commit 63b0f854d1cef490624de5d6cf9039735c7de5ca
Author: Gary Lowell <gary.lowell@inktank.com>
Date:   Mon Mar 25 11:02:31 2013 -0700

    v0.56.4

commit 90ab120a3c08f30654207c9c67fb0a4ff1476aff
Author: Yehuda Sadeh <yehuda@inktank.com>
Date:   Mon Mar 25 09:50:33 2013 -0700

    rgw: bucket index ops on system buckets shouldn't do anything
    
    Fixes: #4508
    Backport: bobtail
    On certain bucket index operations we didn't check whether
    the bucket was a system bucket, which caused the operations
    to fail. This triggered an error message on bucket removal
    operations.
    
    Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
    Reviewed-by: Greg Farnum <greg@inktank.com>
    (cherry picked from commit 70e0ee8ba955322832f0c366537ddf7a0288761e)

commit b8657fa08d0fe176c71a60c1654c53c3156a9c53
Author: Josh Durgin <josh.durgin@inktank.com>
Date:   Mon Feb 25 15:02:50 2013 -0800

    systest: restrict list error acceptance
    
    Only ignore errors after the midway point if the midway_sem_post is
    defined.
    
    Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
    (cherry picked from commit 5b24a68b6e7d57bac688021b822fb2f73494c3e9)

commit 5af3d64f62b30cbaac13ccc7c23fc884ee1b7ae3
Author: Josh Durgin <josh.durgin@inktank.com>
Date:   Mon Feb 25 14:55:34 2013 -0800

    systest: fix race with pool deletion
    
    The second test have pool deletion and object listing wait on the same
    semaphore to connect and start. This led to errors sometimes when the
    pool was deleted before it could be opened by the listing process. Add
    another semaphore so the pool deletion happens only after the listing
    has begun.
    
    Fixes: #4147
    Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
    (cherry picked from commit b0271e390564119e998e18189282252d54f75eb6)

commit 3569489b541ac0643520d20b08c788c26dfaff7f
Author: Sage Weil <sage@inktank.com>
Date:   Tue Mar 19 14:26:16 2013 -0700

    os/FileJournal: fix aio self-throttling deadlock
    
    This block of code tries to limit the number of aios in flight by waiting
    for the amount of data to be written to grow relative to a function of the
    number of aios.  Strictly speaking, the condition we are waiting for is a
    function of both aio_num and the write queue, but we are only woken by
    changes in aio_num, and were (in rare cases) waiting when aio_num == 0 and
    there was no possibility of being woken.
    
    Fix this by verifying that aio_num > 0, and restructuring the loop to
    recheck that condition on each wakeup.
    
    Fixes: #4079
    Signed-off-by: Sage Weil <sage@inktank.com>
    Reviewed-by: Samuel Just <sam.just@inktank.com>
    (cherry picked from commit e5940da9a534821d0d8f872c13f9ac26fb05a0f5)

commit 664ffa7d2178e486b00fa0706067f19b1bb9ab82
Author: Sage Weil <sage@inktank.com>
Date:   Fri Mar 22 13:25:49 2013 -0700

    common/MemoryModel: remove logging to /tmp/memlog
    
    This was a hack for dev purposes ages ago; remove it.  The predictable
    filename is a security issue.
    
    CVE-2013-1882
    
    Reported-by: Michael Scherer <misc@zarb.org>
    Signed-off-by: Sage Weil <sage@inktank.com>
    Reviewed-by: Dan Mick <dan.mick@inktank.com>
    (cherry picked from commit c524e2e01da41ab5b6362c117939ea1efbd98095)

commit 901fa435bb4be29b72cec39e5ff77570d13cbc6c
Author: Sage Weil <sage@inktank.com>
Date:   Fri Mar 22 13:25:43 2013 -0700

    init-ceph: clean up temp ceph.conf filename on exit
    
    Signed-off-by: Sage Weil <sage@inktank.com>
    Reviewed-by: Dan Mick <dan.mick@inktank.com>
    (cherry picked from commit 6a7ad2eac1db6abca3d7edb23ca9b80751400a23)

commit 951f7e562afb4ae91b8d037ef793a96779461e96
Author: Sage Weil <sage@inktank.com>
Date:   Fri Mar 22 13:25:33 2013 -0700

    init-ceph: push temp conf file to a unique location on remote host
    
    The predictable file name is a security problem.
    
    CVE-2013-1882
    
    Reported-by: Michael Scherer <misc@zarb.org>
    Signed-off-by: Sage Weil <sage@inktank.com>
    Reviewed-by: Dan Mick <dan.mick@inktank.com>
    (cherry picked from commit 051734522fea92878dd8139f28ec4e6b01371ede)

commit dd670afccfd3ae58c03ea3b4c213dd8db22f618e
Author: Sage Weil <sage@inktank.com>
Date:   Fri Mar 22 13:25:23 2013 -0700

    mkcephfs: make remote temp directory name unique
    
    The predictable file name is a security problem.
    
