osd: rebuild attrs from clients

Fixes: http://tracker.ceph.com/issues/24768
Signed-off-by: Kefu Chai <kchai@redhat.com>

osd: fix the FTBFS

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

Merge PR #22907 into jewel

* refs/pull/22907/head:
qa/suites/krbd: use 8 osds instead of 6

Reviewed-by: Ilya Dryomov <idryomov@redhat.com>

Merge PR #22916 into jewel

* refs/pull/22916/head:
Revert "rgw: fix index update in dir_suggest_changes"

Reviewed-by: J. Eric Ivancich <ivancich@redhat.com>

Merge PR #22920 into jewel

* refs/pull/22920/head:
Revert "Merge pull request #17893 from dzafman/wip-19140-19224"

Reviewed-by: David Zafman <dzafman@redhat.com>

Merge pull request #22913 from smithfarm/wip-24742-jewel

jewel: CLI unit formatting tests are broken

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

Merge remote-tracking branch 'private/wip-mon-snap-caps-jewel' into jewel

Merge remote-tracking branch 'private/wip-cx-jewel' into jewel

Revert "Merge pull request #17893 from dzafman/wip-19140-19224"

This reverts commit b5f71c68608f1c0debae7efd39c5f12787776a21, reversing
changes made to 93d589af3f245cfe45a366ce4ae8473d62685ca0.

This resulted in several failures due to the objecter sending out of order
reads vs writes after going from full -> non-full.  Not clear why.

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

Revert "rgw: fix index update in dir_suggest_changes"

This reverts commit 2e3f9de81959d4dc8becb17e3e11d5a69e927d04.

This causes a regression in the cls_rgw index_suggest test, see
http://tracker.ceph.com/issues/24640.

Marking http://tracker.ceph.com/issues/24627 back to In Progress.

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

Merge pull request #22914 from liewegas/wip-ostest-jewel

jewel: ceph_test_objectstore: don't test kstore

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

ceph_test_objectstore: don't test kstore

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

qa/workunits/rbd: adapt import_export test to handle multiple units

Fixes: http://tracker.ceph.com/issues/24740
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 620d10e0a4608153c8571ea1b6d1c3e1d3f63685)

Conflicts:
qa/workunits/rbd/import_export.sh

qa/suites/krbd: use 8 osds instead of 6

This is done in master.

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

msg/async: recover lock to continue later

Signed-off-by: Haomai Wang <haomai@xsky.com>
(cherry picked from commit 86f76f96df53168f6551270dff94515156c3ddef)

Merge pull request #22881 from ceph/wip-cd-fix-branch

jewel bp: fix ceph-deploy branch for jewel tests

qa/tests: for pre luminous, fix testing branch of ceph-deploy to 1.5.39 as the
new master branch assumes ceph-volume by default.

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>

qa/workunits/rados: test pool op permissions

Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 97e3f0aa013dad49de1a55f8025327de7e801822)
(cherry picked from commit b0d6e82a8c95380183af099a50a638f308b67090)

Conflicts:
qa/workunits/rados/test_pool_access.sh: use CEPH_KEYRING env variable

(cherry picked from commit 40abf0b8ba3feccd55d71de638a9a98c7369c9d6)

qa/workunits/rbd: test self-managed snapshot create/remove permissions

Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit b12dd0bf419ae834abb31c712830fa9c4b5cda9c)
(cherry picked from commit 37926b092dd1434e8fdca4620e86c0168aa26b24)

Conflicts:
qa/workunits/rbd/permissions.sh: use CEPH_KEYRING env variable
(cherry picked from commit 76b03a265fecc07de2eb9d1d014444bb9e9a2887)

Conflicts:
qa/workunits/rbd/permissions.sh: dropped tests for profile caps

Merge pull request #22677 from ivancich/wip-fix-index-update-jewel

jewel: rgw: fix index update in dir_suggest_changes

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

Merge pull request #22407 from pdvian/wip-24291-jewel

jewel: common: fix typo in rados bench write JSON output

Reviewed-By: Kefu Chai <kchai@redhat.com>
Reviewed-By: Nathan Cutler <ncutler@suse.com>

