Merge pull request #29227 from ifed01/wip-ifed-slow-ops2-nau

nautilus: bluestore: add slow op detection for collection_listing

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

Merge pull request #28756 from smithfarm/wip-40231-nautilus

nautilus: core: maybe_remove_pg_upmap can be super inefficient for large clusters

Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>

Merge pull request #28869 from dzafman/wip-backport-40073

nautilus: osd: Fix the way that auto repair triggers after regular scrub

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

Merge pull request #28993 from smithfarm/wip-40730-nautilus

nautilus: mon: auth mon isn't loading full KeyServerData after restart

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

Merge pull request #29115 from liewegas/wip-39693-nautilus

nautilus: osd/PG: fix cleanup of pgmeta-like objects on PG deletion

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

Merge pull request #29140 from liewegas/wip-40441-nautilus

nautilus: msg/async, v2: make the reset_recv_state() unconditional

Reviewed-by: Radoslaw Zarzynski <rzarzyns@redhat.com>

Merge pull request #29141 from liewegas/wip-msg-async-1-nautilus

nautilus: msg/async: avoid unnecessary costly wakeups for outbound messages

Merge pull request #29142 from liewegas/wip-msg-noneed-set-connection-nautilus

nautilus: msg/async: no-need set connection for Message.

Reviewed-by: Ricardo Dias <rdias@suse.com>

Merge pull request #29143 from liewegas/wip-secure-nautilus

nautilus: msg/async: enable secure mode by default, no longer experimental

Reviewed-by: Ricardo Dias <rdias@suse.com>
Reviewed-by: Neha Ojha <nojha@redhat.com>

Merge pull request #29159 from smithfarm/wip-40845-nautilus

nautilus: core: MDSMonitor: use stringstream instead of dout for mds repaired

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

Merge pull request #29188 from badone/wip-nautilus-lazy-omap-stats-backport-tracker-40744

nautilus: core: lazy omap stat collection

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

Merge pull request #28550 from smithfarm/wip-40007-nautilus

nautilus: rgw: fix prefix handling in LCFilter

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

Merge pull request #28648 from pdvian/wip-40107-nautilus

nautilus: rgw: Librgw doesn't GC deleted object correctly

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

Merge pull request #28712 from pdvian/wip-40134-nautilus

nautilus: rgw: TempURL should not allow PUTs with the X-Object-Manifest.

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

Merge pull request #28713 from smithfarm/wip-40505-nautilus

nautilus: rgw: fix miss get ret in STSService::storeARN

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

Merge pull request #28714 from smithfarm/wip-40515-nautilus

nautilus: rgw: multisite: DELETE Bucket CORS is not forwarded to master zone

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

Merge pull request #28715 from smithfarm/wip-40508-nautilus

nautilus: rgw: conditionally allow builtin users with non-unique email addresses

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

Merge pull request #28728 from joke-lee/nautilus

nautilus: rgw, Policy should be url_decode when assume_role

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

Merge pull request #28729 from joke-lee/nautilus-backport-28172

nautilus: rgw: fix rgw crash and set correct error code

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

Merge pull request #28735 from smithfarm/wip-40125-nautilus

nautilus: rgw: hadoop-s3a suite failing with more ansible errors

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

Merge pull request #28736 from smithfarm/wip-40129-nautilus

nautilus: rgw: Swift interface: server side copy fails if object name contains "?"

