common/hobject: rename get_filestore_key* -> get_nibblewise_key*

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

rbd: fix bench-write

Make each IO get a unique offset!

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

ceph-kvstore-tool: take a db type too

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

common/hobject: make hobject_t::set_key avoid dups

If the key matches the oid name, clear it.

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

test_objectstore_memstore.sh on make check

This'll give us some coverage from ceph_test_objectstore, albeit
on the memstore backend.  Upside is that this takes < 10s.

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

ceph_test_objectstore: test many small writes

This exercises the overlay code better.

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

ceph_test_objectstore: strengthen clone test w/ new assert

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

ceph_test_objectstore: test omap clone

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

ceph_test_objectstore: much cleanup, new tests

These changes are necessary to make newstore (which is more picky
about unique collection names and objects that land *inside* the
correct collection) pass.  There are also several new tests.

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

os/fs: fix open_handle when name_to_handle not present

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

os/fs: add zero / hole punch support

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

os/fs: add simple FS abstraction layer

- open by handle
- xfs allocation hint

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

os/ObjectStore: drop sync(Context *onsync)

No users.

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

os/ObjectStore: drop sync()

No callers.

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

os/ObjectStore: drop flush()

No more callers.

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

os/ObjectStore: drop sync_and_flush()

The semantics of this call were FileStore-specific; and there are no more
callers.

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

osd: improve osd bench

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

buffer: add prepare_iov() method

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

osd: drop dead suicide() code

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

ceph-objectstore-tool: drop explicit sync_and_flush() calls

We implicitly sync on shutdown.

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

ceph_test_objectstore: drop useless sync_and_flush

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

osd: sync (no flush) on osd suicide

Not that we even need to sync here!

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

osd: drop explicit sync/flush calls before umount

Do it implicity on umount.

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

osd: fix flush_journal command

We need to sync_and_flush *and* sync() again in order to retire any items
in the FileStore journal.

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

osd: explicitly use a 'meta' Sequencer

Use an explicit Sequencer for all transactions.

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

osd: use osr for scan_snaps

Updates should be ordered with respect to PG metadata.  Not that it matters
that much since we are doing cleanup/repair work here.

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

os/ObjectStore: do not let Sequencer::flush_commit delete the context

Let the caller manage the lifecycle; this allows us to potentially put it
on the stack, for instance.

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

os/ObjectStore: ref count Sequencer_impl

This allows the implementation to have a lifecycle that extends beyond
the Sequencer interface object we expose to the user.

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

os/KeyValueDB: return -ENOENT if key doesn't exist

Previous code return 0 in all cases, caller doesn't
know if the key is exists in the db.

Signed-off-by: Xiaoxi Chen <xiaoxi.chen@intel.com>

os/ObjectStore: pass bits to create_collection

This informs the backend how many bits of the hash range will be covered
by the collection.  Those bits are implicitly matched against the spg_t in
coll_t.

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

osd/osd_types: pg_t: add contains() helper to check objects is in a pg

This calls the ghobject_t method that does the same, along with ps().

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

os/ObjectStore: return const refs for oid, cid

Should be a bit faster.

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

common/WorkQueue: add wait()

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

vstart.sh: allow rocksdb

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

os/RocksDbStore: init pointers in ctor

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

osd/osd_types: make coll_t operator== less picky

For meta coll's do not compare pg.

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

os/KeyValueDB: add sipler get() variant

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

common/Formatter: add dump_object helper

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

os/FileStore: clear cached fd for renamed-over object

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

Merge pull request #5470 from ceph/wip-newstore-sort

osd: change sort order for hobject_t

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

Merge pull request #5424 from lebauce/fix_swift_user_command_line

doc: fix command line for swift user creation

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

Merge pull request #5048 from SUSE/wip-objecter-comment-cleanup

Objecter.cc: fix name of _take_op_budget() function in comment

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

Merge pull request #5606 from liewegas/wip-scrub-errormsg

osd/ReplicatedPG: fix missing whitespace in scrub error message

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

Merge pull request #5604 from oritwas/wip-cmake-fix-rbd

cmake: add DiffIterate.cc to librbd

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

moved to use boost uuid implementation, based on commit 4fe89a7b14c97b2ed7f357132901beb2bdcec551