Merge pull request #22431 from pdvian/wip-24293-jewel

jewel: mon/LogMonitor: call no_reply() on ignored log message

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

pybind/rados: new methods for manipulating self-managed snapshots

Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 21538304d78df9f3b9f54a5b3c9b6b62fc3e6b48)
(cherry picked from commit 5bd343db3903d8aca45a40436a6dc240bdc1a0f3)
(cherry picked from commit a45f60e0d1a940bef21ea00d0defc7a88a8c7700)

mon/OSDMonitor: enforce caps for all remaining pool ops

Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 61c06200fe59996bba2bb65fc402207bc10fd459)
(cherry picked from commit fb4526690ccd29f1a03d3cd7e5484b9e2191b04b)
(cherry picked from commit 877a319a4e9e328b2954c210d1bc6da29931dc31)

mon/OSDMonitor: enforce caps when creating/deleting unmanaged snapshots

The entity will require write access to the OSD service or permission
for the synthetic "osd pool op unmanaged-snap" command.

Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 4972e054b32c8200600f27564d50d443e683153e)

Conflicts:
src/mon/OSDMonitor.cc: use 'bufferlist::begin' instead of 'bufferlist::cbegin'
(cherry picked from commit 3dfb23d036c068ec5166151ed63628738d68908a)

Conflicts:
src/mon/CMakeLists.txt: added 'OSDCap.cc'
(cherry picked from commit 6b242c8bfc469a5cd560a902165d48405099d758)

Conflicts:
src/mon/CMakeLists.txt: revert to automake
src/mon/OSDMonitor.cc: updated 'MonCap::is_capable' signature
       OSDCap doesn't support profiles

rgw: fix index update in dir_suggest_changes

1. do nothing if the index is no longer exists in cls
   rgw_dir_suggest_changes, whatever the pending state we got in
   check_disk_state, to avoid recreate index when list race with
   delete. the situations:

   * if dir_suggest_changes is done before complete op in Delete, the
     index will be deleted finally.

   * if dir_suggest_changes is done after complete op in Delete, the
     index will no exists and do nothing.

2. stale pending write op's index is existent but exists field is
   false, so remove cur_disk.exists check in op CEPH_RGW_UPDATE, make
   it recover the situation again.

fixes: http://tracker.ceph.com/issues/24627

jewel backport of: https://github.com/ceph/ceph/pull/22217

Signed-off-by: Tianshan Qu <tianshan@xsky.com>
(cherry picked from commit 8304c3e934d667fb89dbe0a79b252a5610b5eb62)

Merge pull request #21367 from tchaikov/wip-23673-jewel

jewel: auth: ceph auth add does not sanity-check caps

Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Reviewed-by: Nathan Cutler <ncutler@suse.com>

Merge pull request #21754 from pdvian/wip-23905-jewel

jewel: osdc/Objecter: prevent double-invocation of linger op callback

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

mon/LogMonitor: call no_reply() on ignored log message

If we're dropping it on the floor, we need to tell the mon that, so that
it can tell the forwarding mon to discard its state.

Fixes: https://tracker.ceph.com/issues/24180
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 9661fa08ccc500587d35d7af8a905916167d8314)

common: fix typo in rados bench write JSON output

Fixes typo in `max_latency` key in JSON output from rados bench write.

Fixes: http://tracker.ceph.com/issues/24199
Signed-off-by: Sandor Zeestraten <sandor@zeestrataca.com>
(cherry picked from commit 1491e08e422d6ad5e612205accd0134203cdc563)

Conflicts:
src/common/obj_bencher.cc : Resolved in write_bench

auth/cephx/CephxProtocol: better random

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 721d7e622bb4b301f0a69a6e236d25c49bd621ad)

- get_random_bytes(), not cct->random()

(cherry picked from commit b98db30e41bb623e2d03636aedabb6a5a59af596)

cephx: update docs

Remove confusing references to ancient releases.  Be a bit more clear
about what signatures do/don't protect.