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

Merge pull request #28737 from smithfarm/wip-40137-nautilus

nautilus: rgw: the Multi-Object Delete operation of S3 API wrongly handles the Code response element

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

Merge pull request #28739 from smithfarm/wip-40142-nautilus

nautilus: rgw: multisite: 'radosgw-admin bucket sync status' should call syncs_from(source.name) instead of id

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

Merge pull request #28751 from pdvian/wip-40150-nautilus

nautilus: rgw: add 'GET /admin/realm?list' api to list realms

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

Merge pull request #28854 from pdvian/wip-40226-nautilus

nautilus: rgw_file: include tenant when hashing bucket names

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

Merge pull request #28886 from pdvian/wip-40263-nautilus

nautilus: rgw_file: all directories are virtual with respect to contents

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

Merge pull request #29154 from smithfarm/wip-40591-nautilus

nautilus: rgw: deleting bucket can fail when it contains unfinished multipart uploads

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

Merge pull request #29163 from smithfarm/wip-40851-nautilus

nautilus: rgw: multisite: radosgw-admin commands should not modify metadata on a non-master zone

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

Merge pull request #29205 from smithfarm/wip-40760-nautilus

nautilus: rgw: Save an unnecessary copy of RGWEnv

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

Merge pull request #29265 from smithfarm/wip-40124-nautilus

nautilus: rgw: hadoop-s3a suite failing with more ansible errors

Merge pull request #29286 from smithfarm/wip-40355-nautilus

nautilus: rgw: provide admin-friendly reshard status output

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

Merge pull request #29287 from smithfarm/wip-40358-nautilus

nautilus: rgw: set null version object issues

Merge pull request #29310 from smithfarm/wip-40349-nautilus

nautilus: rgw/OutputDataSocket: append_output(buffer::list&) says it will (but does not) discard output at data_max_backlog

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

Merge pull request #29311 from smithfarm/wip-40352-nautilus

nautilus: rgw: multisite: RGWListBucketIndexesCR for data full sync needs pagination

Merge pull request #29313 from smithfarm/wip-40512-nautilus

nautilus: rgw: Put LC doesn't clear existing lifecycle

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

Merge pull request #29325 from smithfarm/wip-40450-nautilus

nautilus: core: s3tests-test-readwrite failed in rados run (Connection refused)

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

Merge pull request #28549 from smithfarm/wip-39749-nautilus

nautilus: rgw: multisite: fix --bypass-gc flag for 'radosgw-admin bucket rm'

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

Merge pull request #29040 from dcasier/wip-40100-nautilus

nautilus: ceph-volume: lvm.activate: Return an error if WAL/DB devices absent

Reviewed-by: Jan Fajerski <jfajerski@suse.com>
Reviewed-by: Alfredo Deza <adeza@redhat.com>

qa/rgw: update default port in perl workunits

Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 1643879218638915073d5310b859a94d10ffeac6)

qa/rgw: extra s3tests tasks use rgw endpoint configuration

Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 09e992ff01b4ce286540e1230a30df67103f5968)

rgw_lc: use a new bl while encoding RGW_ATTR_LC

The current code seems to append the value to the bucket attr which isn't what
we want

Fixes: https://tracker.ceph.com/issues/39654
Signed-off-by: Abhishek Lekshmanan <abhishek@suse.com>
(cherry picked from commit e23af175aff1d80a7857406f7c17f5a609ad7e5e)

rgw/multisite:RGWListBucketIndexesCR for data full sync pagination

RGWListBucketIndexesCR uses the 'marker' from remote zone's RGWOp_Metadata_List
admin api to list all keys for data sync.

Fixes http://tracker.ceph.com/issues/39551

Signed-off-by: Shilpa Jagannath <smanjara@redhat.com>
(cherry picked from commit 708ba6dcc90525c8c4316760e01ee9b1e6da21b7)

rgw/OutputDataSocket: actually discard data on full buffer

A dout message in OutputDataSocket::append_output() states that
data will be dropped when appending would cause data_max_backlog
to be exceeded--but the method appends it anyway.

Log output discards at level 0, as messages will be lost.  Suppress
repeated warnings mod 100.  Switch to vector.

