ceph.spec.in: package .pyc files on el7

* package .pyc files on el7
  per
  https://docs.fedoraproject.org/en-US/packaging-guidelines/Python/#_byte_compiling

  it's required to package .pyc files

* use python3 to bytecompile .py files on el7

  per
  https://fedoraproject.org/wiki/Changes/No_more_automagic_Python_bytecompilation#Status_quo,
  `brp-python-bytecompile` script bytecompile .py files automatically, the
  .py files under interpreter-specific directories, such as `/usr/lib/python3.6/`,
  are compiled using the appropriate compiler. but the mgr modules are
  located under `/usr/share/ceph/mgr/`. they are compiled using
  `${__python}`, which is still python2 in RHEL/CentOS 7. since we've
  dropped the support of python2 in octopus. we should either stop
  bytecompiling the .py source files or start using python3 to compile
  them. otherwise python2 will fail to compile source files using python3
  specific syntax like:

```
Compiling /home/jenkins-build/build/workspace/ceph-dev-new-build/ARCH/x86_64/AVAILABLE_ARCH/x86_64/AVAILABLE_DIST/centos7/DIST/centos7/MACHINE_SIZE/huge/release/15.0.0-9242-g400d98b/rpm/el7/BUILDROOT/ceph-15.0.0-9242.g400d98b.el7.x86_64/usr/share/ceph/mgr/telemetry/module.py
...
  File "/usr/share/ceph/mgr/telemetry/module.py", line 387
    anon_devid = f"{devid.rsplit('_', 1)[0]}_{uuid.uuid1()}"
                                                           ^
SyntaxError: invalid syntax

error: Bad exit status from /var/tmp/rpm-tmp.4oRsox (%install)
```

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

Merge pull request #32577 from xenolinux/new

doc/releases/nautilus.rst: remove a redundant '*'

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

doc/releases/nautilus.rst: remove a redundant '*'

remove an extra asterisk '*'

Signed-off-by: Servesha Dudhgaonkar <sdudhgao@redhat.com>

Merge PR #32521 into master

* refs/pull/32521/head:
mds: clarify message about blocked scrub
mds: throttle scrub start for multiple active MDS

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

Merge PR #32213 into master

* refs/pull/32213/head:
cephfs-shell: don't catch libcephfs.Error unnecessarily
cephfs-shell: make every command set a return value on failure
pybind/cephfs: move LibCephFSStateError closer to base class
pybind/cephfs: add method to get error code in Error and OSError

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

mds: clarify message about blocked scrub

Also, non-recursive scrub still needs rejected: a given directory's
fragments may be on other ranks.

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

Merge PR #32201 into master

* refs/pull/32201/head:
mds: maintain client provided metric flags in client metadata
mds: add feature_bitset_t::dump()

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

Merge PR #32320 into master

* refs/pull/32320/head:
mds: Reorganize class members in SessionMap header

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

Merge PR #32548 into master

* refs/pull/32548/head:
libcephfs: fix documentation for ceph_seekdir offset
client: propagate -ENOTCONN errors for ll_releasedir

Reviewed-by: Jeff Layton <jlayton@redhat.com>

Merge pull request #32506 from dzafman/wip-deprecated-full

osd: Again remove deprecated full/nearfull from osdmap

Reviewed-by: Neha Ojha <nojha@redhat.com>

Merge pull request #32591 from CypressXt/patch-1

Documentation: Centos ceph-deploy's python dependencies

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

Merge pull request #32514 from majianpeng/osd-remove-osmap_lock_name

osd/OSD: remove unused parameter osdmap_lock_name.

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

Merge pull request #32588 from tchaikov/wip-qa-mgr-setup-with-force

qa/tasks/mgr: set mgr module option with --force

Reviewed-by: Laura Paduano <lpaduano@suse.com>

Merge pull request #32590 from tchaikov/wip-doc-ceph-bluestore-tools

doc/man: improve bluefs-bdev-expand option

Reviewed-by: Zac Dover <zac.dover@gmail.com>

Merge pull request #32538 from liuchang0812/kstore-bug-39665

os/kstore: do not cache in-fight stripes on read ops to avoid leaks

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

Merge pull request #32563 from majianpeng/bluefs-remove-flush-log

os/bluestore/BlueFS: replace flush_log with sync_metadata.

