10.2.5

Merge pull request #12376 from liewegas/wip-msgr-eagain-loop-jewel

msg/simple/Pipe: avoid returning 0 on poll timeout

Reviewed-by: Greg Farnum <gfarnum@redhat.com>

msg/simple/Pipe: avoid returning 0 on poll timeout

If poll times out it will return 0 (no data to read on socket).  In
165e5abdbf6311974d4001e43982b83d06f9e0cc we changed tcp_read_wait from
returning -1 to returning -errno, which means we return 0 instead of -1
in this case.

This makes tcp_read() get into an infinite loop by repeatedly trying to
read from the socket and getting EAGAIN.

Fix by explicitly checking for a 0 return from poll(2) and returning
EAGAIN in that case.

Fixes: http://tracker.ceph.com/issues/18184
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 6c3d015c6854a12cda40673848813d968ff6afae)

10.2.4

Merge pull request #12167 from liewegas/wip-osdmap-encoding-jewel

jewel: osd: condition OSDMap encoding on features

Reviewed-by: Loic Dachary <ldachary@redhat.com>

Merge pull request #12067 from SUSE/wip-17953-jewel

jewel: mon: OSDMonitor: only reject MOSDBoot based on up_from if inst matches

Reviewed-by: Samuel Just <sjust@redhat.com>

OSDMonitor: only reject MOSDBoot based on up_from if inst matches

If the osd actually restarts, there is no guarrantee that the epoch will
advance past up_from.  If the inst is different, it can't really be a
dup.  At worst, it might be a queued MOSDBoot from a previous inst, but
in that case, the real inst would see itself marked up, and then back
down causing it to try booting again.

Fixes: http://tracker.ceph.com/issues/17899
Signed-off-by: Samuel Just <sjust@redhat.com>
(cherry picked from commit 033ad5b46c0492134e72a8372e44e3ef1358d2df)

Merge pull request #12207 from jdurgin/wip-librados-setxattr-overload-jewel

librados: remove new setxattr overload to avoid breaking the C++ ABI

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

Merge pull request #12267 from dachary/wip-17904-jewel

 jewel: Error EINVAL: removing mon.a at 172.21.15.16:6789/0, there will be 1 monitors

Reviewed-by: Samuel Just <sjust@redhat.com>

mon: MonmapMonitor: drop unnecessary 'goto' statements

Signed-off-by: Joao Eduardo Luis <joao@suse.de>
(cherry picked from commit 20dcb597e35e6961db81831facefbe22cecddec3)

mon: MonmapMonitor: return success when monitor will be removed

Fixes: http://tracker.ceph.com/issues/17725
Signed-off-by: Joao Eduardo Luis <joao@suse.de>
(cherry picked from commit c9d46cfbf2512bc3495c6901de2b8f711bef9bae)

Merge pull request #12001 from dachary/wip-17915-jewel

jewel: filestore: can get stuck in an unbounded loop during scrub

Reviewed-by: Samuel Just <sjust@redhat.com>

librados: remove new setxattr overload to avoid breaking the C++ ABI

Fixes: http://tracker.ceph.com/issues/18058
Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit b8ff781ddcf737882163cf56d7c9b11e815fb699)

Conflicts:
src/include/rados/librados.hpp (trivial namespace change in removed line)

crush: condition latest tunable encoding on features

This avoids throwing hammer OSDMap encodings off.

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 9e5ff86487bd1f5979866b5e16300dd4a3979f97)

crush/CrushWrapper: encode with features

No behavior change yet; just fixing callers.

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit b7c9e055848c8aa951bc48c957cff3ef323ea441)

[Updated write_file to use all feaetures]
[Updated OSDMonitor.cc to use mon->quorum_features instead of the
 mon->get_quorum_con_featuers() helper]
[trivial conflict from removed write_file and read_file]

Conflicts:
src/crush/CrushWrapper.h
src/mgr/PyModules.cc
src/mon/OSDMonitor.cc
src/tools/ceph_monstore_tool.cc

crush/CrushWrapper: drop unused 'lean' encode() argument

No callers, no users.

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 638a38bbb55c07ad0358a35a56418e66874d1c26)

Conflicts:
src/crush/CrushWrapper.h

[trivial conflict due to removal of write_file and read_file]

osd/osd_types: encode pg_pool_t like hammer if features indicate hammer

If the target features are missing the new OSDOp encoding, the
first feature we added post-hammer, encode like hammer.