Fixes: http://tracker.ceph.com/issues/40178
Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
(cherry picked from commit c806b825dae649829de8847d36cb21ffd2bbee8e)

Merge pull request #29260 from jan--f/wip-40921-nautilus

nautilus: ceph-volume: missing string substitution when reporting mounts

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

rgw: set null version object acl issues

1.set null version object acl will create empty index
RGWRados::set_attrs did not clear instance, so index prepare, complete got instance=null,
which lead to empty index 1000_<obj>_i_null.
there is no harm to create empty index, but listomapkeys to find that key.

2.if object is exist with versioned key, we can set none exists null version object
order:
1) enable bucket version
2) put obj
3) disable bucket version
4) set versoned_id=null acl will succeed which should not

Fixes: http://tracker.ceph.com/issues/36763
Signed-off-by: Tianshan Qu <tianshan@xsky.com>
(cherry picked from commit ca6e5c89e8d6da498da15f0eea4bcd4d271b44e9)

rgw: provide admin friendly reshard status output

Signed-off-by: Mark Kogan <mkogan@redhat.com>
(cherry picked from commit ba4167c2f30fbb868482d48aaada6388f58de920)

rgw: update the "radosgw-admin reshard status"
command documentation with expected output examples

Fixes: http://tracker.ceph.com/issues/37615
Signed-off-by: Mark Kogan <mkogan@redhat.com>
(cherry picked from commit 8565ec0db69223ae71766ba000eecc0d2f7d14c2)

qa/rgw: add dnsmasq back to s3a-hadoop

Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 8c8a706b6f4cb3b2a5b89aa1eb06b7e47909852d)

qa/rgw: remove ceph-ansible from s3a-hadoop suite

Fixes: http://tracker.ceph.com/issues/39706
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 0fc2c8ecee2b6233292b9fd1325347fd0fdf9171)

Conflicts:
qa/tasks/s3a_hadoop.py
- mimic lacks the "if hadoop_ver == 'trunk'" conditional block

qa/rgw: use default ports (80 or 443) unless overridden

Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 3ff5f0c2406e40d1024d8152e0ac9400302a757a)

qa/rgw: rgw task can override --rgw-dns-name on the command line

the value for rgw_dns_name isn't known until a machine is assigned, so
it can't be set via 'override: conf:'. add a per-client config option
to the rgw task so it can add the endpoint's hostname and/or s3website
hostname on the radosgw command line

Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 658e5932fb79e2d33b73363b6ce76ff299809e16)

qa/rgw: allow rgw client config to override port

Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 921faebb723c11686bf790ca424c952a786f358a)

ceph-volume:util: Use proper param substition

Fixes: https://tracker.ceph.com/issues/25030
Signed-off-by: Shyukri Shyukriev <shshyukriev@suse.com>
(cherry picked from commit 654660cd89fe9d93087d390805651c860bf4e862)

Merge pull request #28958 from smithfarm/wip-40710-nautilus

nautilus: doc: Document more cache modes

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

Merge pull request #28919 from LenzGr/nautilus-documentation

nautilus: doc: Improved dashboard feature overview

Reviewed-by: Tatjana Dehler <tdehler@suse.com>
Reviewed-by: Ricardo Marques <rimarques@suse.com>
Reviewed-by: Tiago Melo <tmelo@suse.com>
Reviewed-by: Patrick Nawracay <pnawracay@suse.com>
Reviewed-by: Laura Paduano <lpaduano@suse.com>
Reviewed-by: Volker Theile <vtheile@suse.com>

Merge pull request #28738 from smithfarm/wip-40140-nautilus

nautilus: doc: steps to disable metadata_heap on existing rgw zones

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

os/bluestore: add slow op detector for collection listing

Signed-off-by: Igor Fedotov <ifedotov@suse.com>
(cherry picked from commit 5218e5ee6e741dd65b9d1c9b366c135b7b000b6b)

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

os/bluestore: parametrize latency threshold for log_latency funcs..

Signed-off-by: Igor Fedotov <ifedotov@suse.com>
(cherry picked from commit 00af2c411dec6c8a70f4a2c8e4b0398e348fd037)

os/bluestore: cleanup around slow op logging.