Reviewed-by: Igor Fedotov <ifedotov@suse.com>

Merge pull request #32547 from aclamk/admin-socket-doxygen

common/admin_socket: Add doxygen for call and call_async

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

Merge pull request #32519 from tchaikov/wip-qa-skip-if-no-pool

qa/tasks/ceph_manager: do not pick a pool is there is no pools

Reviewed-By: Neha Ojha <nojha@redhat.com>

Merge pull request #32549 from tchaikov/wip-qa-rados-whitelist-health-warning

qa: whitelist FS_DEGRADED

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

Merge pull request #32512 from tchaikov/wip-cbt-py3

Revert "qa/tasks/cbt: include py2 deps on ubuntu for now"

Reviewed-By: Neha Ojha <nojha@redhat.com>

Added fresh Centos install's python dependencies

Fixes: https://tracker.ceph.com/issues/43545
Signed-off-by: Clément Hampaï <clement.hampai@cypressxt.net>

Merge pull request #32500 from sebastian-philipp/test_orchestrator-mypy

pybind/mgr: Add test_orchestrator to mypy

Reviewed-by: Thomas Bechtold <tbechtold@suse.com>

doc/man: improve bluefs-bdev-expand option

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

Merge pull request #32525 from BenoitKnecht/dereference-lvm-path-symlink

ceph-volume: Dereference symlink in lvm list

common/admin_socket: Add doxygen for call and call_async

This is only doxygen, an attempt to solidify interface contract.

Signed-off-by: Adam Kupczyk <akupczyk@redhat.com>

osd/OSD: remove unused parameter osdmap_lock_name.

Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>

mds: throttle scrub start for multiple active MDS

* add check to "scrub start" command handler to dishonor request if
  multiple MDS are "active" for a recursive scrub request
* add check to MDSRank::handle_mds_map() to see if scrubbing is in
  progress and max_mds is greater than 1, if so then scrubbing is
  aborted. This is supposed to take care of the condition when scrubbing
  has been started with max_mds == 1 and then bumped to a higher value

Fixes: https://tracker.ceph.com/issues/43483
Signed-off-by: Milind Changire <mchangir@redhat.com>

qa/tasks/mgr: set mgr module option with --force

if mgr is not active, monitor will refuse to set any option consumed by
mgr modules.

the reason the tests pass somtimes is that, we have a racing here:

1. stop all mgr daemons
2. MgrMonitor gets updated and updates its mgr_module_options
accordingly.
3. in TestDashboard.setUp(), we reset the port number for dashboard
using "ceph config set mgr mgr/dashboard/y/ssl_server_port 7789"
4. restart all mgr daemons

but the 2nd step and 3rd step could race with each other, if the 2nd
step happens after 3rd step, the test passes. otherwise it fails.

in this change, "--force" is passed to the "ceph config set" command,
so ConfigMonitor can bypass the sanity test for the option, and just
set this option.

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

Merge pull request #32571 from tchaikov/wip-qa-test-runner-run-testcase

qa/tasks/cephfs_test_runner: setattr to class not instance

Reviewed-by: Laura Paduano <lpaduano@suse.com>

Merge PR #32532 into master

* refs/pull/32532/head:
qa: only restart MDS between tests

Reviewed-by: Jos Collin <jcollin@redhat.com>

Merge pull request #31834 from yuvalif/wip-yuval-add-kafka-security2

rgw/kafka: add ssl+sasl security to kafka

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

Merge pull request #32462 from alimaredia/wip-s3-tests-branch-override-remove

qa: remove force-branch from overrides of s3-tests

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

Merge PR #32533 into master

* refs/pull/32533/head:
qa: fix various py3 cephfs qa bugs x2

Reviewed-by: Kefu Chai <kchai@redhat.com>
Reviewed-by: Thomas Bechtold <tbechtold@suse.com>

Merge PR #32507 into master

* refs/pull/32507/head:
qa: detect RHEL8 for yum package installation

Reviewed-by: Rishabh Dave <ridave@redhat.com>

Merge pull request #30257 from ligangbin117/wip-ligangbin-rgw-data-error

rgw:[bug] fix data consistency error casued by rgw sent timeout

Reviewed-by: Yuli Yang <yuliyang@cmss.chinamobile.com>
Reviewed-by: Casey Bodley <cbodley@redhat.com>

Merge pull request #31346 from zhangsw/fix-rgw-orphan-error-handle

rgw: build_linked_oids_for_bucket and build_buckets_instance_index should return negative value if it fails