    CVE-2013-1882
    
    Reported-by: Michael Scherer <misc@zarb.org>
    Signed-off-by: Sage Weil <sage@inktank.com>
    Reviewed-by: Dan Mick <dan.mick@inktank.com>
    (cherry picked from commit f463ef78d77b11b5ad78b31e9a3a88d0a6e62bca)

commit 1cd51eb63a91fa0aa395c41572cfee4d53fbc7f5
Author: Samuel Just <sam.just@inktank.com>
Date:   Fri Mar 22 13:51:14 2013 -0700

    PG::GetMissing: need to check need_up_thru in MLogRec handler
    
    Backport: bobtail
    Fixes: #4534
    Signed-off-by: Samuel Just <sam.just@inktank.com>
    Reviewed-by: Sage Weil <sage@inktank.com>
    (cherry picked from commit 4fe4deafbe1758a6b3570048aca57485bd562440)

commit d866814285667c7f9cd9205dca6cf95f3634c5fc
Author: Samuel Just <sam.just@inktank.com>
Date:   Fri Mar 22 13:48:49 2013 -0700

    PG,osd_types: improve check_new_interval debugging
    
    Backport: bobtail
    Signed-off-by: Samuel Just <sam.just@inktank.com>
    Reviewed-by: Sage Weil <sage@inktank.com>
    (cherry picked from commit d611eba9caf45f2d977c627b123462a073f523a4)

commit fc3406f395103737e5784611e2b38fd1c4f26369
Author: Samuel Just <sam.just@inktank.com>
Date:   Tue Mar 5 16:06:20 2013 -0800

    FileStore: fix reversed collection_empty return value
    
    Backport: bobtail
    Fixes: #4380
    Signed-off-by: Samuel Just <sam.just@inktank.com>
    Reviewed-by: Greg Farnum <greg@inktank.com>
    (cherry picked from commit 7a434d10da0f77e1b329de0b06b6645cd73cc81b)
    
    Conflicts:
    	src/os/FileStore.cc

commit 9f7c6ad395653b1039ea957cfaa7689699f510e7
Author: Samuel Just <sam.just@inktank.com>
Date:   Mon Feb 11 12:52:07 2013 -0800

    FileStore: set replay guard on create_collection
    
    This should prevent sequences like:
    
    rmcoll a
    mkcoll a
    touch a foo
    <crash>
    
    from causing trouble by preventing the rmcoll
    and mkcoll from being replayed.
    
    Fixes: 4064
    Backport: bobtail
    Signed-off-by: Samuel Just <sam.just@inktank.com>
    (cherry picked from commit 411770c45734c9827745ddc4018d86c14f2858a6)

commit 9932c9f0a0503f1902549a3c39e5d51488fff3ef
Author: Samuel Just <sam.just@inktank.com>
Date:   Mon Feb 11 12:24:14 2013 -0800

    FileStore: _split_collection should not create the collection
    
    This will simplify adding a replay guard to create_collection.
    
    Backport: bobtail
    Signed-off-by: Samuel Just <sam.just@inktank.com>
    (cherry picked from commit b184ff581a08c9e6ce5b858f06ccbe9d0e2a170b)

commit 7e7ea8266ee45d96863e1edc5a567dcc2977680c
Author: Sage Weil <sage@inktank.com>
Date:   Fri Feb 22 15:15:27 2013 -0800

    client: use 4MB f_bsize and f_frsize for statfs
    
    Old stat(1) reports:
    
      Block size: 1048576    Fundamental block size: 1048576
    
    and the df(1) arithmetic works out.  New stat(1) reports:
    
      Block size: 1048576    Fundamental block size: 4096
    
    which is what we are shoving into statvfs, but we have the b_size and
    fr_size arithmetic swapped.  However, doing the *correct* reporting would
    then break the old stat by making both sizes appear to be 4KB (or
    whatever).
    
    Sidestep the issue by making *both* values 4MB.. which is both large enough
    to report large FS sizes, and also the default stripe size and thus a
    "reasonable" value to report for a block size.
    
    Perhaps in the future, when we no longer care about old userland, we can
    report the page size for f_bsize, which is probably the "most correct"
    thing to do.
    
    Fixes: #3794.  See also #3793.
    Signed-off-by: Sage Weil <sage@inktank.com>
    Reviewed-by: Greg Farnum <greg@inktank.com>
    (cherry picked from commit 7c94083643891c9d66a117352f312b268bdb1135)

commit 66dca7730006dffbce12e6b27016ea5469b7691b
Author: Sage Weil <sage@inktank.com>
Date:   Mon Feb 18 17:39:46 2013 -0800

    os/FileStore: check replay guard on src for collection rename
    
    This avoids a problematic sequence like:
    
         - rename A/ -> B/
         - remove B/1...100
         - destroy B/
         - create A/
         - write A/101...
         <crash>
         - replay A/ -> B/
         - remove B/1...100  (fails but tolerated)
         - destroy B/        (fails with ENOTEMPTY)
    
    Signed-off-by: Sage Weil <sage@inktank.com>
    Reviewed-by: Samuel Just <sam.just@inktank.com>
    (cherry picked from commit 5fc83c8d9887d2a916af11436ccc94fcbfe59b7a)

commit ea570591b0d9dda102bece0f4cc7d263c3e2a3b0
Author: Samuel Just <sam.just@inktank.com>
Date:   Fri Feb 22 14:12:28 2013 -0800

    PG::proc_replica_log: oinfo.last_complete must be *before* first entry in omissing
    
    Fixes: #4189
    Signed-off-by: Samuel Just <sam.just@inktank.com>
    (cherry picked from commit 2dae6a68ee85a20220ee940dbe33a2144d43457b)

commit aa38d87c6237ccfb8b3b03bc7c28dcadb927237b
Author: Sage Weil <sage@inktank.com>
Date:   Thu Feb 21 17:55:21 2013 -0800

    osd/PG: fix typo, missing -> omissing
    
    From ce7ffc34408bf32c66dc07e6f42d54b7ec489d41.
    
    Signed-off-by: Sage Weil <sage@inktank.com>
    (cherry picked from commit dc181224abf6fb8fc583730ae3d90acdf0b80f39)

commit ac914834ba4fb0e4770f63c60713be7a48aa3fa1
Author: Samuel Just <sam.just@inktank.com>
Date:   Thu Feb 21 15:31:36 2013 -0800

    PG::proc_replica_log: adjust oinfo.last_complete based on omissing
    
    Otherwise, search_for_missing may neglect to check the missing
    set for some objects assuming that if the need version is
    prior to last_complete, the replica must have it.
    