We don't need macros any more.

Signed-off-by: Igor Fedotov <ifedotov@suse.com>
(cherry picked from commit 2f5bd9c48f3818b85a502b650db1056004267ccd)

os/bluestore: fix origin reference in logging slow ops.

Signed-off-by: Igor Fedotov <ifedotov@suse.com>
(cherry picked from commit 76c5993601e7fa3346a7958f6e8d2f4e68b07c9a)

rgw: Save an unnecessary copy of RGWEnv

fixes: https://tracker.ceph.com/issues/40183

Signed-off-by: Mark Kogan <mkogan@redhat.com>
(cherry picked from commit 0693ec241ec40b3b7c21e9182b99213024b41049)

osd: Modify lazy omap stats to only show one copy

Only show stats from the primary to get more in line with other pg stats
representations.

Signed-off-by: Brad Hubbard <bhubbard@redhat.com>
(cherry picked from commit 490829c8eaf14f41de978ac0f93ecda3df8bcdef)

tests: Add test for lazy omap stat collection

Signed-off-by: Brad Hubbard <bhubbard@redhat.com>
(cherry picked from commit 88e9ca58a06ba572ff4c6f6125a8ba7f8e0a7795)

Merge pull request #29137 from yuriw/wip-yuriw-40832-nautilus

qa/tests: added 14.2.2 to the mix

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

Merge pull request #28768 from dzafman/wip-40265

nautilus: osd: Don't randomize deep scrubs when noscrub set

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

Merge pull request #29050 from rhcs-dashboard/fix-40768-nautilus

nautilus: mgr/dashboard: switch ng2-toastr to ngx-toastr

Reviewed-by:  Tiago Melo <tmelo@suse.com>
Reviewed-by: Volker Theile <vtheile@suse.com>
Reviewed-by: Yuri Weinstein <yweinste@redhat.com>

qa/tests: added 14.2.2 to the mix

Fixes: http://tracker.ceph.com/issues/40251
Fixes: http://tracker.ceph.com/issues/40832
Signed-off-by: Yuri Weinstein <yweinste@redhat.com>

Merge pull request #28230 from ivancich/nautilus-wip-rgw-admin-unordered

nautilus: rgw: allow radosgw-admin to list bucket w --allow-unordered

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

Merge pull request #28769 from dillaman/wip-40572-nautilus

nautilus: librbd: properly track in-flight flush requests

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

Merge pull request #28816 from trociny/wip-40462-nautilus

nautilus: journal: properly advance read offset after skipping invalid range

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

Merge pull request #28817 from trociny/wip-40594-nautilus

nautilus: rbd-mirror: handle duplicates in image sync throttler queue

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

Merge pull request #28937 from smithfarm/wip-systemd-dep-suse-nautilus

nautilus: ceph.spec.in: Drop systemd BuildRequires in case of building for SUSE

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

Added single check to avoid duplication. Included few more commands.

Signed-off-by: Shilpa Jagannath <smanjara@redhat.com>
(cherry picked from commit 14c3b4b3ebb6ef2fc44d24f8020beb2977c8e46e)

rgw: Fail radosgw-admin commands on non-master zone that modify metadata
but with an option to override, allowing changes only on the local zone.

Fixes: http://tracker.ceph.com/issues/39548
Signed-off-by: Shilpa Jagannath <smanjara@redhat.com>
(cherry picked from commit 79d884eb018cd2fea9ee9e763d0339b3e8626e6d)

mon/MDSMonitor: use stringstream instead of dout for mds repaired

Fixes: http://tracker.ceph.com/issues/40472
Signed-off-by: Zhi Zhang <zhangz.david@outlook.com>
(cherry picked from commit 7ad76ea2280f772cb3ebed54530e4848b841da78)

rgw: minor code clean-up

Adding some const correctness and reformatting.