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 1004c050a1d750d850379b7f66b8559302faac8b)
(cherry picked from commit 0f1ecf11732dd59ccb100da09f291f5b4aec59c6)

auth/cephx: add authorizer challenge

Allow the accepting side of a connection to reject an initial authorizer
with a random challenge.  The connecting side then has to respond with an
updated authorizer proving they are able to decrypt the service's challenge
and that the new authorizer was produced for this specific connection
instance.

The accepting side requires this challenge and response unconditionally
if the client side advertises they have the feature bit.  Servers wishing
to require this improved level of authentication simply have to require
the appropriate feature.

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit f80b848d3f830eb6dba50123e04385173fa4540b)

# Conflicts:
# src/auth/Auth.h
# src/auth/cephx/CephxProtocol.cc
# src/auth/cephx/CephxProtocol.h
# src/auth/none/AuthNoneProtocol.h
# src/msg/Dispatcher.h
# src/msg/async/AsyncConnection.cc

- const_iterator
- ::decode vs decode
- AsyncConnection ctor arg noise
- get_random_bytes(), not cct->random()

(cherry picked from commit 5ead97120e07054d80623dada90a5cc764c28468)

# Conflicts:
# src/auth/cephx/CephxAuthorizeHandler.h
# src/auth/cephx/CephxProtocol.h
# src/auth/none/AuthNoneAuthorizeHandler.h
# src/auth/none/AuthNoneProtocol.h
# src/auth/unknown/AuthUnknownAuthorizeHandler.h
# src/mds/MDSDaemon.cc
# src/mds/MDSDaemon.h
# src/mgr/DaemonServer.cc
# src/mgr/DaemonServer.h
# src/mon/Monitor.cc
# src/mon/Monitor.h
# src/msg/async/AsyncConnection.cc
# src/osd/OSD.cc
# src/osd/OSD.h
# src/test/messenger/simple_dispatcher.h
# src/test/msgr/perf_msgr_client.cc
# src/test/msgr/perf_msgr_server.cc
# src/test/msgr/test_msgr.cc

- lots of override annotation conflicts
  - _refused() callbacks no present in jewel
- lots of msg/async conflicts (code has changed a fair bit)
- we inherited some upstream rotating keys checks, see
  f159a093ecab4aa53693ec106c00d5ecb383c467

mon,msg: implement cephx_*_require_version options

Add a new set of options that are parallel to

 cephx_require_signatures
 cephx_cluster_require_signatures
 cephx_service_require_signatures

that indicate which version of cephx is required to be supported.

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 9bcbc2a3621f9ee4558c9748ab7c73c6e1b22510)
(cherry picked from commit 94b1ca198115c867691babdae7e7ac046689a749)

# Conflicts:
# src/common/legacy_config_opts.h
# src/common/options.cc

- config_opts.h
- no MGR
- FEATURE not FEATUREMASK

Merge pull request #20877 from theanalyst/jewel-23338

jewel: rgw: radosgw-admin: add an option to reset user stats

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

Merge pull request #21212 from smithfarm/wip-22939-jewel

jewel: rgw: system user can't delete bucket completely

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

Merge pull request #21294 from joke-lee/jewel

jewel: rgw: copying part without http header x-amz-copy-source-range will be mistaken for copying object

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

mds: change libmon to .la archive

before this change, it is a plain .a archive. because MDSAuthCaps.cc is
included by both libmon.a and libmds.la, automake complains that

src/Makefile.am: error: object 'mds/MDSAuthCaps.$(OBJEXT)' created both
with libtool and without

so we need to consolidate them. as most of the convenience libraries are
are created using libtool. to be more consistent, i am chaning libmon.a
to libmon.la.

Signed-off-by: Kefu Chai <kchai@redhat.com>
Conflicts:
   src/Makefile-env.am
   src/mds/Makefile-server.am: this change is not cherry-picked from
master. as we switched from autotools to cmake since luminous.

doc: mention stricter caps checking in PendingReleaseNotes

Signed-off-by: Nathan Cutler <ncutler@suse.com>

qa/workunits/cephtool: test with valid auth syntax

'allow' is not a valid mon cap; use 'allow *'.  Add some negative tests
to ensure the validation is working.