    Fixes: #4994
    Signed-off-by: Samuel Just <sam.just@inktank.com>
    (cherry picked from commit ce7ffc34408bf32c66dc07e6f42d54b7ec489d41)

commit 7f39704b0903fb11e3f1e102599bf0c540f93623
Author: Sage Weil <sage@inktank.com>
Date:   Sat Feb 9 00:05:33 2013 -0800

    osd: fix load_pgs collection handling
    
    On a _TEMP pg, is_pg() would succeed, which meant we weren't actually
    hitting the cleanup checks.  Instead, restructure this loop as positive
    checks and handle each type of collection we understand.
    
    This fixes _TEMP cleanup.
    
    Signed-off-by: Sage Weil <sage@inktank.com>
    Reviewed-by: Samuel Just <sam.just@inktank.com>
    (cherry picked from commit b19b6dced85617d594c15631571202aab2f94ae8)

commit 0a09be99f3e9d2b68a052ff63ea858a7c8e5626c
Author: Sage Weil <sage@inktank.com>
Date:   Sat Feb 9 00:04:29 2013 -0800

    osd: fix load_pgs handling of pg dirs without a head
    
    If there is a pgid that passes coll_t::is_pg() but there is no head, we
    will populate the pgs map but then fail later when we try to do
    read_state.  This is a side-effect of 55f8579.
    
    Take explicit note of _head collections we see, and then warn when we
    find stray snap collections.
    
    Signed-off-by: Sage Weil <sage@inktank.com>
    Reviewed-by: Samuel Just <sam.just@inktank.com>
    (cherry picked from commit 1f80a0b576c0af1931f743ad988b6293cbf2d6d9)

commit 2dc25dfd47de8d3602bdb3699c82b8b118b0fb28
Author: Samuel Just <sam.just@inktank.com>
Date:   Thu Feb 7 13:34:47 2013 -0800

    OSD::load_pgs: first scan colls before initing PGs
    
    Backport: bobtail
    Signed-off-by: Samuel Just <sam.just@inktank.com>
    Reviewed-by: Sage Weil <sage@inktank.com>
    (cherry picked from commit 073f58ede2e473af91f76d01679631c169274af7)

commit 2fd8db29ccf56fd5473826f20e351edfe200419d
Author: David Zafman <david.zafman@inktank.com>
Date:   Tue Jan 8 19:24:13 2013 -0800

    osd: Add digest of omap for deep-scrub
    
    Add ScrubMap encode/decode v4 message with omap digest
    Compute digest of header and key/value.  Use bufferlist
    to reflect structure and compute as we go, clearing
    bufferlist to reduce memory usage.
    
    Signed-off-by: David Zafman <david.zafman@inktank.com>
    Reviewed-by: Samuel Just <sam.just@inktank.com>
    (cherry picked from commit 509a93e89f04d7e9393090563cf7be8e0ea53891)

commit ee59f517ccc195e591e0b1b6ddf6d1d054c4e2fd
Author: Samuel Just <sam.just@inktank.com>
Date:   Fri Mar 15 15:13:46 2013 -0700

    OSD: split temp collection as well
    
    Otherwise, when we eventually remove the temp collection, there might be
    objects in the temp collection which were independently pulled into the child
    pg collection.  Thus, removing the old stale parent link from its temp
    collection also blasts the omap entries and snap mappings for the real child
    object.
    
    Backport: bobtail
    Fixes: #4452
    Signed-off-by: Samuel Just <sam.just@inktank.com>
    Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
    (cherry picked from commit f8d66e87a5c155b027cc6249006b83b4ac9b6c9b)

commit 7ec68e2d45b431a3a77458939622e6ea1e396d0e
Author: Samuel Just <sam.just@inktank.com>
Date:   Thu Mar 14 19:59:36 2013 -0700

    PG: ignore non MISSING pg query in ReplicaActive
    
    1) Replica sends notify
    2) Prior to processing notify, primary queues query to replica
    3) Primary processes notify and activates sending MOSDPGLog
    to replica.
    4) Primary does do_notifies at end of process_peering_events
    and sends to Query.
    5) Replica sees MOSDPGLog and activates
    6) Replica sees Query and asserts.
    
    In the above case, the Replica should simply ignore the old
    Query.
    
    Fixes: #4050
    Backport: bobtail
    Signed-off-by: Samuel Just <sam.just@inktank.com>
    Reviewed-by: Sage Weil <sage@inktank.com>
    (cherry picked from commit 8222cbc8f35c359a35f8381ad90ff0eed5615dac)

commit b279b148b38c755d2dcca6524d7b8a95daf123a0
Author: Samuel Just <sam.just@inktank.com>
Date:   Wed Mar 13 16:04:23 2013 -0700

    FileJournal: queue_pos \in [get_top(), header.max_size)
    
    If queue_pos == header.max_size when we create the entry
    header magic, the entry will be rejected at get_top() on
    replay.
    
    Fixes: #4436
    Backport: bobtail
    Signed-off-by: Samuel Just <sam.just@inktank.com>
    Reviewed-by: Sage Weil <sage@inktank.com>
    (cherry picked from commit de8edb732e3a5ce4471670e43cfe6357ae6a2758)

commit 56e0a129b0d9ad463f4300999ed2c15173cbe603
Author: Samuel Just <sam.just@inktank.com>
Date:   Thu Mar 14 18:52:02 2013 -0700

    OSD: expand_pg_num after pg removes
    
    Otherwise:
    1) expand_pg_num removes a splitting pg entry
    2) peering thread grabs pg lock and starts split
    3) OSD::consume_map grabs pg lock and starts removal
    
    At step 2), we run afoul of the assert(is_splitting)
    check in split_pgs.  This way, the would be splitting
    pg is marked as removed prior to the splitting state
    being updated.
    