Signed-off-by: J. Eric Ivancich <ivancich@redhat.com>
(cherry picked from commit fb862909202c12c4bd908a0efe140c0f0ef0bd5a)

rgw: allow multipart upload abort to proceed

Cleaning up after aborting unfinished multipart uploads is a
best-effort process. Due to the history of the multipart upload, not
all components may be found and the process needs to be robust to
that.

Presently, if a component is not found after the final call to
list_bucket_multiparts, the error propogates to the caller, but not if
the component is not found in an earlier call. This change does not
propogate that error up ever.

Additional logging is also added to multipart-related functions.

Signed-off-by: J. Eric Ivancich <ivancich@redhat.com>
(cherry picked from commit 51aff62ac7cbb4653e0de667dccfa46795ffd8cb)

common/options: allow (but to not prefer or require) secure mode

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 1aa28cde19a87d2fbf5d0fec8d659bc726e9968a)

# Conflicts:
# src/common/options.cc
- STARTUP flag backport ordering

common/options: make clients prefer to connect to mons via secure mode

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit cd4b4f01b9b46587fc7f3ce65e956213af18f300)

# Conflicts:
# src/common/options.cc
- STARTUP flag was already backported

common/options: allow connections to mons in secure mode

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 3bdc49dc80aa75a68e7c0da2fcf7c1ccdcf03fe4)

# Conflicts:
# src/common/options.cc
- the STARTUP flag was already backported

common/options: prefer secure mode between monitors

We still allow crc here in order to facilitate upgrades.

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit a39273b79430cc0683b29fb7550cdbf18c760d98)

# Conflicts:
# src/common/options.cc
 - the STARTUP flag was already backported

auth/AuthRegistry: remove experimental flag for 'secure' mode

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit ac3abbef48168a97a7374cc10fa3dc1c78123a98)

msg/async: no-need set connection for Message.

In function decode_message, it already did.

Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
(cherry picked from commit 7e1fcf82b61bf624fc5a1a8e647f74378ebccf4a)

msg/async: avoid unnecessary costly wakeups for outbound messages

If a wakeup for an outbound message has already been scheduled or is
currently executing within the worker thread, avoid re-adding a wakeup.
For small IO sizes under high queue depths, these extra syscalls start
to add up. For larger IO sizes or small queue depths, it doesn't hurt
performance.

fio --ioengine=rbd results:

IOPS pre-change post-change
4K: 84.9k 98.3k
32K: 58.4k 59.5k
256K: 12.1k 12.2k
4M: 803 802

Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 294c41f18adada6ab762d5b3f9f45d7c8bed4316)

qa/valgrind.supp: be slightly less specific on suppression

I am seeing this trace, which matches except for the
'fun:_ZN15AsyncConnection7processEv' frame.