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 2f8cfb632823ba4e63eaff394392d6af7979d7c8)

osd/osd_types: conditional pg_pool_t encoding

Align this with decode.

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 01d9e8a20bbc3c039f67b040da95018e2c7b00b6)

os/filestore/HashIndex: fix list_by_hash_* termination on reaching end

If we set *next to max, then the caller (a few lines up) doesn't terminate
the loop and will keep trying to list objects in every following hash
dir until it reaches the end of the collection.  In fact, if we have an
end bound we will never to an efficient listing unless we hit the max
first.

For one user, this was causing OSD suicides when scrub ran because it
wasn't able to list all objects before the timeout.  In general, this would
cause scrub to stall a PG for a long time and slow down requests.

Broken by refactor in 921c4586f165ce39c17ef8b579c548dc8f6f4500.

Fixes: http://tracker.ceph.com/issues/17859
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit c5180262a086c2d3895aff4bf0fb0ff9a6666149)

Merge pull request #11822 from SUSE/wip-17816-jewel

jewel: Missing comma in ceph-create-keys causes concatenation of arguments

Reviewed-by: Loic Dachary <ldachary@redhat.com>

ceph-create-keys: add missing argument comma

The arguments "get" and "client.admin" were being concatenated into
"getclient.admin".

Found using ceph-ansible + strace:

    13031 execve("/usr/bin/ceph", ["ceph", "--cluster=ceph", "--name=mon.", "--keyring=/var/lib/ceph/mon/ceph-ceph-mon0/keyring", "auth", "getclient.admin"], ["PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin", "LANG=en_US.UTF-8", "CLUSTER=ceph", "TCMALLOC_MAX_TOTAL_THREAD_CACHE_BYTES=134217728", "CEPH_AUTO_RESTART_ON_UPGRADE=no"] <unfinished ...>

Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
(cherry picked from commit 482022233d845b75876b04ca23fb137281a9f6ab)

Merge pull request #11679 from dachary/wip-17734-jewel

jewel: Upgrading 0.94.6 -> 0.94.9 saturating mon node networking

Reviewed-by: Kefu Chai <kchai@redhat.com>

mon: expose require_jewel_osds flag to user

Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
(cherry picked from commit 83ffc2b761742d563777e50959faa6a6010edae0)

mon/OSDMonitor: encode OSDMap::Incremental with same features as OSDMap

The Incremental encode stashes encode_features, which is
what we use later to reencode the updated OSDMap.  Use
the same features so that the encoding will match!

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 916ca6a0aaa32bd9c2b449e0d7fbd312c29f06e5)

mon/OSDMonitor: health warn if require_{jewel,kraken} flags aren't set

We want to prompt users to set these flags as soon as their
upgrades complete.

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 12e508313dbd5d1d38c76859cb7de2ce22404e12)

Conflicts:
   src/mon/OSDMonitor.cc: remove references to kraken

    if ((osdmap.get_up_osd_features() & CEPH_FEATURE_SERVER_KRAKEN) &&
!osdmap.test_flag(CEPH_OSDMAP_REQUIRE_KRAKEN)) {
      string msg = "all OSDs are running kraken or later but the"
" 'require_kraken_osds' osdmap flag is not set";
      summary.push_back(make_pair(HEALTH_WARN, msg));
      if (detail) {
detail->push_back(make_pair(HEALTH_WARN, msg));
      }
    } else

mon/OSDMonitor: encode canonical full osdmap based on osdmap flags

If the JEWEL or KRAKEN flags aren't set, encode the full map without
those features.  This ensure that older OSDs in the cluster will be able
to correctly encode the full map with a matching CRC.  At least, that is
true as long as the encoding changes are guarded by those feature bits.
That appears to be true currently, and we plan to ensure that it is true
in the future as well.

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 5e0daf6642011bf1222c4dc20aa284966fa5df9f)

Conflicts:
   src/mon/OSDMonitor.cc: removed reference to kraken

    if (!tmp.test_flag(CEPH_OSDMAP_REQUIRE_KRAKEN)) {
      dout(10) << __func__ << " encoding without feature SERVER_KRAKEN" << dendl;
      features &= ~CEPH_FEATURE_SERVER_KRAKEN;
    }