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 497a63251b0d7b8b953bb01d0edb838b8b35919a)

qa/workunits/rest/test.py: update this change also, as the rest proxy
was removed in master.

mon/AuthMonitor: check caps validity for all cap-related commands

Add a validity check for 'auth add' and 'fs authorize'.  'auth caps' and
'auth get-or-create[-key]' already had the check.

Fixes: http://tracker.ceph.com/issues/22525
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit fa588730bfea17eb1ca556e20ffcfd0b7628d093)

Conflicts:
src/mon/AuthMonitor.cc (no "fs autorize" in jewel)

mon: expand cap validity check for mgr, osd, mds

Also strenghen the check to declare an unknown cap type as invalid.

Note that this means that in a mixed-version cluster, an older mon
would apply the syntax check for the older caps, even if the (say) OSDs
are newer and could parse something different.  This is judged to be fine:
it is not unreasonable to ask for the mons *and* daemons to be upgraded
before using the new cap syntax or feature.

Signed-off-by: Jing Li <lijing@gohighsec.com>
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 113fa941e3a4feaed0891d1fb7d321ade3e1bca7)

Conflicts:
src/mon/AuthMonitor.cc (no mgr in jewel)
src/mon/CMakeLists.txt (no cmake in jewel; backported the changes
                               manually to src/mon/Makefile.am)

auth/cephx/CephxSessionHandler: implement CEPHX_V2 calculation mode

Derive the signature from the entire buffer (both cipher blocks).

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 8f396cf35a3826044b089141667a196454c0a587)

# Conflicts:
# src/auth/cephx/CephxSessionHandler.cc

- master and mimic use no_bl crypto interface, which doesn't exist here,
so fall back to using bufferlist-based calls into encrypt().

(cherry picked from commit a2b04cc337a6f6f7b7a8b02bf31a8f3448670645)

# Conflicts:
# src/auth/cephx/CephxSessionHandler.cc

- mswab differences

include/ceph_features: define CEPHX2 feature

This indicates support for cephx v2.

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit d672a7afe62ade4960e3e7f935c004f15945f39b)
(cherry picked from commit 52318ad035d2e1698b021e85a34d9d98c4bc078c)

# Conflicts:
# src/include/ceph_features.h

Merge pull request #17818 from smithfarm/wip-upgrade-rh-74

jewel: tests: make upgrade tests use supported distros

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

Merge pull request #21194 from dzafman/wip-23267-jewel

jewel: scrub errors not cleared on replicas can cause inconsistent pg state when replica takes over primary

Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Reviewed-by: David Zafman <dzafman@redhat.com>

Merge pull request #21747 from vumrao/wip-vumrao-rhbz1568897

jewel: OSD: print log when osd want to kill self

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

Merge pull request #22050 from tchaikov/jewel-22530

jewel: mon/OSDMonitor.cc: fix expected_num_objects interpret error

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

Merge pull request #22188 from tchaikov/jewel-23966

jewel: Deleting a pool with active notify linger ops can result in seg fault

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

Merge pull request #22189 from theanalyst/wip-jewel-24244

mon/MDSMonitor: no_reply on MMDSLoadTargets

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

Merge pull request #21162 from smithfarm/wip-22378-jewel

jewel: ceph-fuse: failure to remount in startup test does not handle client_die_on_failed_remount properly

Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>

Merge pull request #21734 from joscollin/wip-23932-jewel

jewel: client: avoid second lock on client_lock

Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>

mon/MDSMonitor: no_reply on MMDSLoadTargets

If we don't note that we don't reply then we don't close out the routed
mon request and the op will appear as slow on the forwarding mon.