Reviewed-By: Casey Bodley <cbodley@redhat.com>
Reviewed-By: Abhishek Lekshmanan <abhishek@suse.com>

Merge pull request #32575 from tchaikov/wip-43227

qa/tasks/mgr/dashboard: set pg_num to 16

Reviewed-by: Volker Theile <vtheile@suse.com>
Reviewed-by: Sebastian Wagner <swagner@suse.com>
Reviewed-by: Tatjana Dehler <tdehler@suse.com>
Reviewed-by: Laura Paduano <lpaduano@suse.com>

Merge pull request #32572 from varshar16/wip-admin-build-doxgen-typo

admin/build-doc: Fix doxygen typo

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

qa/tasks/mgr/dashboard: set pg_num to 16

as the default pg_num is now 16 since 78bf9244, and the minimum pg_num
is also 16 in autoscaler since 78bf9244.

Fixes: https://tracker.ceph.com/issues/43227
Signed-off-by: Kefu Chai <kchai@redhat.com>

qa/tasks/cephfs_test_runner: setattr to class not instance

before this change, `setattr()` sets the instance specialized with a certain method
of test case, so in `MgrTestCase.setUpClass()`

assert cls.mgr_cluster is not None

fails,

after this change, instead of test case, the class of test suite is updated with the
specified params, even if we pass a certain test to test runner.
so we can

./run-backend-api-tests.sh tasks.mgr.test_dashboard.TestDashboard.test_standby

now. before this change, we can only:

./run-backend-api-tests.sh tasks.mgr.test_dashboard.TestDashboard

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

Merge pull request #32234 from votdev/better_orch_exception

mgr/dashboard: Throw a more meaningful exception

Reviewed-by: Sebastian Wagner <swagner@suse.com>
Reviewed-by: Tatjana Dehler <tdehler@suse.com>
Reviewed-by: Tiago Melo <tmelo@suse.com>

admin/build-doc: Fix doxygen typo

Signed-off-by: Varsha Rao <varao@redhat.com>

cephfs-shell: don't catch libcephfs.Error unnecessarily

Signed-off-by: Rishabh Dave <ridave@redhat.com>

cephfs-shell: make every command set a return value on failure

Fixes: https://tracker.ceph.com/issues/43249
Signed-off-by: Rishabh Dave <ridave@redhat.com>

pybind/cephfs: move LibCephFSStateError closer to base class

At first look it appears that class OSError is the only class derived
from class Error which makes class Error look redundant. Therefore, it's
better to move class LibCephFSStateError from the bunch of exception
classes derived from OSError to closer to class Error.

Signed-off-by: Rishabh Dave <ridave@redhat.com>

pybind/cephfs: add method to get error code in Error and OSError

Add method to class Error and override it class OSError.

Signed-off-by: Rishabh Dave <ridave@redhat.com>

Merge pull request #32551 from theanalyst/doc/releases/14.2.6

doc: release notes for v14.2.6

Reviewed-By: Josh Durgin <jdurgin@redhat.com>
Reviewed-By: Neha Ojha <nojha@redhat.com>

Merge pull request #32564 from tchaikov/wip-cmake-radosgwd

cmake: let vstart depend on radosgwd

Reviewed-by: Laura Paduano <lpaduano@suse.com>

ceph-volume: Dereference symlink in lvm list

This allows for a symlink to be passed to

```
ceph-volume lvm list <path>
```

which makes it possible to use `/dev/disk/by-path/*` devices, for
instance.

Fixes: https://tracker.ceph.com/issues/43497
Signed-off-by: Benoît Knecht <bknecht@protonmail.ch>

cmake: let vstart depend on radosgwd

in f528f173, in cmake, the target of executable "radosgw" is renamed
to "radosgwd",  and the static library of "radosgw_a" was renamed to
"radosgw". this broke the tests which expected radosgw to be available
if "tests" was built.

in this change, both "vstart" and "tests" now depend on "radosgwd"
instead of "radosgw".

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

os/bluestore/BlueFS: replace flush_log with sync_metadata.

Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>

qa/tasks/ceph_manager: do not pick a pool is there is no pools

random.choice(seq) raises IndexError if seq is empty. we cannot ensure
there is always one or more pools in the cluster while using pool
related thrasher. so skip the thrasher action if there is no pools at
that moment.

