test/cls_rbd: mask newer feature bits to support upgrade tests

Fixes: http://tracker.ceph.com/issues/21217
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit aeb8f29d21625f3570e2f3e6dd26594d1a6df810)

11.2.1

Merge pull request #14702 from ceph/wip-bp-kraken-systemd

kraken: qa/tasks: misc systemd updates

Reviewed-by: Nathan Cutler <ncutler@suse.com

Merge pull request #15509 from ceph/wip-ceph-disk-fix-kraken

kraken: selinux: Do parallel relabel on package install

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

Merge pull request #14612 from smithfarm/wip-19651

tests: backport Sage's fixes to qa/suites/upgrade/jewel-x

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

Merge pull request #16298 from smithfarm/wip-20517-kraken

kraken: rbd: cli: map with cephx disabled results in error message

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

Merge pull request #16175 from smithfarm/wip-20263-kraken

kraken: rgw: datalog trim can't work as expected

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

Merge pull request #15985 from prallabh/kraken

kraken: rgw: Custom data header support

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

Merge pull request #16186 from smithfarm/wip-20264-kraken

kraken: rbd: [cli] ensure positional arguments exist before casting

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

Merge pull request #16342 from dillaman/wip-20630-kraken

kraken: tests: qa/tasks: rbd-mirror daemon not properly run in foreground mode

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

Merge pull request #16290 from smithfarm/wip-19759-kraken

kraken: rgw: multisite: after CreateBucket is forwarded to master, local bucket may use different value for bucket index shards

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

Merge pull request #16180 from smithfarm/wip-20347-kraken

kraken: rgw: meta sync thread crash at RGWMetaSyncShardCR

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

Merge pull request #16190 from smithfarm/wip-20026-kraken

kraken: mds: unresponsive when truncating a very large file

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

Merge pull request #15526 from badone/wip-async-sleep-timer-fix-kraken

kraken: osd: Implement asynchronous scrub sleep

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

Merge pull request #16137 from smithfarm/wip-20024-kraken

kraken: tests: HEALTH_WARN pool rbd pg_num 244 > pgp_num 224 during upgrade

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

Merge pull request #14734 from smithfarm/wip-19670-kraken

kraken: build/ops: logrotate is missing from debian package (kraken, master)

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

Merge pull request #16166 from smithfarm/wip-19840-kraken

kraken: rgw: civetweb frontend segfaults in Luminous

Reviewed-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>

qa/suites/upgrade/jewel-x: do not thrash cluster full during upgrade

The mon command varies.

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 39fdc53fe5f33678fbbd00cf8810b6d523d0040c)

qa/suites/upgrade/jewel-x/parallel: expand workload matrix

These should run independently against a racing upgrade.

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 1a0ad2b48839fea75d4d3339f7d198404ff1ac37)

qa/suites/upgrade/jewel-x/stress-split-erasure-code: box thrashosds

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit b2d354d56359d08b35cbea58f811c7bafb700d31)

qa/suites/upgrade/jewel-x/stress-split: finish client.0 upgrade too

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 56f9387736eed136b38c087a4805821063e9f8ab)

qa/suites/upgrade/jewel-x: remove kraken references

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 468285b65d6bef2d86c80ebdfecf0920294ca5cd)

qa/suite/upgrade/jewel-x: drop x86 ec test

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 1e0e53c80d8fa88970cee1d61aaa8514004547ac)

qa/suites/upgrade/jewel-x: fix upgrade order; fix split grouping

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 269eafb2027ea6dfbad730f6fb1481a1cabc5e60)

Conflicts:
    stress-split/6-kraken.yaml (do not delete)
    stress-split/6-luminous.yaml (do not add)
    stress-split-erasure-code/6-kraken.yaml (do not delete)
    stress-split-erasure-code/6-luminous.yaml (do not add)

qa/suites/upgrade/jewel-x/parallel: upgrade mons before osds

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 3c80e15c3b34ac2adc4e70f09929e3bc01785594)

qa/suites/upgrade/jewel-x/parallel: fix upgrade to luminous

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 2e3ea53b41cacc4b4bbc4ffe7342fb00233009cf)

Conflicts:
   4-kraken.yaml (do not delete)
   4-luminous.yaml (do not add)