Merge pull request #11742 from tchaikov/wip-17728-jewel

jewel: test/ceph_test_msgr: do not use Message::middle for holding transient…

Reviewed-by: Loic Dachary <ldachary@redhat.com>

Merge pull request #11746 from liewegas/wip-post-file-key-jewel

jewel: ceph-post-file: use new ssh key

Reviewed-by: Loic Dachary <ldachary@redhat.com>
Reviewed-by: Brad Hubbard <bhubbard@redhat.com>

ceph-post-file: Ignore keys offered by ssh-agent

In my case, I had multiple private keys in ssh-agent which resulted in
the sftp connection failing despite explicitly specifying the private
key to use

Signed-off-by: David Galloway <dgallowa@redhat.com>
(cherry picked from commit a61fcb2eac35a149b49efdc9b2ffa675afb968e8)

ceph-post-file: migrate to RSA SSH keys

DSA keys are being deprecated: http://www.openssh.com/legacy.html

drop.ceph.com will continue to allow the old DSA key but eventually,
users submitting logs using ceph-post-file will run into issues when
OpenSSH completely drops support for the algorithm.

Fixes: http://tracker.ceph.com/issues/14267
Signed-off-by: David Galloway <dgallowa@redhat.com>
(cherry picked from commit ecd02bf3f1c7a07a3271b2736a9e12dd6e897821)

# Conflicts:
# src/CMakeLists.txt

msg: adjust byte_throttler from Message::encode

Normally we never call encode on a message that has a byte_throttler set
because we only use it for messages we received.  However, for forwarded
messages that we clear_payload() before resending, we *do* reencode, and in
that case we need to retake the appropriate number of bytes from the
throttler--just like we release them in clear_payload().

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit a9651282f7c16df872757b82d3d2995d92458d5c)

msg/Message: fix set_middle vs throttler

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit e7bf50b27a495ed75def67025d1ceca83861ba35)

messages/MForward: reencode forwarded message if target has differing features

This ensures we reencode the payload with the
appropriate set of features if the client, us, or the
target do not have identical features.  Otherwise we
may forward an encoding with more features than the
target can handle.

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit a433455e59067a844c3df4a0d6080db2ceb4ec59)

messages/MForward: fix encoding features

We were encoding the message with the sending client's
features, which makes no sense: we need to encode with
the recipient's features so that it can decode the
message.

The simplest way to fix this is to rip out the bizarre
msg_bl handling code and simply keep a decoded Message
reference, and encode it when we send.

We encode the encapsulated message with the intersection
of the target mon's features and the sending client's
features.  This probably doesn't matter, but it's
conceivable that there is some feature-dependent
behavior in the message encode/decode that is important.

Fixes: http://tracker.ceph.com/issues/17365
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit d4f5e88f36e5388ae9e062c4bc49ac1c684a3f3c)

all: add const to operator<< param

Signed-off-by: Michal Jarzabek <stiopa@gmail.com>
(cherry picked from commit 0a157e088b2e5eb66177421f19f559ca427240eb)

test/ceph_test_msgr: do not use Message::middle for holding transient data

Message::middle is used for holding encoded data, so we we can not stuff
it with payload and leave the "payload" field empty. this change
refactors the ceph_test_msgr by introducing a Payload class which
encodes all test data in it.

Fixes: http://tracker.ceph.com/issues/17728
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 56896a7ed20869ce91ade4c77c1d6cbab8d50de1)
Conflicts:
src/test/msgr/test_msgr.cc: do not use the new-style DENC()
framework for implementing the encoder of Payload class. DENC() was
introduced after jewel was released.

test/ceph_test_msgr: fix circular locking dependency

* do not acquire lock when sending message
* remove lock in session
* reduce the scope guarded by locks for better performance.

Fixes: http://tracker.ceph.com/issues/16955
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit cf1801c260c42aa93850538eea7a194440ebe350)

ceph_test_msgr: use ceph log infrastructure to output

because we want to get the right log sequence which mixes ceph logginer and
cerr. Otherwise, cerr output make the logs a little disordered.

Signed-off-by: Haomai Wang <haomai@xsky.com>
(cherry picked from commit d1268a6aa895ee93cd9fee6fc6c759317e681a85)

Merge pull request #11728 from ceph/wip-librados-upgrade-jewel

qa: remove EnumerateObjects from librados upgrade tests

Merge pull request #11669 from ceph/wip-jewel-11566

rgw: fix put_acls for objects starting and ending with underscore

Reviewed-by: Loic Dachary <ldachary@redhat.com>

Merge pull request #11472 from dachary/wip-17510-jewel

jewel: ERROR: got unexpected error when trying to read object: -2

Reviewed-by: Yehuda Sadeh <ysadehwe@redhat.com>

rgw: fix put_acls for objects starting and ending with underscore

Fixes: http://tracker.ceph.com/issues/17625
Signed-off-by: Orit Wasserman <owasserm@redhat.com>
(cherry picked from commit 14d4d912c7b47c56b16ae0bdc6bc08d208de3461)