Fixes: https://tracker.ceph.com/issues/43412
Signed-off-by: Kefu Chai <kchai@redhat.com>

doc: update release table with 14.2.6

Signed-off-by: Abhishek Lekshmanan <abhishek@suse.com>

doc: releases: release notes for v14.2.6 nautilus

Signed-off-by: Abhishek Lekshmanan <abhishek@suse.com>

Merge PR #32508 into master

* refs/pull/32508/head:
doc: remove reference to obsolete scrub command

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

Merge PR #32535 into master

* refs/pull/32535/head:
qa: define centos version for fs:verify

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

qa: whitelist FS_DEGRADED

`admin_socket_output --all` sends "respawn" to mds, so when the mds
restarts, FS_DEGRADED.

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

Merge pull request #32517 from tchaikov/wip-mon-tell-command

mon: should not take non-tell commands as tell ones

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

Merge pull request #32539 from tchaikov/wip-debian-librgw

debian/radosgw.install: correct path to libradosgw.so*

Reviewed-by: Kaleb S. KEITHLEY <kkeithle@redhat.com>

libcephfs: fix documentation for ceph_seekdir offset

The offset should be a value returned by telldir, not seekdir.

Signed-off-by: David Disseldorp <ddiss@suse.de>

Merge pull request #32331 from smithfarm/wip-spec-drop-vim-header

rpm: drop vim-specific header

Reviewed-by: Ken Dreyer <kdreyer@redhat.com>
Reviewed-by: Jan Fajerski <jfajerski@suse.com>
Reviewed-by: Mykola Golub <mgolub@suse.com>

rgw/pubsub: add ssl+sasl security to kafka

Signed-off-by: Yuval Lifshitz <yuvalif@yahoo.com>

client: propagate -ENOTCONN errors for ll_releasedir

When initially added via e2dc1b6bf284, the unconditional
ceph_ll_releasedir() zero-return doesn't appear to have had any effect,
as Client::ll_releasedir() always returned zero.
efca340596ef added -ENOTCONN errors when Client::ll_releasedir() is
called during unmount. Ensure that this error is propagated.

Signed-off-by: David Disseldorp <ddiss@suse.de>

Merge pull request #32461 from s0nea/wip-dashboard-fix-user-test

mgr/dashboard: fix failing user test

Reviewed-by: Volker Theile <vtheile@suse.com>
Reviewed-by: Patrick Seidensal <pseidensal@suse.com>
Reviewed-by: Lenz Grimmer <lgrimmer@suse.com>
Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge PR #32537 into master

* refs/pull/32537/head:
Switch spelling of utilization

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

debian/radosgw.install: correct path to libradosgw.so*

it was a regression introduced by f528f173e4

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

qa: only restart MDS between tests

"fs fail" will only fail the MDS that are part of the file system which
will generally allow us to avoid spurious MDS_INSUFFICIENT_STANDBY
warnings. Further, only restart the MDS, there's no reason to leave them
offline.

Fixes: https://tracker.ceph.com/issues/43514
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>

os/kstore: do not cache in-fight stripes on read ops to avoid memory leaks

Fixes: http://tracker.ceph.com/issues/39665
Signed-off-by: Chang Liu <liuchang0812@gmail.com>
Signed-off-by: lvshanchun lvshanchun@gmail.com

Switch spelling of utilization

Prefer the non-British spelling of utilization since that's what the majority
of the code base seems to use.

Signed-off-by: Bryan Stillwell <bstillwell@godaddy.com>

doc: remove reference to obsolete scrub command

Fixes: e9a5ce0897efc6126caeebea9900bf05ec3d2174
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>

Merge PR #30802 into master

* refs/pull/30802/head:
cephfs-shell: Modify Quota Test
cephfs-shell: Fix permission errors on quota tests
cephfs-shell: Add tests for quota
cephfs-shell: Print max_bytes and max_files as string on get Quota
cephfs-shell: Remove extra length argument passed to setxattr()

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

Merge PR #32347 into master

* refs/pull/32347/head:
cephfs-shell: set shell.exit_code to 1 when reading conf fails
cephfs-shell: cephfs-shell breaks when conf_get() returns None
cephfs-shell: use libcephfs for exceptions
cephfs-shell: fix incorrect use of str.strip

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

Merge PR #32459 into master

* refs/pull/32459/head:
qa/tasks: Nothing to clean up if the volume was not created
mgr/volumes: drop obsolete comment in _cmd_fs_volume_create
mgr/volumes: cleanup on fs create error