Merge pull request #13871 from smithfarm/wip-19162-kraken

kraken: rgw: rgw_file: fix marker computation

Merge pull request #16133 from smithfarm/wip-18378-kraken

kraken: msg/simple/SimpleMessenger.cc: 239: FAILED assert(!cleared)

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

Merge pull request #16134 from smithfarm/wip-19340-kraken

kraken: An OSD was seen getting ENOSPC even with osd_failsafe_full_ratio passed

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

osdc/Filer: truncate large file party by party

Fixes: http://tracker.ceph.com/issues/19755
Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
(cherry picked from commit 5fab215e461e5ecc36c0f9d9ea867f6c45e80263)

Conflicts:
src/osdc/Filer.h
src/osdc/Filer.cc
src/mds/MDCache.cc

Merge pull request #16131 from smithfarm/wip-swift-kraken

tests: swift.py: clone the ceph-jewel branch

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

osd: Move scrub sleep timer to osdservice

PR 14886 erroneously creates a scrub sleep timer for every pg resulting
in a proliferation of threads. Move the timer to the osd service so
there can be only one.

Fixes: http://tracker.ceph.com/issues/19986
Signed-off-by: Brad Hubbard <bhubbard@redhat.com>
(cherry picked from commit f110a82437df79dc20207d296e8229fc0e9ce18b)

osd: Implement asynchronous scrub sleep

Rather than blocking the main op queue just do an async sleep.

Fixes: http://tracker.ceph.com/issues/19497
Signed-off-by: Brad Hubbard <bhubbard@redhat.com>
(cherry picked from commit 7af3e86c2e4992db35637864b83832535c94d0e6)

Merge pull request #16493 from smithfarm/wip-suppress-upgrade-fail

tests: run certain upgrade/jewel-x tests on Xenial only

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

Merge pull request #14597 from gregsfortytwo/wip-kraken-snaptrim

kraken: core: improve control and throttling of the snap trimmer

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

tests: upgrade/jewel-x/parallel: drop duplicate kraken.yaml

This yaml file has the effect of re-running "ceph osd set require_kraken_osds"
at the very end of the test. Drop it.

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

tests: run certain upgrade/jewel-x tests on Xenial only

This PR drops two upgrade/jewel-x test cases that are not compatible with
https://github.com/ceph/ceph/pull/14597

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

tests: swift.py: clone the ceph-jewel branch

The master branch of ceph/swift.git contains tests that are incompatible with
Jewel and Hammer. The ceph-jewel branch omits these tests.

Signed-off-by: Nathan Cutler <ncutler@suse.com>
(cherry picked from commit a86ce728954a765797ce634025d43650d990e480)

Conflicts:
    qa/tasks/swift.py: clone ceph-kraken branch instead of ceph-jewel

Merge pull request #16111 from smithfarm/wip-20497-kraken

kraken: tests: insufficient timeout in radosbench task

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

rgw: lease_stack: use reset method instead of assignment

It seems that the intent of 45877d38fd9a385b2f8b13e90be94d784898b0b3 was to
change all instances of "lease_stack = ..." to "lease_stack.reset(...)", but
this one was missed.

Signed-off-by: Nathan Cutler <ncutler@suse.com>
(cherry picked from commit fc425afeb2e2a6ba3c98c612b3977aea619c9f73)

Conflicts:
src/rgw/rgw_sync.cc (trivial)

Merge pull request #16178 from smithfarm/wip-20268-kraken

kraken: rgw: get wrong content when download object with specific range when compression was enabled

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

rpm: Fix undefined FIRST_ARG

If FIRST_ARG is undefined, the rpms will show an error on upgrade
because the condition in the postun script gets broken.

This was a regression introduced by commit for issue 20077 that moved
ceph-disk unit files to ceph-base.

Fixes: http://tracker.ceph.com/issues/20077
Signed-off-by: Boris Ranto <branto@redhat.com>
(cherry picked from commit 562816914ccca8e4e7d9c31f333db2f0da6f7c99)

selinux: Install ceph-base before ceph-selinux

We need to have ceph-base installed before ceph-selinux to use ceph-disk
in %post script. The default ordering is random and so the installation
randomly failed to relabel the files.