Signed-off-by: Rohan Mars <code@rohanmars.com>
Reviewed-by: Casey Bodley <cbodley@redhat.com>

Merge pull request #5433 from zqkkqz/master

common/Thread: pthread_attr_destroy(thread_attr) when done with it

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

osd/ReplicatedPG: fix missing whitespace in scrub error message

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

Merge pull request #5538 from ceph/wip-selinux-improve-update

selinux: Relabel files if and only if the policy version changed

Reviewed-by: Milan Broz <mbroz@redhat.com>

ceph.spec.in: Require awk for selinux post script

Signed-off-by: Boris Ranto <branto@redhat.com>

selinux: Relabel files if and only if the policy version changed

Currently, the ceph files are being relabelled every time the package is
rebuilt. Fix this by checking the policy versions and relabel the files
only if the policy actually changed (different policy version was
detected).

Signed-off-by: Boris Ranto <branto@redhat.com>

Merge pull request #5596 from mbroz/wip-selinux-systemd

Update ceph-selinux rpm spec

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

cmake: add DiffIterate.cc to librbd

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

Merge pull request #5600 from liewegas/wip-osd-bind

doc: ceph-osd binds to up to 4 ports, not 3

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

doc: ceph-osd binds to up to 4 ports, not 3

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

Merge remote-tracking branch 'gh/next'

Merge pull request #5597 from dachary/wip-12718-restorecon

ceph-disk: only call restorecon when available

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

ceph-disk: only call restorecon when available

9db80da12803d42bb676d67f37442c0c54d83448 added an unconditional call to
restorecon after mounting the filesystem. It fails when restorecon is
not available and must be made conditional.

http://tracker.ceph.com/issues/12718 Fixes: #12718

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

Do not use hardcoded paths in spec post section for selinux commands.

Signed-off-by: Milan Broz <mbroz@redhat.com>

Add dependence to selinux-policy-base to fix installation in kickstart.

Signed-off-by: Milan Broz <mbroz@redhat.com>

Update rpm spec to properly restart systemd ceph.target.

If we are using systemd, ceph target should be used.
Also fail to restart service should not stop policy to load.

Signed-off-by: Milan Broz <mbroz@redhat.com>

Merge pull request #5595 from ceph/wip-12710

qa/fsstress.sh: fix 'cp not writing through dangling symlink'

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

Merge pull request #5594 from ceph/wip-12711

mds: properly set client incarnation

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

qa/fsstress.sh: fix 'cp not writing through dangling symlink'

On some test machines, /usr/lib/ltp/testcases/bin/fsstress is
dangling symlink. 'cp -f' is impotent in this case.

Fixes: #12710
Signed-off-by: Yan, Zheng <zyan@redhat.com>

mds: properly set client incarnation

Fixes: #12711
Signed-off-by: Yan, Zheng <zyan@redhat.com>

Merge pull request #5593 from ceph/revert-4927-snapset-obc

Revert "osd/ReplicatedPG: snapset is not persisted"

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

Revert "osd/ReplicatedPG: snapset is not persisted"

Merge pull request #5577 from oritwas/wip-next-12363

rgw: we should not overide Swift sent content type

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

Merge pull request #5558 from s09816/rbd-fix

rbd:improve the error handle of rbd,check the return value.

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

osd/OSDMap: test_flag returns bool

Otherwise comparing the result to a bool will promote the bool to
an int and fail the comparison.

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

osdc/Objecter: restart listing this PG if sort order changes

If the cluster sort order changes mid-way through our listing, our
cursor within this pg is meaningless and we need to restart at the
beginning of the PG.

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

osd/ReplicatedPG: fix missing set sort order on [N]PGLS

This is the only thing that cares about the missing set sort order,
so only adjust it here as needed.

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

osd/osd_types: add pg_missing_t::resort() method

Set/change sort order as needed.

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

osd/osd_types: make pg_missing_t sort order dynamic

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

osd: refuse to boot if SORTBITWISE not set but backend cannot sort nibblewise

This prevenents OSDs with new backends from starting if the
cluster is still using the legacy sort order.