Fixes: http://tracker.ceph.com/issues/23769
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit b462b59065424520170956581e72e16481b16f0a)

librados: fix potential race condition if notify immediately fails

Fixes: http://tracker.ceph.com/issues/23966
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit cc06fbab08ef5b7f1b18f2af333c019ad86de03d)

osdc: guard op->on_notify_finish with lock

Fixes: http://tracker.ceph.com/issues/23966
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit c16371965cc785f2e2b87d8c6684854fa2de65e1)

osdc: invoke notify finish context on linger commit failure

Fixes: http://tracker.ceph.com/issues/23966
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit fcd414f6324136ce0695fcf30a60eec6ff338a98)

Merge pull request #21503 from smithfarm/wip-23783-jewel

jewel: table of contents doesn't render for luminous/jewel docs

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

Merge pull request #22166 from tchaikov/jewel-javah

jewel: makefile: use javac for creating JNI headers

Reviewed-by: Yuri Weinstein <yweins@redhat.com>

makefile: bump java source option from 5 to 6

this kills the errors like

error: Source option 5 is no longer supported. Use 6 or later.
error: Target option 1.5 is no longer supported. Use 1.6 or later.

Signed-off-by: Kefu Chai <kchai@redhat.com>
Conflicts:
    src/java/Makefile.am: this change is not cherry-picked from
master. as we've moved to java 8 in master. see df27ed3a .

makefile: use javac for creating JNI headers

JDK 1.10 does not offer javah anymore, so we need to use "javac -h"
instead.

Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit edc504b9d446cc1739f3e8e087914cfe66d6cac3)
Conflicts: configure.ac:
           src/java/Makefile.am: we don't use cmake back in jewel, so
port the cmake changes to autotools.

Merge pull request #22130 from ceph/wip-yuriw-clients-fix-jewel

qa/test: added supported distro

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

qa/test: added supported ditro

Signed-off-by: Yuri Weinstein <yweinste@redhat.com>

Merge pull request #21210 from pdvian/wip-23303-jewel

jewel: rgw: add radosgw-admin sync error trim to trim sync error log

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

mon/OSDMonitor.cc: fix expected_num_objects interpret error

Fixes: http://tracker.ceph.com/issues/22530
Signed-off-by: Yang Honggang <joseph.yang@xtaotech.com>
(cherry picked from commit c24f2baec968ff66e792991d829591505ffba3ae)

Conflicts:
qa/standalone/mon/osd-pool-create.sh: jewel does not have
qa/standalone tests by then. and in jewel, "ruleset" is not renamed to
"rule" yet.
src/mon/OSDMonitor.cc
src/pybind/mgr/localpool/module.py: trivial resolution

Merge pull request #21190 from smithfarm/wip-21033-jewel

jewel: build/ops: rpm: bump epoch ahead of ceph-common in RHEL base

Reviewed-by: Ken Dreyer <kdreyer@redhat.com>
Reviewed-by: Boris Ranto <branto@redhat.com>

Merge pull request #21213 from smithfarm/wip-23026-jewel

jewel: rgw: data sync of versioned objects, note updating bi marker

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

osdc/Objecter: prevent double-invocation of linger op callback

Fixes: http://tracker.ceph.com/issues/23872
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 1bc6f31f8f5ce33176b8a346492b087728947a71)

OSD: print log when osd want to kill self

Signed-off-by: Haomai Wang <haomai@xsky.com>
(cherry picked from commit 9bc7d9667076473deda4a9214eb74955458eb36f)

client: avoid second lock on client_lock

Avoid a second nested lock on client lock. As a result, drop unused function ll_get_inodeno().
Fixes: https://tracker.ceph.com/issues/23815
Signed-off-by: Supriti Singh <supriti.singh@suse.com>
(cherry picked from commit 7be8d559a11507255f847ce6279d23abaabde70a)

 Conflicts:
src/client/Client.cc (dropped function call ll_get_inodeno(in))

rgw: rgw-rados, rgw-admin add an option to recalculate user stats

Adds a method in rgw-rados to reset user stats calling the earlier implemented
cls user reset stats.
In rgw-admin we add an option called --reset-stats that invokes this method.

Fixes: http://tracker.ceph.com/issues/23335
Signed-off-by: Abhishek Lekshmanan <abhishek@suse.com>
(cherry picked from commit 0b72dd7d3beea47b5c479eeb979b73c4cd70f7bc)

 Conflicts:
  src/rgw/rgw_admin.cc
  whitespace conflict
  src/rgw_rados.cc
  switch to jewel version of get_obj_ref (where raw version isn't available)

jewel only admin fixes

cls/user: implement reset user stats

This is an implementation of reset user stats, that recalculates the user stats
purely based on the values of bucket entries in user.buckets object. This is
helpful in cases when user stats has been improperly set in case of manual
resharding etc.