    Backport: bobtail
    Fixes: #4449
    Signed-off-by: Samuel Just <sam.just@inktank.com>
    Reviewed-by: Sage Weil <sage@inktank.com>
    (cherry picked from commit f1b031b3cf195cf6df3d3c47c7d606fba63ed4c4)

commit d6e678a62950ccc16d010a76500705e176ba9c73
Author: Sage Weil <sage@inktank.com>
Date:   Mon Feb 11 06:23:54 2013 -0800

    osd: update snap collections for sub_op_modify log records conditionaly
    
    The only remaining caller is sub_op_modify().  If we do have a non-empty
    op transaction, we want to do this update, regardless of what we think
    last_backfill is (our notion may be not completely in sync with the
    primary).  In particular, our last_backfill may be the same object but
    a different snapid, but the primary disagrees and is pushing an op
    transaction through.
    
    Instead, update the collections if we have a non-empty transaction.
    
    Signed-off-by: Sage Weil <sage@inktank.com>
    Reviewed-by: Samuel Just <sam.just@inktank.com>
    (cherry picked from commit 31e911b63d326bdd06981ec4029ad71b7479ed70)

commit 51c2c4d7ccb25617967689d7e531eb145ee7011d
Author: Sage Weil <sage@inktank.com>
Date:   Sun Feb 10 17:02:45 2013 -0800

    osd: include snaps in pg_log_entry_t::dump()
    
    Signed-off-by: Sage Weil <sage@inktank.com>
    Reviewed-by: Samuel Just <sam.just@inktank.com>
    (cherry picked from commit 715d8717a0e8a08fbe97a3e7d3ffd33aa9529d90)

commit f0bf68ff500f7337103aef2a9d6d10c3597e410f
Author: Sage Weil <sage@inktank.com>
Date:   Sun Feb 10 16:59:48 2013 -0800

    osd: unconditionally encode snaps buffer
    
    Previously we would only encode the updated snaps vector for CLONE ops.
    This doesn't work for MODIFY ops generated by the snap trimmer, which
    may also adjust the clone collections.  It is also possible that other
    operations may need to populate this field in the future (e.g.,
    LOST_REVERT may, although it currently does not).
    
    Fixes: #4071, and possibly #4051.
    
    Signed-off-by: Sage Weil <sage@inktank.com>
    Reviewed-by: Samuel Just <sam.just@inktank.com>
    (cherry picked from commit 54b6dd924fea3af982f3d729150b6449f318daf2)

commit aa74fabd90b776e9938f3d59ea5ed04bd4027dbb
Author: Sage Weil <sage@inktank.com>
Date:   Sun Feb 10 10:57:12 2013 -0800

    osd: improve debug output on snap collections
    
    Signed-off-by: Sage Weil <sage@inktank.com>
    Reviewed-by: Samuel Just <sam.just@inktank.com>
    (cherry picked from commit 8b05492ca5f1479589bb19c1ce058b0d0988b74f)

commit a30903c6adaa023587d3147179d6038ad37ca520
Author: Samuel Just <sam.just@inktank.com>
Date:   Thu Mar 7 12:53:51 2013 -0800

    PG: check_recovery_sources must happen even if not active
    
    missing_loc/missing_loc_sources also must be cleaned up
    if a peer goes down during peering:
    
    1) pg is in GetInfo, acting is [3,1]
    2) we find object A on osd [0] in GetInfo
    3) 0 goes down, no new peering interval since it is neither up nor
    acting, but peer_missing[0] is removed.
    4) pg goes active and try to pull A from 0 since missing_loc did not get
    cleaned up.
    
    Backport: bobtail
    Fixes: #4371
    Signed-off-by: Samuel Just <sam.just@inktank.com>
    Reviewed-by: Greg Farnum <greg@inktank.com>
    (cherry picked from commit de22b186c497ce151217aecf17a8d35cdbf549bb)

commit 19f6df28337307ce6d8617524f88f222c0ffcab3
Author: Samuel Just <sam.just@inktank.com>
Date:   Tue Mar 5 15:49:26 2013 -0800

    HashIndex: _collection_list_partial must tolerate NULL next
    
    Backport: bobtail
    Fixes: #4379
    Signed-off-by: Samuel Just <sam.just@inktank.com>
    Reviewed-by: Greg Farnum <greg@inktank.com>
    (cherry picked from commit ce4432adc67dc2fc06dd21ea08e59d179496bcc6)

commit 1b5c1564bad3574db257cb17d48190958d870f51
Author: Samuel Just <sam.just@inktank.com>
Date:   Tue Mar 5 14:35:39 2013 -0800

    OSD: lock not needed in ~DeletingState()
    
    No further refs to the object can remain at this point.
    Furthermore, the callbacks might lock mutexes of their
    own.
    
    Backport: bobtail
    Fixes: #4378
    Signed-off-by: Samuel Just <sam.just@inktank.com>
    Reviewed-by: Greg Farnum <greg@inktank.com>
    (cherry picked from commit e4bf1bcab159d7c5b720f5da01877c0f67c16d16)

commit 50d2d4a77f339d489703c1125dc5c1159d08ca8a
Author: Samuel Just <sam.just@inktank.com>
Date:   Sun Mar 10 12:50:01 2013 -0700

    ReplicatedPG: don't leak reservation on removal
    
    Fixes: 4431
    Signed-off-by: Samuel Just <sam.just@inktank.com>
    Reviewed-by: Greg Farnum <greg@inktank.com>
    (cherry picked from commit 32bf131e0141faf407b5ff993f75f97516b27c12)
    
    Conflicts:
    
    	src/osd/ReplicatedPG.cc

commit c6b7c4be3fc44d065c29ea76b744b39272c2ba05
Author: Yehuda Sadeh <yehuda@inktank.com>
Date:   Tue Mar 12 12:56:01 2013 -0700

    rgw: set up curl with CURL_NOSIGNAL
    
    Fixes: #4425
    Backport: bobtail
    Apparently, libcurl needs that in order to be thread safe. Side
    effect is that if libcurl is not compiled with c-ares support,
    domain name lookups are not going to time out.
    Issue affected keystone.
    
    Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
    Reviewed-by: Greg Farnum <greg@inktank.com>
    (cherry picked from commit 88725316ddcfa02ff110e659f7a8131dc1ea2cfc)

commit 8c16a376a9cbfd6812824fb71e2cc9935e14e667
Author: Sage Weil <sage@inktank.com>
Date:   Fri Mar 8 08:56:44 2013 -0800

    osd: mark down connections from old peers
    
    Close out any connection with an old peer.  This avoids a race like:
    
    - peer marked down
    - we get map, mark down the con
    - they reconnect and try to send us some stuff
    - we share our map to tell them they are old and dead, but leave the con
      open
    ...
    - peer marks itself up a few times, eventually reuses the same port
    - sends messages on their fresh con
    - we discard because of our old con
    
    This could cause a tight reconnect loop, but it is better than wrong
    behavior.
    
    Other possible fixes:
     - make addr nonce truly unique (augment pid in nonce)
     - make a smarter 'disposable' msgr state (bleh)
    
    Signed-off-by: Sage Weil <sage@inktank.com>
    (cherry picked from commit 881e9d850c6762290f8be24da9e74b9dc112f1c9)

commit 1affc8b6e8cb82510aed524f90a6165c6edf7513
Author: Sage Weil <sage@inktank.com>
Date:   Fri Mar 8 08:53:40 2013 -0800

    osd/PG: rename require_same_or_newer_map -> is_same_or_newer_map
    
    This avoids confusion with the OSD method of the same name, and better
    matches what the function tests (and does not do).
    
    Signed-off-by: Sage Weil <sage@inktank.com>
    (cherry picked from commit ba7e815a18cad110525f228db1b3fe39e011409e)
    
    Conflicts:
    
    	src/osd/ReplicatedPG.cc

commit 7ca55d36c4007be5880102fa1ed314d3d66e9d87
Author: Sage Weil <sage@inktank.com>
Date:   Mon Mar 11 16:25:16 2013 -0700

    log: drop default 'log max recent' from 100k -> 10k
    
    Use less memory.
    
    Signed-off-by: Sage Weil <sage@inktank.com>
    (cherry picked from commit c021c5ccf0c063cccd7314964420405cea6406de)

commit 33ac786ddc14d0904f4835cbf684a51f4815d45b
Author: Jan Harkes <jaharkes@cs.cmu.edu>
Date:   Fri Mar 8 12:45:57 2013 -0500

    Fix radosgw actually reloading after rotating logs.
    
    The --signal argument to Debian's start-stop-daemon doesn't
    make it send a signal, but defines which signal should be send
    when --stop is specified.
    
    Signed-off-by: Jan Harkes <jaharkes@cs.cmu.edu>
    (cherry picked from commit 44f1cc5bc42f9bb6d5a386037408d2de17dc5413)

commit 6838b8db9a3d4ceaf121a261e2ded4af9bfeaeb9
Author: Josh Durgin <josh.durgin@inktank.com>
Date:   Wed Mar 6 17:42:03 2013 -0800

    common: reduce default in-memory logs for non-daemons
    
    The default of 100000 can result in hundreds of MBs of extra memory
    used. This was most obvious when using librbd with caching enabled,
    since there was a dout(0) accidentally left in the ObjectCacher.
    
    refs: #4352
    backport: bobtail
    Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
    (cherry picked from commit 7c208d2f8e3f28f4055a4ae51eceae892dcef1dc)

commit f80f64cf024bd7519d5a1fb2a5698db97a003ce8
Author: Sage Weil <sage@inktank.com>
Date:   Fri Feb 22 17:01:53 2013 -0800

    osd: allow (some) log trim when degraded, but not during recovery
    
    We allow some trim during degraded, although we keep more entries around to
    improve our chances of a restarting OSD of doing log-based recovery.
    
    Still disallow during recovery...
    
    Signed-off-by: Sage Weil <sage@inktank.com>
    Reviewed-by: Samuel Just <sam.just@inktank.com>
    (cherry picked from commit 6d89b34e5608c71b49ef33ab58340e90bd8da6e4)

commit e502a65da1779564059e74c09ab87aca1d901bd2
Author: Sage Weil <sage@inktank.com>
Date:   Mon Feb 25 15:33:35 2013 -0800

    osd: restructure calc_trim
    
    No functional change, except that we log more debug, yay!
    
    Signed-off-by: Sage Weil <sage@inktank.com>
    Reviewed-by: Samuel Just <sam.just@inktank.com>
    (cherry picked from commit 86df164d04f6e31a0f20bbb94dbce0599c0e8b3d)

commit 3a3123033d7d9e1574be3fb18c807eeb0f4678eb
Author: Sage Weil <sage@inktank.com>
Date:   Fri Feb 22 16:48:02 2013 -0800

    osd: allow pg log trim during (non-classic) scrub
    
    Chunky (and deep) scrub do not care about PG log trimming.  Classic scrub
    still does.
    
    Deep scrub can take a long time, so not trimming the log during that period
    may eat lots of RAM; avoid that!
    