<error>
  <unique>0x2399</unique>
  <tid>11</tid>
  <threadname>msgr-worker-1</threadname>
  <kind>UninitCondition</kind>
  <what>Conditional jump or move depends on uninitialised value(s)</what>
  <stack>
    <frame>
      <ip>0x5366B18</ip>
      <obj>/usr/lib64/ceph/libceph-common.so.0</obj>
      <fn>ceph::crypto::onwire::AES128GCM_OnWireRxHandler::authenticated_decrypt_update_final(ceph::buffer::v14_2_0::list&amp;&amp;, unsigned int)</fn>
      <dir>/usr/src/debug/ceph-15.0.0-1717-g8d72af7/src/msg/async</dir>
      <file>crypto_onwire.cc</file>
      <line>274</line>
    </frame>
    <frame>
      <ip>0x5355E60</ip>
      <obj>/usr/lib64/ceph/libceph-common.so.0</obj>
      <fn>ProtocolV2::handle_read_frame_epilogue_main(std::unique_ptr&lt;ceph::buffer::v14_2_0::ptr_node, ceph::buffer::v14_2_0::ptr_node::disposer&gt;&amp;&amp;, int)</fn>
      <dir>/usr/src/debug/ceph-15.0.0-1717-g8d72af7/src/msg/async</dir>
      <file>ProtocolV2.cc</file>
      <line>1311</line>
    </frame>
    <frame>
      <ip>0x533E2A3</ip>
      <obj>/usr/lib64/ceph/libceph-common.so.0</obj>
      <fn>ProtocolV2::run_continuation(Ct&lt;ProtocolV2&gt;&amp;)</fn>
      <dir>/usr/src/debug/ceph-15.0.0-1717-g8d72af7/src/msg/async</dir>
      <file>ProtocolV2.cc</file>
      <line>45</line>
    </frame>
    <frame>
      <ip>0x534FB1C</ip>
      <obj>/usr/lib64/ceph/libceph-common.so.0</obj>
      <fn>ProtocolV2::reuse_connection(boost::intrusive_ptr&lt;AsyncConnection&gt; const&amp;, ProtocolV2*)::{lambda(ConnectedSocket&amp;)#3}::operator()(ConnectedSocket&amp;)::{lambda()#2}::operator()()</fn>
      <dir>/usr/src/debug/ceph-15.0.0-1717-g8d72af7/src/msg/async</dir>
      <file>ProtocolV2.cc</file>
      <line>2739</line>
    </frame>
    <frame>
      <ip>0x534FF57</ip>
      <obj>/usr/lib64/ceph/libceph-common.so.0</obj>
      <fn>ProtocolV2::reuse_connection(boost::intrusive_ptr&lt;AsyncConnection&gt; const&amp;, ProtocolV2*)::{lambda(ConnectedSocket&amp;)#3}::operator()(ConnectedSocket&amp;)</fn>
      <dir>/usr/src/debug/ceph-15.0.0-1717-g8d72af7/src/msg/async</dir>
      <file>ProtocolV2.cc</file>
      <line>2745</line>
    </frame>
    <frame>
      <ip>0x535001E</ip>
      <obj>/usr/lib64/ceph/libceph-common.so.0</obj>
      <fn>__invoke_impl&lt;void, ProtocolV2::reuse_connection(const AsyncConnectionRef&amp;, ProtocolV2*)::&lt;lambda(ConnectedSocket&amp;)&gt;&amp;, ConnectedSocket&amp;&gt;</fn>
      <dir>/opt/rh/devtoolset-8/root/usr/include/c++/8/bits</dir>
      <file>invoke.h</file>
      <line>60</line>
    </frame>
    <frame>
      <ip>0x535001E</ip>
      <obj>/usr/lib64/ceph/libceph-common.so.0</obj>
      <fn>__invoke&lt;ProtocolV2::reuse_connection(const AsyncConnectionRef&amp;, ProtocolV2*)::&lt;lambda(ConnectedSocket&amp;)&gt;&amp;, ConnectedSocket&amp;&gt;</fn>
      <dir>/opt/rh/devtoolset-8/root/usr/include/c++/8/bits</dir>
      <file>invoke.h</file>
      <line>95</line>
    </frame>
    <frame>
      <ip>0x535001E</ip>
      <obj>/usr/lib64/ceph/libceph-common.so.0</obj>
      <fn>__call&lt;void, 0&gt;</fn>
      <dir>/opt/rh/devtoolset-8/root/usr/include/c++/8</dir>
      <file>functional</file>
      <line>400</line>
    </frame>
    <frame>
      <ip>0x535001E</ip>
      <obj>/usr/lib64/ceph/libceph-common.so.0</obj>
      <fn>operator()&lt;&gt;</fn>
      <dir>/opt/rh/devtoolset-8/root/usr/include/c++/8</dir>
      <file>functional</file>
      <line>484</line>
    </frame>
    <frame>
      <ip>0x535001E</ip>
      <obj>/usr/lib64/ceph/libceph-common.so.0</obj>
      <fn>EventCenter::C_submit_event&lt;std::_Bind&lt;ProtocolV2::reuse_connection(boost::intrusive_ptr&lt;AsyncConnection&gt; const&amp;, ProtocolV2*)::{lambda(ConnectedSocket&amp;)#3} (ConnectedSocket)&gt; &gt;::do_request(unsigned long)</fn>
      <dir>/usr/src/debug/ceph-15.0.0-1717-g8d72af7/src/msg/async</dir>
      <file>Event.h</file>
      <line>227</line>
    </frame>
    <frame>
      <ip>0x535FCD6</ip>
      <obj>/usr/lib64/ceph/libceph-common.so.0</obj>
      <fn>EventCenter::process_events(unsigned int, std::chrono::duration&lt;unsigned long, std::ratio&lt;1l, 1000000000l&gt; &gt;*)</fn>
      <dir>/usr/src/debug/ceph-15.0.0-1717-g8d72af7/src/msg/async</dir>
      <file>Event.cc</file>
      <line>441</line>
    </frame>
    <frame>
      <ip>0x5365086</ip>
      <obj>/usr/lib64/ceph/libceph-common.so.0</obj>
      <fn>operator()</fn>
      <dir>/usr/src/debug/ceph-15.0.0-1717-g8d72af7/src/msg/async</dir>
      <file>Stack.cc</file>
      <line>53</line>
    </frame>
    <frame>
      <ip>0x5365086</ip>
      <obj>/usr/lib64/ceph/libceph-common.so.0</obj>
      <fn>std::_Function_handler&lt;void (), NetworkStack::add_thread(unsigned int)::{lambda()#1}&gt;::_M_invoke(std::_Any_data const&amp;)</fn>
      <dir>/opt/rh/devtoolset-8/root/usr/include/c++/8/bits</dir>
      <file>std_function.h</file>
      <line>297</line>
    </frame>
    <frame>
      <ip>0x55F519E</ip>
      <obj>/usr/lib64/ceph/libceph-common.so.0</obj>
      <fn>execute_native_thread_routine</fn>
    </frame>
    <frame>
      <ip>0x1076BDD4</ip>
      <obj>/usr/lib64/libpthread-2.17.so</obj>
      <fn>start_thread</fn>
    </frame>
    <frame>
      <ip>0x118E0EAC</ip>
      <obj>/usr/lib64/libc-2.17.so</obj>
      <fn>clone</fn>
    </frame>
  </stack>
</error>