Merge pull request #11662 from linuxbox2/jewel-17635

jewel: rgw: handle empty POST condition

Merge pull request #11634 from dillaman/wip-17590-jewel

jewel: journal: do not prematurely flag object recorder as closed

Reviewed-by: Jason Dillaman <dillaman@redhat.com>

qa: remove EnumerateObjects from librados upgrade tests

These rely on new rados functionality not present in jewel

Signed-off-by: Josh Durgin <jdurgin@redhat.com>

Merge pull request #11644 from ceph/wip-17695

jewel: librbd: discard after write can result in assertion failure

Reviewed-by: Jason Dillaman <dillaman@redhat.com>

librbd: discard after write can result in assertion failure

With journaling enabled, the proper lock is not held when handling
a discard after write to overlapping extents. This issue is only present
on the jewel branch due to design changes on the master branch.

Fixes: http://tracker.ceph.com/issues/17695
Signed-off-by: Jason Dillaman <dillaman@redhat.com>

rgw: handle empty POST condition

Fixes: http://tracker.ceph.com/issues/17635
Before accessing json entity, need to check that iterator is valid.
If there is no entry return appropriate error code.

Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
(cherry picked from commit 23cb642243e09ca4a8e104f62a3bb7b2cbb6ea12)

Merge pull request #11657 from dachary/wip-17707-jewel

jewel: ceph-disk: using a regular file as a journal fails

Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #11321 from linuxbox2/jewel-11051

jewel: rgw: Do not archive metadata by default

Reviewed-by: Matt Benjamin <mbenjamin@redhat.com>

Merge pull request #11626 from ceph/wip-jewel-11567

jewel: don't loop forever when reading data from 0 sized segment.

Reviewed-by: Orit Wasserman <owasserm@redhat.com>

Merge pull request #11478 from dachary/wip-17312-jewel

jewel: build/ops: allow building RGW with LDAP disabled

Reviewed-by: Orit Wasserman <owasserm@redhat.com>

ceph-disk: allow using a regular file as a journal

Because of a missing return, ceph-disk prepare would fail if given a
regular file as a journal. If the journal file does not exist, ceph-disk
will create it but fail to ensure that the ceph user owns it. The
symlink to the journal file is not set when the journal file is
specified on the command line and the journal file does not exist at
all. The ceph-osd daemon will silently create it as a file but it will
not be the file given in argument.

Add a test case to verify using a regular file as a journal works as
expected.

Fixes: http://tracker.ceph.com/issues/17662
Signed-off-by: Jayashree Candadai <jayaajay@indiana.edu>
Signed-off-by: Loic Dachary <ldachary@redhat.com>
(cherry picked from commit db917d50eb5f86a07a5487e130f46a6b1d27672a)

ceph-disk: PEP8ify

Signed-off-by: Anirudha Bose <ani07nov@gmail.com>
(cherry picked from commit f7a15ccfa2cc214c3688a2b41ac45be1abfb7700)

Conflicts:
src/ceph-disk/setup.py: python 2.7 restriction not backported

ceph-disk: Set space_symlink to the path, not file object

Signed-off-by: Anirudha Bose <ani07nov@gmail.com>
(cherry picked from commit d290454cf5f660c1681eefd70a38e79da683525f)

ceph-disk: Use os.path.isabs instead of .startswith('/')

Signed-off-by: Anirudha Bose <ani07nov@gmail.com>
(cherry picked from commit 6f795356de00dd7c33282183b11a03da800fde7b)

ceph-disk: Use context manager with FileLock

acquire and release methods of FileLock are dropped

Signed-off-by: Anirudha Bose <ani07nov@gmail.com>
(cherry picked from commit c1011d514ce0c7d340a5acd6f9c640165e169156)

ceph-disk: Fix bug in FileLock

Python fcntl.lockf() accepts a file descriptor, not a file object

Signed-off-by: Anirudha Bose <ani07nov@gmail.com>
(cherry picked from commit df9cc2ce938a969f4044b63fd80030d00f64f060)

ceph-disk: Use true integer division in get_dev_size

Signed-off-by: Anirudha Bose <ani07nov@gmail.com>
(cherry picked from commit f1bb72c82806cc03e85e0b19c83d61409c6b2d51)

ceph-disk: Compatibility fixes for Python 3

ceph-disk: Misc cleanups

Signed-off-by: Anirudha Bose <ani07nov@gmail.com>
(cherry picked from commit d0e29c74f84a2ed3014a516c0106172619314bdc)

