->=13.2.0
-========
-
-(some or all of the notes in this section seem to have been omitted
-from the 13.2.0 release notes (?), so include them in 13.2.1 release
-notes (?))
-
-* The Telegraf module for the Manager allows for sending statistics to
- an Telegraf Agent over TCP, UDP or a UNIX Socket. Telegraf can then
- send the statistics to databases like InfluxDB, ElasticSearch, Graphite
- and many more.
-
-* The graylog fields naming the originator of a log event have
- changed: the string-form name is now included (e.g., ``"name":
- "mgr.foo"``), and the rank-form name is now in a nested section
- (e.g., ``"rank": {"type": "mgr", "num": 43243}``).
-
-* If the cluster log is directed at syslog, the entries are now
- prefixed by both the string-form name and the rank-form name (e.g.,
- ``mgr.x mgr.12345 ...`` instead of just ``mgr.12345 ...``).
-
-* The JSON output of the ``osd find`` command has replaced the ``ip``
- field with an ``addrs`` section to reflect that OSDs may bind to
- multiple addresses.
-
-13.2.1
-======
-
-* CephFS clients without the 's' flag in their authentication capability
- string will no longer be able to create/delete snapshots. To allow
- ``client.foo`` to create/delete snapshots in the ``bar`` directory of
- filesystem ``cephfs_a``, use command:
-
- - ``ceph auth caps client.foo mon 'allow r' osd 'allow rw tag cephfs data=cephfs_a' mds 'allow rw, allow rws path=/bar'``
-
-13.2.3
-======
-
-* The default memory utilization for the mons has been increased
- somewhat. Rocksdb now uses 512 MB of RAM by default, which should
- be sufficient for small to medium-sized clusters; large clusters
- should tune this up. Also, the ``mon_osd_cache_size`` has been
- increase from 10 OSDMaps to 500, which will translate to an
- additional 500 MB to 1 GB of RAM for large clusters, and much less
- for small clusters.
-
-* Ceph v13.2.2 includes a wrong backport, which may cause mds to go into
- 'damaged' state when upgrading Ceph cluster from previous version.
- The bug is fixed in v13.2.3. For ceph v13.2.2 installation, upgrading
- to v13.2.3 does not requires special action.
-
-* The bluestore_cache_* options are no longer needed. They are replaced
- by osd_memory_target, defaulting to 4GB. BlueStore will expand
- and contract its cache to attempt to stay within this
- limit. Users upgrading should note this is a higher default
- than the previous bluestore_cache_size of 1GB, so OSDs using
- BlueStore will use more memory by default.
- For more details, see `BlueStore docs <http://docs.ceph.com/docs/mimic/rados/configuration/bluestore-config-ref/#automatic-cache-sizing>_`
-
-13.2.5
-======
-
-* This release fixes the pg log hard limit bug that was introduced in
- 13.2.2, https://tracker.ceph.com/issues/36686. A flag called
- `pglog_hardlimit` has been introduced, which is off by default. Enabling
- this flag will limit the length of the pg log. In order to enable
- that, the flag must be set by running `ceph osd set pglog_hardlimit`
- after completely upgrading to 13.2.2. Once the cluster has this flag
- set, the length of the pg log will be capped by a hard limit. Once set,
- this flag *must not* be unset anymore. In luminous, this feature was
- introduced in 12.2.11. Users who are running 12.2.11, and want to
- continue to use this feauture, should upgrade to 13.2.5 or later.
-
-* This release also fixes a cve on civetweb, CVE-2019-3821 where ssl fds were
- not closed in civetweb in case the initial negotiation fails.
-
-13.2.6
+13.2.7
======
-* Ceph v13.2.6 now packages python bindings for python3.6 instead of
- python3.4, because EPEL7 recently switched from python3.4 to
- python3.6 as the native python3. see the `announcement <https://lists.fedoraproject.org/archives/list/epel-announce@lists.fedoraproject.org/message/EGUMKAIMPK2UD5VSHXM53BH2MBDGDWMO/>_`
- for more details on the background of this change.