Fixes: http://tracker.ceph.com/issues/20184
Signed-off-by: Boris Ranto <branto@redhat.com>
(cherry picked from commit e69086645d3411a2ed781609b670eb5f16ac4810)
Conflicts:
ceph.spec.in: No _epoch_prefix in kraken.

Merge pull request #16114 from smithfarm/wip-20500-kraken

kraken: cephfs: src/test/pybind/test_cephfs.py fails

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

Merge pull request #16108 from smithfarm/wip-19763-kraken

kraken: cephfs: non-local quota changes not visible until some IO is done

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

Merge pull request #16107 from smithfarm/wip-19710-kraken

kraken: mds: enable daemon to start when session ino info is corrupt

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

Merge pull request #16106 from smithfarm/wip-19680-kraken

kraken: mds: damage reporting by ino number is useless

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

Merge pull request #16105 from smithfarm/wip-19678-kraken

kraken: cephfs: ceph-fuse does not recover after lost connection to MDS

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

Merge pull request #16104 from smithfarm/wip-19676-kraken

kraken: cephfs: Test failure: test_data_isolated (tasks.cephfs.test_volume_client.TestVolumeClient)

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

Merge pull request #16103 from smithfarm/wip-19674-kraken

kraken: cephfs: mds is crushed, after I set about 400 64KB xattr kv pairs to a file

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

Merge pull request #16102 from smithfarm/wip-19672-kraken

kraken: mds: assert fail when shutting down

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

Merge pull request #16101 from smithfarm/wip-19669-kraken

kraken: mds: daemon goes readonly writing backtrace for a file whose data pool has been removed

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

Merge pull request #16100 from smithfarm/wip-19667-kraken

kraken: cephfs: mount point break off problem after mds switch

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

Merge pull request #16099 from smithfarm/wip-19664-kraken

kraken: mds: C_MDSInternalNoop::complete doesn't free itself

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

Merge pull request #14998 from jan--f/wip-19845-kraken

kraken: cephfs: normalize file open flags internally used by cephfs

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

Merge pull request #16168 from smithfarm/wip-20031-kraken

kraken: rgw: Swift's at-root features (/crossdomain.xml, /info, /healthcheck) are broken

Reviewed-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>

Merge pull request #16164 from smithfarm/wip-19777-kraken

kraken: rgw: swift: disable revocation thread under certain circumstances

Reviewed-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>

Merge pull request #16191 from smithfarm/wip-20028-kraken

kraken: cephfs: Deadlock on two ceph-fuse clients accessing the same file

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

Merge pull request #16183 from smithfarm/wip-20405-kraken

kraken: rgw: Lifecycle thread will still handle the bucket even if it has been removed

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

Merge pull request #16181 from smithfarm/wip-20363-kraken

kraken: rgw: VersionIdMarker and NextVersionIdMarker are not returned when listing object versions

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

Merge pull request #16179 from smithfarm/wip-20269-kraken

kraken: rgw: wrong object size after copy of uncompressed multipart objects

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

Merge pull request #16174 from smithfarm/wip-20261-kraken

kraken: rgw: 'radosgw-admin usage show' listing 0 bytes_sent/received

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

Merge pull request #16173 from smithfarm/wip-20156-kraken

kraken: fix: rgw crashed caused by shard id out of range when listing data log

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

Merge pull request #16165 from smithfarm/wip-19839-kraken

kraken: rgw: reduce log level of 'storing entry at' in cls_log

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

Merge pull request #16163 from smithfarm/wip-19766-kraken

kraken: rgw: when uploading the objects continuesly in the versioned bucket, some objects will not sync

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

Merge pull request #16162 from smithfarm/wip-19725-kraken

kraken: rgw: S3 v4 authentication issue with X-Amz-Expires

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

Merge pull request #16161 from smithfarm/wip-19614-kraken

kraken: multisite: rest api fails to decode large period on 'period commit'

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

Merge pull request #16139 from smithfarm/wip-20147-kraken

kraken: rgw: 'gc list --include-all' command infinite loop the first 1000 items

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

Merge pull request #16096 from smithfarm/wip-19336-kraken

kraken: rbd: refuse to use an ec pool that doesn't support overwrites

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

Merge pull request #16097 from smithfarm/wip-19609-kraken

kraken: tests: [librados_test_stub] cls_cxx_map_get_XYZ methods don't return correct value