Conflicts:
     src/ceph-disk/tox.ini: python3 is not supported in jewel

ceph-disk: change ownership of init file to ceph:ceph

Fixes: http://tracker.ceph.com/issues/16280
Signed-off-by: Shylesh Kumar <shmohan@redhat.com>
(cherry picked from commit aab9d03e1b50ba10a383663088400b9fabe306cb)

rgw: Do not archive metadata by default

Fixes: http://tracker.ceph.com/issues/17256
Signed-off-by: Pavan Rallabhandi <PRallabhandi@walmartlabs.com>
Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
(cherry picked from commit c617ea83e0cca0061af18e0811c7ef8b4e836519)

Merge pull request #11642 from tchaikov/wip-17685-jewel

jewel: mon: fix missing osd metadata (again)

Reviewed-by: Loic Dachary <ldachary@redhat.com>

Merge pull request #11467 from dachary/wip-17262-jewel

jewel: rbd-nbd IO hang

Reviewed-by: Jason Dillaman <dillaman@redhat.com>

mon: fix missing osd metadata (again)

The JSON output was getting broken by continuing
in the wrong place.

Fixes: http://tracker.ceph.com/issues/17685
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit c5700ce4b45b3a385fe4c2111da852bea7d86da2)

journal: do not prematurely flag object recorder as closed

Fixes: http://tracker.ceph.com/issues/17590
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 2be6367cb0119d86dfecfa94eb4a3a102c40162a)

Don't loop forever when reading data from 0 sized segment.

The 0 sized segment can arise depending on how
the client uploads the object in the first place.
The cpu loop then happens during a swift `GET'.

Signed-off-by: Marcus Watts <mwatts@redhat.com>
(cherry picked from commit 46c5f9773246522e66bb2cca49345d0b62a16c42)

Merge pull request #10757 from dachary/wip-17056-jewel

jewel: mon/osdmonitor: decouple adjust_heartbeat_grace and min_down_reporters

Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #10759 from dachary/wip-16866-jewel

jewel: OSD: ceph osd df does not show summarized info correctly if one or more OSDs are out

Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #11590 from dachary/wip-17642-jewel

jewel: TestJournalReplay: sporadic assert(m_state == STATE_READY || m_state == STATE_STOPPING) failure

Reviewed-by: Loic Dachary <ldachary@redhat.com>

test: TestJournalReplay test cases need to wait for committed journal event

Fixes: http://tracker.ceph.com/issues/17566
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 1bdf7a425eb7436838c9445ea2c8ab42dfd3a1b6)

librbd: ignore cache busy errors when shrinking an image

Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 4ce663845679dc35f2f15b893c6f988c4a60b25b)

Conflicts:
     src/test/librbd/operation/test_mock_ResizeRequest.cc:
     when_resize does not have the allow_shrink argument because
     d1f2c557b2c039730baca9efa3f5244bc19dcb1a has not been
     backported

librbd: invalidate cache before trimming image

Any potential writeback outside the extents of a shrunk image
would result in orphaned objects.

Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 3f93a1917437ba7b69b306e4ff971b79e8b79c89)

rbd-nbd: mask out-of-bounds IO errors caused by image shrink

Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit c6cfb616a846959a2cd1c0e540b14668c61a2afd)

rbd-nbd: fix kernel deadlock during teuthology testing

Fixes: http://tracker.ceph.com/issues/16921
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit ce7c1520a8019f011fa34dd898af317f78974577)

Merge pull request #11519 from dachary/wip-17576-jewel

jewel: RGW loses realm/period/zonegroup/zone data: period overwritten if somewhere in the cluster is still running Hammer

Reviewed-by: Casey Bodley <cbodley@redhat.com>

Merge pull request #11471 from dachary/wip-17511-jewel

jewel: s3tests-test-readwrite failing with 500

Reviewed-by: Casey Bodley <cbodley@redhat.com>

Merge pull request #11469 from dachary/wip-17538-jewel

jewel: rgw:user email can modify to empty when it has values

Reviewed-by: Casey Bodley <cbodley@redhat.com>

Merge pull request #11466 from dachary/wip-17290-jewel

jewel: ImageWatcher: use after free within C_UnwatchAndFlush

Reviewed-by: Jason Dillaman <dillaman@redhat.com>

Merge pull request #11464 from dachary/wip-17373-jewel

jewel: image.stat() call in librbdpy fails sometimes

Reviewed-by: Jason Dillaman <dillaman@redhat.com>

Merge pull request #11463 from dachary/wip-17384-jewel

jewel: helgrind: TestLibRBD.TestIOPP potential deadlock closing an image with read-ahead enabled