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit f019fc094ecec0a0109d9afd28bdf87e11c69a0d)

msg/async, v2: make the reset_recv_state() unconditional.

Fixes: http://tracker.ceph.com/issues/40115
Signed-off-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
(cherry picked from commit eb8b9c50e4fa5459d32ddb4ad93d011a28f74179)

Merge PR #29079 into nautilus

* refs/pull/29079/head:
client: do not return EEXIST for mkdirs
pybind/mgr/volumes: print exceptions in purge thread
pybind/mgr/volumes: refactor trash readdir
pybind/mgr/volumes: use existing client provided recursive mkdir
pybind/mgr/volumes: cleanup fs removal
pybind/mgr/subvolumes: use bytes for paths
pybind/mgr/volumes: remove unused property
test: cleanup removing all subvolumes before removing subvolume group
mgr / volumes: wrap rmtree() call within try..except block
mgr / volumes: use negative error codes everywhere
test: add basic purge queue validation test
mgr / volumes: schedule purge job for volumes on init
mgr / volumes: purge queue for async subvolume delete
mgr / volumes: maintain connection pool for fs volumes
mgr/volumes: do not import unused module

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

os/bluestore: be verbose about objects that existing on rmcoll

This is always a bug (OSD doesn't try to remove a collection unless it
thinks it is empty), and not seeing it at default debug levels makes it
hard to track down.