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

Merge PR #32318 into master

* refs/pull/32318/head:
mds: don't add metadata to session close message

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

Merge PR #31250 into master

* refs/pull/31250/head:
mds: Reorganize class members in MDSTableServer header

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

Merge PR #31481 into master

* refs/pull/31481/head:
mds: Reorganize struct members in Mutation header

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

Merge PR #32452 into master

* refs/pull/32452/head:
cephfs-journal-tool: fix the usage
cephfs-journal-tool: fix crash with empty argv

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

osd: Again remove deprecated full/nearfull from osdmap

Fixes: https://tracker.ceph.com/issues/43485
Caused by: dff411f1905cc69bfb2cfa8b62a00b4702e6aa46

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

qa: detect RHEL8 for yum package installation

Fixes: https://tracker.ceph.com/issues/43487
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>

Merge PR #32510 into master

* refs/pull/32510/head:
mon: use non-obsolete mon scrub cmd

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

qa: define centos version for fs:verify

Otherwise it uses the teuthology default of 7.6.

Fixes: https://tracker.ceph.com/issues/43516
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>

qa: fix various py3 cephfs qa bugs x2

Fixes: https://tracker.ceph.com/issues/43515
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>

Merge pull request #32247 from dzafman/wip-43307

Remove use of rules batching for upmap balancer and default for upmap_max_deviation to 5

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

Merge pull request #32412 from tchaikov/wip-spec-more-deps-for-el8

ceph.spec.in: re-enable "make check" deps for el8

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

Merge pull request #32483 from trociny/wip-pool_replayer-print_status

rbd-mirror: fix "pool replayer status" for case when init failed

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

Merge pull request #32463 from trociny/wip-unlink-peer

librbd: when unlinking peer from mirror snaps do it in all namespaces

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

Merge pull request #32456 from trociny/wip-43428

rbd-mirror: make logrotate work

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

Merge pull request #32404 from kalebskeithley/master

rgw: build radosgw daemon as a shared lib + small executable

Merge pull request #32494 from ronen-fr/clang_3

crimson: fix lambda captures of non-variables

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

mds: maintain client provided metric flags in client metadata

Fixes: https://tracker.ceph.com/issues/43251
Signed-off-by: Venky Shankar <vshankar@redhat.com>

mds: add feature_bitset_t::dump()

Signed-off-by: Venky Shankar <vshankar@redhat.com>

mgr/dashboard: fix failing user tests

Choose another future_date_1 date (9 days instead of 10) to
make sure the expiration date will be between future_date_1
and future_date_2. Otherwise the expiration date might be
equal to future_date_1.
Also wait for 10 seconds before the password refresh in
test_pwd_expiration_date_update to prevent date equality.

Fixes: https://tracker.ceph.com/issues/43431
Signed-off-by: Tatjana Dehler <tdehler@suse.com>

Merge pull request #32499 from sebastian-philipp/orchestrator-cli-mon-update-count-only

mgr/orchestrator_cli: _update_mons require host spec only

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

crimson: fix lambda captures of non-variables

One cannot just capture a structured binding "non-variable".
(From the C++ standard, $8.4.5.2:
"If a lambda-expression [...] captures a structured binding (explicitly or implicitly),
the program is ill-formed.")

Co-authored-by: Kefu Chai <kchai@redhat.com>
Signed-off-by: Ronen Friedman <rfriedma@redhat.com>

Merge pull request #32376 from idryomov/wip-krbd-fsx-object-map

qa/suites/krbd: fsx with object-map and fast-diff

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

Merge pull request #32518 from tchaikov/wip-orchestrator-service-action-tests

qa/tasks/mgr/test_orchestrator_cli: fix service action tests

Reviewed-by: Thomas Bechtold <tbechtold@suse.com>
Reviewed-by: Sebastian Wagner <swagner@suse.com>

mgr/rook: Add some more type annotations

Signed-off-by: Sebastian Wagner <sebastian.wagner@suse.com>

mgr/orchestrator_cli: _update_mons require host spec only

* `mgr/cephadm` requres a host spec right now
* `mgr/rook` only supports `spec.count` right now

Signed-off-by: Sebastian Wagner <sebastian.wagner@suse.com>

Merge pull request #32496 from majianpeng/osd-remove-unused-func2

osd/OSD: remove unused func enqueue_peering_evt_front

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

Merge pull request #32484 from bzed/bluefs-use-uint64_t-for-len

os/bluestore/Blue(FS|Store): uint64_t alloc_size

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