Reviewed-by: Jason Dillaman <dillaman@redhat.com>

Merge pull request #11462 from dachary/wip-17404-jewel

jewel: update_features API needs to support backwards/forward compatibility

Reviewed-by: Jason Dillaman <dillaman@redhat.com>

Merge pull request #11459 from dachary/wip-17483-jewel

jewel: RBD should restrict mirror enable/disable actions on parents/clones

Reviewed-by: Jason Dillaman <dillaman@redhat.com>

Merge pull request #11460 from dachary/wip-17482-jewel

jewel: Enable/Disable of features is allowed even the features are already enabled/disabled

Reviewed-by: Jason Dillaman <dillaman@redhat.com>

Merge pull request #11461 from dachary/wip-17481-jewel

jewel: Proxied operations shouldn't result in error messages if replayed

Reviewed-by: Jason Dillaman <dillaman@redhat.com>

Merge pull request #11431 from ceph/wip-jewel-tarball

jewel: build/ops: include more files in "make dist" tarball

Reviewed-by: Nathan Cutler <ncutler@suse.cz>
Reviewed-by: Boris Ranto <branto@redhat.com>

Merge pull request #11474 from dachary/wip-17350-jewel

jewel: rgw:response information is error when geting token of swift account

Reviewed-by: Orit Wasserman <owasserm@redhat.com>

Merge pull request #11492 from SUSE/wip-17575-jewel

jewel: aarch64: Compiler-based detection of crc32 extended CPU type is broken

Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #11475 from dachary/wip-17349-jewel

jewel: Modification for TEST S3 ACCESS section in INSTALL CEPH OBJECT GATEWAY page

Reviewed-by: Loic Dachary <ldachary@redhat.com>

Merge pull request #11473 from dachary/wip-17509-jewel

jewel: Config parameter rgw keystone make new tenants in radosgw multitenancy does not work

Reviewed-by: Loic Dachary <ldachary@redhat.com>

Merge pull request #11201 from ceph/wip-backport-logrotate-jewel

jewel: build/ops: backport 'logrotate: Run as root/ceph'

Reviewed-by: Loic Dachary <ldachary@redhat.com>

Merge pull request #11126 from tchaikov/wip-17179-jewel

jewel: add a tool to rebuild mon store from OSD

Reviewed-by: huanwen ren <ren.huanwen@zte.com.cn>
Reviewed-by: Ken Dreyer <kdreyer@redhat.com>

Merge pull request #11563 from ceph/wip-jewel-acl-underscore

rgw: fix regression with handling double underscore

Reviewed-by: Loic Dachary <ldachary@redhat.com>

rgw: set correct instance on the object

Fixes: http://tracker.ceph.com/issues/17443
This was broken by commit bc840afafdfe5e528e5c1b711e71420ac3cb5a67

Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
(cherry picked from commit d2ada35c8c0d938a704ecf9974988ea30a9c0105)

Merge pull request #11548 from dachary/wip-17609-jewel

jewel: tests: ceph-disk must ignore debug monc

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

Merge pull request #11411 from dachary/wip-17245-jewel

jewel: tests: scsi_debug fails /dev/disk/by-partuuid

Reviewed-by: Loic Dachary <ldachary@redhat.com>

Merge pull request #10884 from dachary/wip-17149-jewel

jewel: ceph-disk: expected systemd unit failures are confusing

Reviewed-by: Loic Dachary <ldachary@redhat.com>

rgw: fix regression with handling double underscore

Fixes: http://tracker.ceph.com/issues/16856
Signed-off-by: Orit Wasserman <owasserm@redhat.com>
(cherry picked from commit bc840afafdfe5e528e5c1b711e71420ac3cb5a67)

Merge pull request #11408 from dachary/wip-17345-jewel

jewel: Ceph Status - Segmentation Fault

Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #11407 from dachary/wip-17360-jewel

jewel: ceph-objectstore-tool crashes if --journal-path <a-directory>

Reviewed-by: Kefu Chai <kchai@redhat.com>