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

Merge pull request #16184 from smithfarm/wip-20154-kraken

kraken: rbd: Potential IO hang if image is flattened while read request is in-flight

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

Merge pull request #16187 from smithfarm/wip-20266-kraken

kraken: rbd: [api] is_exclusive_lock_owner shouldn't return -EBUSY

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

Merge pull request #16195 from dillaman/wip-20351-kraken

kraken: tests: test/librbd: decouple ceph_test_librbd_api from libceph-common

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

Merge pull request #15486 from dillaman/wip-20022-kraken

kraken: rbd-mirror: ensure missing images are re-synced when detected

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

Merge pull request #14540 from smithfarm/wip-18910-kraken

kraken: rbd-nbd: check /sys/block/nbdX/size to ensure kernel mapped correctly

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

Merge pull request #16140 from smithfarm/wip-20271-kraken

kraken: tests: LibRadosMiscConnectFailure.ConnectFailure hang

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

Merge pull request #16112 from smithfarm/wip-20499-kraken

kraken: tests: ObjectStore/StoreTest.OnodeSizeTracking/2 fails on bluestore

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

Merge pull request #14960 from yehudasa/wip-19704

kraken: civetweb: move to post 1.8 version

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

osd: do not default-abort on leaked pg refs

Signed-off-by: Greg Farnum <gfarnum@redhat.com>
(cherry picked from commit 4caf2df0c380a1281db9509b3feb342705512b58)

osd: shutdown our reserver_finisher earlier in the process

This finisher thread has a lot of callbacks which can hold PGRefs. Make
sure we drain them out before checking that all the PGs have finished
and have no outstanding references.

Moving this should be safe; we've already stopped the op thread et al
and the only things still running are the OSDService's objecter_finisher,
recovery_request_timer, and snap_sleep_timer (which has definitely been emptied
by the time we get here as it's synchronously cleared out on PG shutdown).

Signed-off-by: Greg Farnum <gfarnum@redhat.com>
(cherry picked from commit 66ea9c1f66ae31035e62bd4335b08948b1e8e5e2)

osd: Reset() the snaptrimmer on shutdown

We were failing to exit various wait states which held PGRefs. Error!

Fixes: http://tracker.ceph.com/issues/19931
Signed-off-by: Greg Farnum <gfarnum@redhat.com>
(cherry picked from commit b0e9deeea8a8e90f6d7e9d56b6b4aed890e01d7b)

osd: print out pg debug ref counts on acquire/drop

Signed-off-by: Greg Farnum <gfarnum@redhat.com>
(cherry picked from commit a3b028a6d1ba74ed975ebd665b1b50fb7e5039a4)

mon: Fix status output warning for mon_warn_osd_usage_min_max_delta

Fixes: http://tracker.ceph.com/issues/20544
Caued by: 489e810c37ed6fb9d32d1015634041a577501ee4

Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit 56f9808016ab289bdc0ce7cfbb0503c78b509593)

Conflicts:
    src/mon/PGMap.cc:
        kraken has no daa0793c393a21bd2dd2ec52a0efd181e1032400 so we manually
        apply the changes from 56f9808016ab289bdc0ce7cfbb0503c78b509593 to
        src/mon/PGMonitor.cc instead of to src/mon/PGMap.cc

mon/PGMonitor: clean up min/max span warning

Clean up option naming.

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 489e810c37ed6fb9d32d1015634041a577501ee4)

bluestore: Fix BlueStore::statfs available amount to not include bluefs min reserved

This fixes OSD crashes because checking osd_failsafe_full_ratio won't work
without accurate statfs information.

Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit 72d83f848a35e8831d66e8529c4e26f51e845da6)

osd: Round fullness in message to correspond to df -h

This really only works after journal drains because
we adjust for the journal.

Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit 26dcb591f9af01ed444aa758c3d601e7f67261b2)

filestore: Account for dirty journal data in statfs

Fixes: http://tracker.ceph.com/issues/16878
Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit 78adb70c21c6b8e6a9191b76917919b125a9490f)

mon: Add warning if diff in OSD usage > config mon_warn_osd_usage_percent (10%)

Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit c8004e6558359fb542e45bb4b483a6c91afdc0b4)

mon: Bump min in ratio to 75%

Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit 830cc7aa7be1ccd8f54f056b6a58e923cadd1c2b)

osd: Fix ENOSPC crash message text

Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit 50cfe03fcba253c8380b21043ed03879134d6836)

Merge pull request #16143 from smithfarm/wip-20365-kraken

kraken: mon: osd crush set crushmap need sanity check

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

Merge pull request #16138 from smithfarm/wip-20034-kraken

kraken: ceph-disk: Racing between partition creation & device node creation

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

Merge pull request #16135 from smithfarm/wip-20010-kraken

kraken: ceph-disk: separate ceph-osd --check-needs-* logs

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

qa/tasks: rbd-mirror daemon not properly run in foreground mode

Fixes: http://tracker.ceph.com/issues/20630
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 4fa1918717b25a5ffafbf649eedcfe7d5ab829c2)

rbd: do not attempt to load key if auth is disabled

Fixes: http://tracker.ceph.com/issues/19035
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 8b9c8df6d7f0b75c5451953bb322bc1f9afb6299)

rgw: when create_bucket use the same num_shards with info.num_shards

pr #14388 only fix the num_shards in BucketInfo, "init_bucket_index" function still use local num_shards

Fixes: http://tracker.ceph.com/issues/19745
Signed-off-by: Shasha Lu <lu.shasha@eisoo.com>
(cherry picked from commit 4ce64a190b4ff36985e785e574c077d39796feea)

Conflicts:
src/rgw/rgw_rados.cc - init_bucket_index() called earlier

rgw: using the same bucket num_shards as master zg when create bucket in secondary zg

create bucket in secondary zonegroup will forward to master. The master may have different num_shards option.
So when create bucket in local, should use master's num_shards instead of local num_shards option.

Fixes: http://tracker.ceph.com/issues/19745
Signed-off-by: Shasha Lu <lu.shasha@eisoo.com>
(cherry picked from commit a34c4b8fb13dd5590eb3c6ecb5e55207ed8e3ee8)

Conflicts:
    src/rgw/rgw_op.cc - no RGWBulkUploadOp:: methods in kraken; modifications
        to RGWBulkUploadOp::handle_dir() omitted

rgw: add a field to store generic user data in the bucket index,
that can be populated/fetched via a configurable custom http header

Signed-off-by: Pavan Rallabhandi <prallabhandi@walmartlabs.com>
(cherry picked from commit abca7a86c3cfbb58fafb5d057d9d6f5017a53704)

Conflicts:
src/rgw/rgw_op.cc
          Signature fixes for RGWPutObjProcessor_Multipart::do_complete().

src/rgw/rgw_op.h
         A new member field `crypt_http_responses` is added in the class `RGWPutObj` in master version, which is not required in Kraken.

src/rgw/rgw_rados.cc
         In RGWRados::Bucket::UpdateIndex::complete, RGWObjEnt has been removed in master, which has to be retained in Kraken.
         In RGWRados::cls_obj_complete_op, user_data is added to the rgw_bucket_dir_entry_meta
         In RGWRados::cls_bucket_list, the user_data field of RGWObjEnt is populated.

src/rgw/rgw_rados.h
         In UpdateIndex::complete(), remove_objs is of type rgw_obj_key in Kraken instead of rgw_obj_index_key
         RGWPutObjProcessor_Multipart is not part of this file in Kraken.

src/rgw/rgw_rest_swift.cc
In RGWListBucket_ObjStore_SWIFT::send_response(), there is no meta struct in Kraken.

        src/rgw/rgw_common.h
         Add user_data field in RGWObjEnt structure.

        src/rgw/rgw_json_enc.cc
         Add user_data field while dumping RGWObjEnt.

osd: repair the PG_DEBUG_REFS build

Signed-off-by: Greg Farnum <gfarnum@redhat.com>
(cherry picked from commit ec4185d1980b48b01687a177248d0894f326dc37)

Conflicts:
src/osd/OSD.cc
src/osd/Session.h

osd: use PGRef instead of boost::intrusive_ptr<PG> everywhere

Signed-off-by: Greg Farnum <gfarnum@redhat.com>
(cherry picked from commit 306ad85f0fb4597ba3bb3c04a67abdaba556ba74)