    Might fix: #4179
    Signed-off-by: Sage Weil <sage@inktank.com>
    Reviewed-by: Samuel Just <sam.just@inktank.com>
    (cherry picked from commit 0ba8db6b664205348d5499937759916eac0997bf)

commit cde0f224dac21b2e17e06d396ac52e18034a4262
Author: Sage Weil <sage@inktank.com>
Date:   Thu Feb 28 12:46:00 2013 -0800

    msgr: drop messages on cons with CLOSED Pipes
    
    Back in commit 6339c5d43974f4b495f15d199e01a141e74235f5, we tried to make
    this deal with a race between a faulting pipe and new messages being
    queued.  The sequence is
    
    - fault starts on pipe
    - fault drops pipe_lock to unregister the pipe
    - user (objecter) queues new message on the con
    - submit_message reopens a Pipe (due to this bug)
    - the message managed to make it out over the wire
    - fault finishes faulting, calls ms_reset
    - user (objecter) closes the con
    - user (objecter) resends everything
    
    It appears as though the previous patch *meant* to drop *m on the floor in
    this case, which is what this patch does.  And that fixes the crash I am
    hitting; see #4271.
    
    Signed-off-by: Sage Weil <sage@inktank.com>
    Reviewed-by: Greg Farnum <greg@inktank.com>
    (cherry picked from commit 0f42eddef5da6c1babe9ed51ceaa3212a42c2ec4)

commit 2fa2546c90e7a6eab9a3de7ab4e4271f8b25bad6
Author: Concubidated <tyler.brekke@inktank.com>
Date:   Fri Mar 8 13:44:39 2013 -0800

    Fix output of 'ceph osd tree --format=json'
    
    Signed-off-by: Tyler Brekke <tyler.brekke@inktank.com>
    (cherry picked from commit 9bcba944c6586ad5f007c0a30e69c6b5a886510b)

commit b70e2c270b9eb3fce673b7e51b527ebf88214f14
Author: Sam Lang <sam.lang@inktank.com>
Date:   Tue Feb 12 11:32:29 2013 -0600

    deb:  Add ceph-coverage to ceph-test deb package
    
    Teuthology uses the ceph-coverage script extensively
    and expects it to be installed by the ceph task.  Add
    the script to the ceph-test debian package so that it
    gets installed for that use case.
    
    Signed-off-by: Sam Lang <sam.lang@inktank.com>
    (cherry picked from commit 376cca2d4d4f548ce6b00b4fc2928d2e6d41038f)

commit ca9aac785eb9e2d1ee955792d2f4d1d911727fb3
Author: Yehuda Sadeh <yehuda@inktank.com>
Date:   Fri Feb 22 15:04:37 2013 -0800

    rgw: set attrs on various list bucket xml results (swift)
    
    Fixes: #4247
    The list buckets operation was missing some attrs on the different
    xml result entities. This fixes it.
    
    Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
    (cherry picked from commit 4384e59ad046afc9ec53a2d2f1fff6a86e645505)

commit e39660c901756d5e722308e72a9d8ee4893f70f7
Author: Yehuda Sadeh <yehuda@inktank.com>
Date:   Fri Feb 22 15:02:02 2013 -0800

    formatter: add the ability to dump attrs in xml entities
    
    xml entities may have attrs assigned to them. Add the ability
    to set them. A usage example:
    
    formatter->open_array_section_with_attrs("container",
         FormatterAttrs("name", "foo", NULL));
    
    This will generate the following xml entity:
    <container name="foo">
    
    Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
    Reviewed-by: Greg Farnum <greg@inktank.com>
    (cherry picked from commit 7cb6ee28073824591d8132a87ea09a11c44efd66)
    
    Conflicts:
    	src/common/Formatter.cc

commit 0304b85e5c12c30459b94ec5d332dfaa04d20d5b
Author: Yehuda Sadeh <yehuda@inktank.com>
Date:   Wed Mar 6 19:32:21 2013 -0800

    rgw: don't iterate through all objects when in namespace
    
    Fixes: #4363
    Backport: argonaut, bobtail
    When listing objects in namespace don't iterate through all the
    objects, only go though the ones that starts with the namespace
    prefix
    
    Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
    (cherry picked from commit 6669e73fa50e3908ec825ee030c31a6dbede6ac0)

commit d0dafaa77f96fa91785df0796806ea07fd93f319
Author: Josh Durgin <josh.durgin@inktank.com>
Date:   Thu Feb 28 12:13:45 2013 -0800

    ObjectCacher: fix debug log level in split
    
    Level 0 should never be used for this kind of debugging.
    
    Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
    Reviewed-by: Dan Mick <dan.mick@inktank.com>
    (cherry picked from commit cb3ee33532fb60665f39f6ccb1d69d67279fd5e1)

commit ff014ed26bfa4fbf40acffec4eb4805b95a9982c
Author: Dan Mick <dmick@danceorelse.org>
Date:   Thu Jan 24 13:38:25 2013 -0800

    rados: remove unused "check_stdio" parameter
    
    Signed-off-by: Dan Mick <dan.mick@inktank.com>
    (cherry picked from commit bb860e49a7faeaf552538a9492ef0ba738c99760)

commit e89884da9f76b713372a79b772ba3cc2f3b03048
Author: Sage Weil <sage@inktank.com>
Date:   Wed Jan 23 21:31:11 2013 -0800

    rados: obey op_size for 'get'
    
    Otherwise we try to read the whole object in one go, which doesn't bode
    well for large objects (either non-optimal or simply broken).
    
    Signed-off-by: Sage Weil <sage@inktank.com>
    Reviewed-by: Dan Mick <dan.mick@inktank.com>
    (cherry picked from commit 234becd3447a679a919af458440bc31c8bd6b84f)

commit ee943c8bcf36f1e2218d8e25edfa38ec5fe4bec2
Author: Samuel Just <sam.just@inktank.com>
Date:   Wed Feb 27 16:58:45 2013 -0800

    FileJournal::wrap_read_bl: adjust pos before returning
    
    Otherwise, we may feed an offset past the end of the journal to
    check_header in read_entry and incorrectly determine that the entry is
    corrupt.
    