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit e387cbc45037db78db5dfbb01483b8d76e1c276f)

osd/PrimaryLogPG: disallow ops on objects with an empty name

These may conflict with pgmeta objects.  And are going to cause other
problems later (e.g., https://tracker.ceph.com/issues/38724).

Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit bf6e8a4a4db8b7d7101ca80b844ee953aa4580b2)

osd/PG: fix cleanup of pgmeta-like objects on PG deletion

If an object has an empty 'name' field, it "looks" like a pgmeta object,
and the PG cleanup code was skipping it.  However, we were letting these
objects get created.

Fix by only skipping *our* pgmeta object.  If there are other pgmeta-like
objects in the PG collection, clean them up.

Fixes: https://tracker.ceph.com/issues/38724
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit a0f48db5cbfc48761eeac0daa259fecbaa573969)

# Conflicts:
# src/osd/PG.cc
- location of this code is do_delete_work in master, _delete_some in n

14.2.2

mgr/dashboard: Switch ng2-toastr for ngx-toastr

Fixes: https://tracker.ceph.com/issues/40768
Signed-off-by: Tiago Melo <tmelo@suse.com>
(cherry picked from commit 67f5e5b18b71ccabcaa8d9e35c57630d17395b06)
Signed-off-by: Ernesto Puerta <epuertat@redhat.com>
Required manual (trivial) fixing of the following files:
Conflicts:
      src/pybind/mgr/dashboard/frontend/package.json
      src/pybind/mgr/dashboard/frontend/src/app/ceph/cluster/prometheus/alert-list/alert-list.component.spec.ts
      src/pybind/mgr/dashboard/frontend/src/app/ceph/cluster/prometheus/silence-form/silence-form.component.spec.ts
      src/pybind/mgr/dashboard/frontend/src/app/ceph/cluster/prometheus/silence-list/silence-list.component.spec.ts
      src/pybind/mgr/dashboard/frontend/src/app/ceph/pool/pool-list/pool-list.component.spec.ts
      src/pybind/mgr/dashboard/frontend/src/app/shared/services/prometheus-notification.service.spec.ts

client: do not return EEXIST for mkdirs

Behavior should be similar to `mkdir -p`.

Introduced-by: 26905ca82776e8119a347a9a2e8d65c272c7ed62
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
(cherry picked from commit b049eb9bbcdaef547c3559e9dbe5a3ca272f60ec)

pybind/mgr/volumes: print exceptions in purge thread

Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
(cherry picked from commit ecf85023954f464c328671c897d610c8ebc8c789)

pybind/mgr/volumes: refactor trash readdir

Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
(cherry picked from commit e71e9187ec69ce80e9d7a2fc2c58946dd3db88e1)

pybind/mgr/volumes: use existing client provided recursive mkdir

Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
(cherry picked from commit e9c314abd06e2d91a58cbd681978a3abb6134952)

pybind/mgr/volumes: cleanup fs removal

In Nautilus, the simplest sequence is:

    fs fail name
    fs rm name --yes-i-really-mean-it

Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
(cherry picked from commit 7cc4b72305a895e99ae79e6a4d09f45c23a6257c)

pybind/mgr/subvolumes: use bytes for paths

Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
(cherry picked from commit 3d63cd947f55a8508f972ee249ea6a04836429ae)

pybind/mgr/volumes: remove unused property

Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
(cherry picked from commit f7987f5fcba322879046e86720f1db55d076d881)

test: cleanup removing all subvolumes before removing subvolume group

Test `test_subvolume_create_with_desired_mode_in_group()` creates three
subvolume in a subvolume group. During cleanup, it only removed two of
the three subvolumes. This causes failure when removing the subvolume
group since it's not empty.

Signed-off-by: Venky Shankar <vshankar@redhat.com>
(cherry picked from commit ae470a64ecf8ded8bbe3dbe02b5a470fecb7e2c3)