Fixes: http://tracker.ceph.com/issues/23335
Signed-off-by: Abhishek Lekshmanan <abhishek@suse.com>
(cherry picked from commit 63f9127dd685890d938a612739320c942a96f910)

* conflicts:
cls_user.cc:
jewel only cls fixes, truncated is not a part of omap get vals, so iterating
over the range by checking the output of omap_get_vals and continuing.
cls_user_ops.h:
explicit ::encode/::decode

doc update template call for toctree

Up until version 1.1.3 of Sphinx, this call would show hidden toctree
entries. Most of Ceph's toctree's are hidden, so when the version of
Sphinx was bumped to >1.6, the sidebar stopped showing entries

Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 9bb3397a5876ded89853f11cd35b3ca5e8aeec5c)

osd: Fix stale scrub stats when a primary takes over

Fixes: http://tracker.ceph.com/issues/23267
Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit 5cfb8241f482ed53c63bb97262425b2acb733d7d)

Conflicts:
src/osd/PG.cc (trivial)
test/Makefile.am (Add to make check for Jewel)
test/osd/CMakeLists.txt (Add to make check for Jewel)
src/test/osd/osd-scrub-test.sh (from qa/standalone/scrub/osd-scrub-test.sh)
  Different location of ceph-helpers.sh
  No manager so remove run_mgr
  Remove default rbd pool
  Use 4 OSDs because remapped PG doesn't go clean

client: quit on failed remount during dentry invalidate test

Fixes: http://tracker.ceph.com/issues/22269
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
(cherry picked from commit 5b2b1d14468c290c56ee6c95ea557c99464e0098)

Conflicts:
PendingReleaseNotes - put the release note under 10.2.11
src/ceph_fuse.cc - changed order of client_try_dentry_invalidate and
    can_invalidate_dentries assignments to match master, and use jewel
            form of g_conf for accessing config option values
src/client/Client.cc - code being replaced by _do_remount() calls is
    different in jewel, and use jewel convention for accessing config
            option values
src/common/legacy_config_opts.h - file does not exist in jewel
src/common/options.cc - file does not exit in jewel; made the changes
            manually in src/common/config_opts.h

Merge pull request #21185 from smithfarm/wip-23158-jewel

jewel: mds: underwater dentry check in CDir::_omap_fetched is racy

Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>

Merge pull request #18010 from jdurgin/wip-rocksdb-jewel

jewel: core: enable rocksdb for filestore

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

Merge pull request #21199 from smithfarm/wip-22403-jewel

jewel: osd: replica read can trigger cache promotion

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

Merge pull request #21200 from ovh/bp-jewel-backport-22448

jewel: osd/PrimaryLogPG: dump snap_trimq size

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

Merge pull request #21197 from smithfarm/wip-23521-jewel

jewel: ceph_authtool: add mode option

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

Merge pull request #20381 from pdvian/wip-22941-jewel

jewel: core: librados: Double free in rados_getxattrs_next

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

rgw: If use 'copy part' without 'x-amz-copy-source-range',

it will copy an entire source object

Signed-off-by: Malcolm Lee <fengxueyu35@126.com>
(cherry picked from commit 753b3a7)

Conflicts:
 1. in original commit, copy_source is std::string, jewel
    version is const char *.
 2. in original commit, url_decode take only one param,
    jewel version take 2 params.
 3. in original commit, get_obj_state take 6 params, jewel
    version takes 4 params.