    Fixes: 4296
    Backport: bobtail
    Backport: argonaut
    Reviewed-by: Sage Weil <sage@inktank.com>
    Signed-off-by: Samuel Just <sam.just@inktank.com>
    (cherry picked from commit 5d54ab154ca790688a6a1a2ad5f869c17a23980a)

commit 8c6f52215240f48b5e4d5bb99a5f2f451e7ce70a
Author: Sage Weil <sage@inktank.com>
Date:   Wed Jan 16 13:14:00 2013 -0800

    osd: leave osd_lock locked in shutdown()
    
    No callers expect the lock to be dropped.
    
    Fixes: #3816
    Signed-off-by: Sage Weil <sage@inktank.com>
    (cherry picked from commit 98a763123240803741ac9f67846b8f405f1b005b)

commit a324d999804b4648f245ee36b3bd611b3d139d5d
Author: Sage Weil <sage@inktank.com>
Date:   Tue Feb 26 14:07:12 2013 -0800

    msg: fix entity_addr_t::is_same_host() for IPv6
    
    We weren't checking the memcmp return value properly!  Aie...
    
    Backport: bobtail
    Signed-off-by: Sage Weil <sage@inktank.com>
    (cherry picked from commit c8dd2b67b39a8c70e48441ecd1a5cc3c6200ae97)

commit 79d68ae8581141c3fb4cfafd76d5111ff009b762
Author: Sage Weil <sage@inktank.com>
Date:   Sun Feb 17 22:35:50 2013 -0800

    osd: requeue pg waiters at the front of the finished queue
    
    We could have a sequence like:
    
    - op1
    - notify
    - op2
    
    in the finished queue.  Op1 gets put on waiting_for_pg, the notify
    creates the pg and requeues op1 (and the end), op2 is handled, and
    finally op1 is handled.  That breaks ordering; see #2947.
    
    Instead, when we wake up a pg, queue the waiting messages at the front
    of the dispatch queue.
    
    Signed-off-by: Sage Weil <sage@inktank.com>
    Reviewed-by: Samuel Just <sam.just@inktank.com>
    (cherry picked from commit 56c5a07708d52de1699585c9560cff8b4e993d0a)

commit 79c4e7e91becc497843d96251776bdc176706aa0
Author: Sage Weil <sage@inktank.com>
Date:   Sun Feb 17 20:49:52 2013 -0800

    osd: pull requeued requests off one at a time
    
    Pull items off the finished queue on at a time.  In certain cases, an
    event may result in new items betting added to the finished queue that
    will be put at the *front* instead of the back.  See latest incarnation
    of #2947.
    
    Note that this is a significant changed in behavior in that we can
    theoretically starve if an event keeps resulting in new events getting
    generated.  Beware!
    
    Signed-off-by: Sage Weil <sage@inktank.com>
    Reviewed-by: Samuel Just <sam.just@inktank.com>
    (cherry picked from commit f1841e4189fce70ef5722d508289e516faa9af6a)

commit 3a6ce5d0355beaa56199465e94666cae40bd8da1
Author: Sage Weil <sage@inktank.com>
Date:   Thu Jan 17 22:00:42 2013 -0800

    mds: open mydir after replay
    
    In certain cases, we may replay the journal and not end up with the
    dirfrag for mydir open.  This is fine--we just need to open it up and
    fetch it below.
    
    Signed-off-by: Sage Weil <sage@inktank.com>
    (cherry picked from commit e51299fbce6bdc3d6ec736e949ba8643afc965ec)

commit 36ed407e0f939a9bca57c3ffc0ee5608d50ab7ed
Author: Greg Farnum <greg@inktank.com>
Date:   Thu Feb 21 09:21:01 2013 -0800

    mds: use inode_t::layout for dir layout policy
    
    Remove the default_file_layout struct, which was just a ceph_file_layout,
    and store it in the inode_t.  Rip out all the annoying code that put this
    on the heap.
    
    To aid in this usage, add a clear_layout() function to inode_t.
    
    Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
    Signed-off-by: Greg Farnum <greg@inktank.com>
    Reviewed-by: Sage Weil <sage@inktank.com>

commit b4fd196cae5ab76aeb8f82f0581d0a6a2133a3ed
Author: Sage Weil <sage@inktank.com>
Date:   Sun Jan 20 21:53:37 2013 -0800

    mds: parse ceph.*.layout vxattr key/value content
    
    Use qi to parse a strictly formatted set of key/value pairs.  Be picky
    about whitespace.  Any subset of recognized keys is allowed.  Parse the
    same set of keys as the ceph.*.layout.* vxattrs.
    
    Signed-off-by: Sage Weil <sage@inktank.com>
    (cherry picked from commit 5551aa5b3b5c2e9e7006476b9cd8cc181d2c9a04)

commit 5f92adca26b509aed5655e85ad53174fa7b574b9
Author: Yehuda Sadeh <yehuda@inktank.com>
Date:   Mon Feb 18 09:10:43 2013 -0800

    rgw: fix multipart uploads listing
    
    Fixes: #4177
    Backport: bobtail
    Listing multipart uploads had a typo, and was requiring the
    wrong resource (uploadId instead of uploads).
    
    Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
    Reviewed-by: Greg Farnum <greg@inktank.com>
    (cherry picked from commit db99fb4417b87301a69cb37b00c35c838b77197e)

commit a44df9343594099fecb3897df393249d3d1992e2
Author: Yehuda Sadeh <yehuda@inktank.com>
Date:   Fri Feb 15 10:22:54 2013 -0800

    rgw: don't copy object when it's copied into itself
    
    Fixes: #4150
    Backport: bobtail
    
    When object copied into itself, object will not be fully copied: tail
    reference count stays the same, head part is rewritten.
    
    Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
    Reviewed-by: Greg Farnum <greg@inktank.com>
    (cherry picked from commit 34f885be536d0ac89c10fd29b1518751d2ffc547)

commit 91d6be8353259ca30dc87062422e9ae334c3c344
Author: Samuel Just <sam.just@inktank.com>
Date:   Tue Feb 19 10:49:33 2013 -0800

    PG: remove weirdness log for last_complete < log.tail
    
    In the case of a divergent object prior to log.tail,
    last_complete may end up before log.tail.
    