\ No newline at end of file
+* The `cache drop` admin socket command has been removed. The `ceph tell mds.X
+ cache drop` remains.
ratio = raw_avail / fs_avail
assert 0.9 < ratio < 1.1
- def _run_drop_cache_cmd(self, timeout, use_tell):
+ def _run_drop_cache_cmd(self, timeout):
drop_res = None
- if use_tell:
- mds_id = self.fs.get_lone_mds_id()
- drop_res = json.loads(
- self.fs.mon_manager.raw_cluster_cmd("tell", "mds.{0}".format(mds_id),
- "cache", "drop", str(timeout)))
- else:
- drop_res = self.fs.mds_asok(["cache", "drop", str(timeout)])
+ mds_id = self.fs.get_lone_mds_id()
+ drop_res = json.loads(
+ self.fs.mon_manager.raw_cluster_cmd("tell", "mds.{0}".format(mds_id),
+ "cache", "drop", str(timeout)))
return drop_res
- def _drop_cache_command(self, timeout, use_tell=True):
+ def _drop_cache_command(self, timeout):
self.mount_b.umount_wait()
ls_data = self.fs.mds_asok(['session', 'ls'])
self.assert_session_count(1, ls_data)
# create some files
self.mount_a.create_n_files("dc-dir/dc-file", 1000)
# drop cache
- drop_res = self._run_drop_cache_cmd(timeout, use_tell)
+ drop_res = self._run_drop_cache_cmd(timeout)
self.assertTrue(drop_res['client_recall']['return_code'] == 0)
self.assertTrue(drop_res['flush_journal']['return_code'] == 0)
- def _drop_cache_command_timeout(self, timeout, use_tell=True):
+ def _drop_cache_command_timeout(self, timeout):
self.mount_b.umount_wait()
ls_data = self.fs.mds_asok(['session', 'ls'])
self.assert_session_count(1, ls_data)
# simulate client death and try drop cache
self.mount_a.kill()
- drop_res = self._run_drop_cache_cmd(timeout, use_tell)
+ drop_res = self._run_drop_cache_cmd(timeout)
self.assertTrue(drop_res['client_recall']['return_code'] == -errno.ETIMEDOUT)
self.assertTrue(drop_res['flush_journal']['return_code'] == 0)
self.mount_a.mount()
self.mount_a.wait_until_mounted()
- def test_drop_cache_command_asok(self):
- """
- Basic test for checking drop cache command using admin socket.
- Note that the cache size post trimming is not checked here.
- """
- self._drop_cache_command(10, use_tell=False)
-
- def test_drop_cache_command_tell(self):
+ def test_drop_cache_command(self):
"""
Basic test for checking drop cache command using tell interface.
Note that the cache size post trimming is not checked here.
"""
self._drop_cache_command(10)
- def test_drop_cache_command_timeout_asok(self):
- """
- Check drop cache command with non-responding client using admin
- socket. Note that the cache size post trimming is not checked here.
- """
- self._drop_cache_command_timeout(5, use_tell=False)
-
- def test_drop_cache_command_timeout_tell(self):
+ def test_drop_cache_command_timeout(self):
"""
Check drop cache command with non-responding client using tell
interface. Note that the cache size post trimming is not checked
asok_hook,
"show cache status");
assert(r == 0);
- r = admin_socket->register_command("cache drop",
- "cache drop name=timeout,type=CephInt,range=0,req=false",
- asok_hook,
- "drop cache");
- assert(r == 0);
r = admin_socket->register_command("dump tree",
"dump tree "
"name=root,type=CephString,req=true "
} else if (command == "cache status") {
Mutex::Locker l(mds_lock);
mdcache->cache_status(f);
- } else if (command == "cache drop") {
- int64_t timeout;
- if (!cmd_getval(g_ceph_context, cmdmap, "timeout", timeout)) {
- timeout = 0;
- }
-
- C_SaferCond cond;
- command_cache_drop((uint64_t)timeout, f, &cond);
- int r = cond.wait();
- if (r != 0) {
- f->flush(ss);
- }
} else if (command == "dump tree") {
command_dump_tree(cmdmap, ss, f);
} else if (command == "dump loads") {
projects / ceph.git / commitdiff