Merge pull request #21203 from smithfarm/wip-22394-jewel

jewel: librbd: cannot copy all image-metas if we have more than 64 key/value pairs

Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Reviewed-by: Mykola Golub <mgolub@suse.com>

Merge pull request #21098 from smithfarm/wip-23348-jewel

jewel: rgw: inefficient buffer usage for PUTs

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

Merge pull request #21215 from smithfarm/wip-23546-jewel

jewel: journal: Message too long error when appending journal

Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Reviewed-by: Mykola Golub <mgolub@suse.com>

Merge pull request #21189 from smithfarm/wip-23356-jewel

jewel: cephfs: client: prevent fallback to remount when dentry_invalidate_cb is true but root->dir is NULL

Reviewed-by: Yan, Zheng <zyan@redhat.com>

Merge pull request #21232 from pdvian/wip-23543-jewel

jewel: rbd-nbd: fix ebusy when do map

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

Merge pull request #21228 from dillaman/wip-22396-jewel

jewel: librbd: cannot clone all image-metas if we have more than 64 key/value pairs

Reviewed-by: Mykola Golub <mgolub@suse.com>

Merge pull request #21227 from dillaman/wip-21442-jewel

jewel: rbd: clean up warnings when mirror commands used on non-setup pool

Reviewed-by: Mykola Golub <mgolub@suse.com>

Merge pull request #21225 from dillaman/wip-20637-jewel

jewel: rbd-mirror: ignore permission errors on rbd_mirroring object

Reviewed-by: Mykola Golub <mgolub@suse.com>

Merge pull request #21224 from dillaman/wip-21689-jewel

jewel: librbd: list_children should not attempt to refresh image

Reviewed-by: Mykola Golub <mgolub@suse.com>

Merge pull request #21223 from dillaman/wip-21915-jewel

jewel: rbd-mirror: strip environment/CLI overrides for remote cluster

Reviewed-by: Mykola Golub <mgolub@suse.com>

Merge pull request #21220 from dillaman/wip-21867-jewel

jewel: librbd: object map batch update might cause OSD suicide timeout

Reviewed-by: Mykola Golub <mgolub@suse.com>

Merge pull request #21219 from dillaman/wip-23305-jewel

jewel: librbd: create+truncate for whole-object layered discards

Reviewed-by: Mykola Golub <mgolub@suse.com>

Merge pull request #21207 from smithfarm/wip-23525-jewel

jewel: rbd: is_qemu_running in qemu_rebuild_object_map.sh and qemu_dynamic_features.sh may return false positive

Reviewed-by: Mykola Golub <mgolub@suse.com>

Merge pull request #21206 from smithfarm/wip-23012-jewel

jewel: rbd: [journal] allocating a new tag after acquiring the lock should use on-disk committed position

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

Merge pull request #21205 from smithfarm/wip-22578-jewel

jewel: rbd: rbd-mirror split brain test case can have a false-positive failure until teuthology

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

Merge pull request #21125 from smithfarm/wip-23508-jewel

jewel: tests: test_admin_socket.sh may fail on wait_for_clean

Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Reviewed-by: Mykola Golub <mgolub@suse.com>

Merge pull request #20627 from pdvian/wip-23153-jewel

jewel: rbd: TestLibRBD.RenameViaLockOwner may still fail with -ENOENT

Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Reviewed-by: Mykola Golub <mgolub@suse.com>

rgw: fix memory fragmentation problem reading data from client.

mg_read returns 0 on EOF.  In some versions of civetweb, mg_read can
return "short" reads.  The logic in the rest of ceph depends on a read
always filling its buffer when possible.  So loop here and fill the
buffer.  Looping to fill the buffer means we must also track when
mg_read returns EOF, because after that further calls to mg_read will return -1.

Fixes: https://tracker.ceph.com/issues/23207
Signed-off-by: Marcus Watts <mwatts@redhat.com>
(cherry picked from commit d9a150b1f1f9ec3952555186722c9b13d7448e2d)

Conflicts:
    src/rgw/rgw_civetweb.cc:
        - master RGWCivetWeb::RGWCivetWeb() becomes RGWMongoose::RGWMongoose() in jewel,
          with slightly different initializer list; adapt mg_read() call for jewel
        - master RGWCivetWeb::read_data() becomes RGWMongoose::read_data() in jewel,
          and inside this function we return -EIO in jewel, instead of throwing an exception

Merge pull request #18304 from smithfarm/wip-21546-jewel

jewel: rgw: file write error

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