    Backport: bobtail
    Fixes #4174
    Signed-off-by: Samuel Just <sam.just@inktank.com>
    Reviewed-by: Sage Weil <sage@inktank.com>
    (cherry picked from commit dbadb3e2921297882c5836c67ca32bb8ecdc75db)
    
    Conflicts:
    
    	src/osd/PG.cc

commit 7889c5412deaab1e7ba53d66814d0f25659229c1
Author: James Page <james.page@ubuntu.com>
Date:   Mon Feb 18 16:24:54 2013 +0000

    Strip any trailing whitespace from rbd showmapped
    
    More recent versions of ceph append a bit of whitespace to the line
    after the name of the /dev/rbdX device; this causes the monitor check
    to fail as it can't find the device name due to the whitespace.
    
    This fix excludes any characters after the /dev/rbdN match.
    (cherry picked from commit ad84ea07cac5096de38b51b8fc452c99f016b8d8)

commit 42fba772c3c0e2396124a9d6ec39297dd9af4cc9
Merge: 290a352 ad6655e
Author: Sage Weil <sage@newdream.net>
Date:   Sun Feb 17 21:52:55 2013 -0800

    Merge pull request #64 from dalgaaf/wip-bobtail-memleaks
    
    cherry-pick some memleak fixes from master to bobtail

commit ad6655e976916d3b62e0c91fd469eeb49fe7da3e
Author: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
Date:   Fri Feb 8 17:17:59 2013 +0100

    rgw/rgw_rest.cc: fix 4K memory leak
    
    Fix 4K memory leak in case RGWClientIO::read() fails in
    read_all_chunked_input().
    
    Error from cppcheck was:
    Checking src/rgw/rgw_rest.cc...
    [src/rgw/rgw_rest.cc:688]: (error) Memory leak: data
    
    Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
    (cherry picked from commit 89df090e04ef9fc5aae29122df106b0347786fab)

commit 3de692753c28ddbeb3c86b51466be16e94c9c458
Author: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
Date:   Fri Feb 8 17:14:19 2013 +0100

    SyntheticClient.cc: fix some memory leaks in the error handling
    
    Fix some memory leaks in case of error handling due to failed
    client->open() calls.
    
    Error from cppcheck was:
    [src/client/SyntheticClient.cc:1980]: (error) Memory leak: buf
    [src/client/SyntheticClient.cc:2040]: (error) Memory leak: buf
    [src/client/SyntheticClient.cc:2090]: (error) Memory leak: buf
    (cherry picked from commit f0ba80756d1c3c313014ad7be18191981fb545be)

commit f19d228c6a49222659c769099aaa4e755b80331d
Author: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
Date:   Fri Feb 8 16:57:20 2013 +0100

    rgw/rgw_xml.cc: fix realloc memory leak in error case
    
    Fix error from cppcheck:
    
    [src/rgw/rgw_xml.cc:212]: (error) Common realloc mistake: 'buf'
      nulled but not freed upon failure
    
    Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
    (cherry picked from commit d48cc789ea075ba2745754035640ada4131b2119)

commit b0c6be95b03d9f3dd2badcdcff359ae7bc9684f4
Author: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
Date:   Fri Feb 8 16:54:33 2013 +0100

    os/FileStore.cc: fix realloc memory leak in error case
    
    Fix error from cppcheck:
    
    [src/os/FileStore.cc:512]: (error) Common realloc mistake: 'fiemap'
      nulled but not freed upon failure
    
    Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
    (cherry picked from commit c92a0f552587a232f66620170660d6b2ab6fb3a5)

commit f3e5cedbc11005701ac0a8e70909a6372cd2fe6f
Author: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
Date:   Fri Feb 8 16:49:36 2013 +0100

    common/fiemap.cc: fix realloc memory leak
    
    Fix error from cppcheck:
    
    [src/common/fiemap.cc:73]: (error) Common realloc mistake: 'fiemap'
      nulled but not freed upon failure
    
    Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
    (cherry picked from commit f26f1470e7af36fa1eb8dc59c8a7c62c3c3a22ba)

commit 290a352c3f9e241deac562e980ac8c6a74033ba6
Author: Sage Weil <sage@inktank.com>
Date:   Thu Feb 14 11:37:57 2013 -0800

    osd/OSDCap: add unit test for parsing pools/objects with _ and -
    
    Hunting #4122, where a user saw
    
    2013-02-13 19:39:25.467916 7f766fdb4700 10 osd.0 10  session 0x2c8cc60 client.libvirt has caps osdcap[grant(object_prefix rbd^@children  class-read),grant(pool libvirt^@pool^@test rwx)] 'allow class-read object_prefix rbd_children, allow pool libvirt-pool-test rwx'
    
    Signed-off-by: Sage Weil <sage@inktank.com>
    (cherry picked from commit 2ce28ef1d7f95e71e1043912dfa269ea3b0d1599)
    (cherry picked from commit a6534bc8a0247418d5263b765772d5266f99229c)

commit f47916353547c77a65c7b70e609a9096f09c6d28
Author: Sage Weil <sage@inktank.com>
Date:   Thu Feb 14 15:39:43 2013 -0800

    osd/OSDCap: tweak unquoted_word parsing in osd caps
    
    Newer versions of spirit (1.49.0-3.1ubuntu1.1 in quantal, in particular)
    dislike the construct with alnum and replace the - and _ with '\0' in the
    resulting string.
    
    Fixes: #4122
    Backport: bobtail
    Signed-off-by: Sage Weil <sage@inktank.com>
    Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
    (cherry picked from commit 6c504d96c1e4fbb67578fba0666ca453b939c218)