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

Merge pull request #5586 from ceph/wip-jni-loader

wip-jni-loader

Signed-off-by: Noah Watkins <noahwatkins@gmail.com>

Merge pull request #5588 from zhouyuan/isal_yasm_fix

configure: Fix checking for yasm compatibility

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

configure: Fix checking for yasm compability

Fix typo when checking yasm

Signed-off-by: Yuan Zhou <yuan.zhou@intel.com>

java: add libcommon to deps

Why do indirect dependencies seem to work randomly:

  undefined symbol: _ZN4ceph3log3Log12create_entryEii

Signed-off-by: Noah Watkins <noahwatkins@gmail.com>

java: search for JNI bits in common dirs

Signed-off-by: Noah Watkins <noahwatkins@gmail.com>

Merge pull request #5572 from xinxinsh/wip-rados-bench-error

fix print error of rados bench

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

rbd:improve the error handle of rbd,check the return value.

Signed-off-by: s09816 <shi.lu@h3c.com>

Merge pull request #5443 from ceph/wip-wrlock

cleanup: remove all traces of rados 'lock' operations

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

Merge remote-tracking branch 'origin/next'

Merge pull request #5537 from s09816/master

rbd:modify the log of purging snaps so that it is more appropriate

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

Merge pull request #5560 from solesoul1127/master

rbd:Check the dest image name, if it is empty string, return error and give a message

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

fix print error of rados bench

Total time run:         12.279167
Total writes made:      92
Write size:             4194304
Bandwidth (MB/sec):     30
Stddev Bandwidth:       23.4
Max bandwidth (MB/sec): 64
Min bandwidth (MB/sec): 2
Average IOPS:           7
Stddev IOPS:            6
Max IOPS:               32767
Min IOPS:               -1537890352
Average Latency:        2.12
Stddev Latency:         1.35
Max latency:            6.05
Min latency:            0.501

Signed-off-by: xinxin shu <xinxin.shu@intel.com>

Merge pull request #5569 from ceph/wip-unused-var

client: fix unused var warning

rbd:modify the log of purging snaps so that it is more appropriate.

Signed-off-by: s09816 <shi.lu@h3c.com>

Merge pull request #4744 from ceph/wip-11625

librbd: diff_iterate should issue concurrent ops

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

Merge pull request #5555 from dachary/wip-mon-test-timeouts

tests: be more generous with mon tests timeouts

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

Merge pull request #5576 from liewegas/wip-dencoder

simplify handling for objects w/ nondeterministic encoding

Reviewed-by: John Spray <john.spray@redhat.com>

rgw: we should not overide Swift sent content type

Fixes: #12363
backport: hammer

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

test/encoding/check-generated: test sorted json dumps for nondeterministic objects

Nondeterministic objects dump nondeterministically (usually due
to unordered_map or _set).  Compare their sorted output.  This
is a weaker test but is better than nothing.

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

Revert "osd/HitSet: make subclasses dump deterministically"

This reverts commit 34c048b3cb2b16658e692ac5110115612031510d.

Revert "mon/PGMap: dump osd_epochs in deterministic order"

This reverts commit 3df6438cd2c9a77a06b7a8d011bafbed00f1fc5e.

Merge pull request #4864 from stevecapperlinaro/for-ceph/journal-page-size-fix

FileJournal: Remove CEPH_PAGE_SIZE assumptions

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

Merge pull request #5100 from stevecapperlinaro/for-ceph/tweak-buffer-append-size

Common: Do not use CEPH_PAGE_SIZE when appending buffers in Ceph

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

Merge pull request #5339 from xinxinsh/wip-use-throttle

use throttle framework to throttle objectstore ops/bytes

Reviewed-by: Li Wang <liwang@ubuntukylin.com>
Reviewed-by: Sage Weil <sage@redhat.com>

crypto: fix unbalanced ceph::crypto::init/ceph::crypto:shutdown

we may create a CephContext without calling common_init_finish(), then
delete the CephContext. In this case, ceph::crypto:init() is not called,
so CephContext::~CephContext() should not call ceph::crypto::shutdown().

Fixes: #12598
Signed-off-by: Yan, Zheng <zyan@redhat.com>
Reviewed-by: John Spray <john.spray@redhat.com>

Merge pull request #5573 from SUSE/wip-restart-on-rpm-removal-only

ceph.spec.in: test %preun argument is zero for removal-only operations