
 
     def __init__(self):
@@ -127,6 +143,9 @@ class CephDisk:
         raise Exception("journal for uuid = " + uuid + " not found in " + str(disks))
 
     def destroy_osd(self, uuid):
+        STATEDIR = '/var/lib/ceph'
+        activate_lock = filelock(STATEDIR + '/tmp/ceph-disk.activate.lock')
+        activate_lock.acquire()
         id = self.sh("ceph osd create " + uuid).strip()
         self.helper("control_osd stop " + id + " || true")
         self.wait_for_osd_down(uuid)
@@ -155,6 +174,7 @@ class CephDisk:
         ceph auth del osd.{id}
         ceph osd crush rm osd.{id}
         """.format(id=id))
+        activate_lock.release()
 
     @staticmethod
     def osd_up_predicate(osds, uuid):