path: root/doc/changelog/v0.67.6.txt

commit c71f8a60cf49fb3d4efb5ea75b2366cd7f167497 (tag: refs/tags/v0.67.6)
Author: Alfredo Deza <alfredo.deza@inktank.com>
Date:   Thu Feb 13 20:20:42 2014 +0000

    release build 67.6
    
    Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>

commit 1c42a6e34bb8e318e001384270cda003add946a7
Merge: 3e473d4 c5e12f9
Author: Sage Weil <sage@inktank.com>
Date:   Thu Feb 13 06:55:37 2014 -0800

    Merge pull request #1232 from ceph/dumpling-7334
    
    backport ceph-disk improvements to dumpling
    
    http://pulpito.ceph.com/ubuntu-2014-02-12_16:52:33-ceph-deploy-dumpling-7334-testing-basic-plana/

commit 3e473d4237697872498094b41fed19b414f8d3c4
Author: David Zafman <david.zafman@inktank.com>
Date:   Fri Sep 27 16:23:09 2013 -0700

    common,os: Remove filestore_xattr_use_omap option
    
    Now we operate just like when this was set to true
    
    Fixes: #6143
    
    Signed-off-by: David Zafman <david.zafman@inktank.com>
    (cherry picked from commit dc0dfb9e01d593afdd430ca776cf4da2c2240a20)

commit c5e12f94efc07f0ff174c0f19494ee2e06d88dfa
Author: Alfredo Deza <alfredo@deza.pe>
Date:   Wed Feb 12 16:43:59 2014 -0500

    add support for absence of PATH
    
    Note that this commit is actually bisecting the changes from
    Loic Dachary that touch ceph-disk only (ad515bf). As that changeset
    also touches other files it causes conflicts that are not resolvable
    for backporting it to dumpling.
    
    Signed-off-by: Alfredo Deza <alfredo@deza.pe>

commit c531d75f7e86cee619e44c247298b50c37b81539
Author: Sage Weil <sage@inktank.com>
Date:   Mon Sep 9 22:27:23 2013 -0700

    ceph-disk: make initial journal files 0 bytes
    
    The ceph-osd will resize journal files up and properly fallocate() them
    so that the blocks are preallocated and (hopefully) contiguous.  We
    don't need to do it here too, and getting fallocate() to work from
    python is a pain in the butt.
    
    Fixes: #5981
    Signed-off-by: Sage Weil <sage@inktank.com>
    (cherry picked from commit a786ad773cd33880075f1deb3691528d1afd03ec)

commit 836ead3e42491597c192c63d3171021c5daecbc0
Author: Alfredo Deza <alfredo@deza.pe>
Date:   Mon Feb 10 15:07:55 2014 -0500

    alert the user about error messages from partx
    
    Signed-off-by: Alfredo Deza <alfredo@deza.pe>
    (cherry picked from commit 9bcc42a3e6b08521694b5c0228b2c6ed7b3d312e)

commit cff3547c60dc6d164f9a95eaa2706633e32756d0
Author: Alfredo Deza <alfredo@deza.pe>
Date:   Fri Feb 7 11:55:01 2014 -0500

    use partx for red hat or centos instead of partprobe
    
    Signed-off-by: Alfredo Deza <alfredo@deza.pe>
    (cherry picked from commit 42900ff9da9f5adcac239a84ebf4d2e407c29699)

commit 5ebd7c4520a9805f985cab1a0ba9311e19efa041
Author: Josh Durgin <josh.durgin@inktank.com>
Date:   Tue Jan 28 17:26:58 2014 -0800

    ceph-disk: run the right executables from udev
    
    When run by the udev rules, PATH is not defined. Thus,
    ceph-disk-activate relies on its which() function to locate the
    correct executable.  The which() function used os.defpath if none was
    set, and this worked for anything using it.
    
    ad6b4b4b08b6ef7ae8086f2be3a9ef521adaa88c added a new default value to
    PATH, so only /usr/bin was checked by callers that did not use
    which(). This resulted in the mount command not being found when
    ceph-disk-activate was run by udev, and thus osds failing to start
    after being prepared by ceph-deploy.
    
    Make ceph-disk consistently use the existing helpers (command() and
    command_check_call()) that use which(), so lack of PATH does not
    matter. Simplify _check_output() to use command(),
    another wrapper around subprocess.Popen.
    
    Fixes: #7258
    Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
    (cherry picked from commit d7b0c7faafd37e4ae8a1680edfa60c22b419cbd8)

commit 4a7c4e0a81753dfb811396e7bfa1df72135ee12b
Author: Loic Dachary <loic@dachary.org>
Date:   Fri Jan 3 16:06:55 2014 +0100

    ceph-disk: implement --sysconfdir as /etc/ceph
    
    Replace hardcoded /etc/ceph with the SYSCONFDIR global variable and
    implement the --sysconfdir option to override the default value.
    
    Signed-off-by: Loic Dachary <loic@dachary.org>
    (cherry picked from commit a71025d33621257b6fd6632516cfed2849ff1637)

commit c8b9798a69bd1655a7131c7bce8a14c2c74aba0c
Author: Loic Dachary <loic@dachary.org>
Date:   Fri Jan 3 16:03:09 2014 +0100

    ceph-disk: implement --statedir as /var/lib/ceph
    
    Replace hardcoded /var/lib/ceph with the STATEDIR global variable and
    implement the --statedir option to override the default value.
    
    Signed-off-by: Loic Dachary <loic@dachary.org>
    (cherry picked from commit ca713f48ae7a1fece2869f1a1c97d23ab33fb441)

commit 79564874be35d4414092fbb80df3214d7f282b1c
Author: Loic Dachary <loic@dachary.org>
Date:   Wed Jan 1 22:05:03 2014 +0100

    ceph-disk: add copyright notice
    
    Signed-off-by: Loic Dachary <loic@dachary.org>
    (cherry picked from commit 51ee3c04777aaf6b9609dde9bc318b5c66c70787)

commit c4b66bbdf09a8bb85ed14fbb643f40f87345463b
Author: Loic Dachary <loic@dachary.org>
Date:   Wed Jan 1 22:11:30 2014 +0100

    ceph-disk: create the data directory if it does not exist
    
    Instead of failing if the OSD data directory does not exist, create
    it. Only do so if the data directory is not enforced to be a device via
    the use of the --data-dev flag. The directory is not recursively created.
    
    Signed-off-by: Loic Dachary <loic@dachary.org>
    (cherry picked from commit 306b099ab093bfac466d68fe1cb87367bc01e577)

commit c473c65d4b6235fe20a97bd1d4741c12f86858e8
Author: Loic Dachary <loic@dachary.org>
Date:   Tue Dec 31 00:03:42 2013 +0100

    ceph-disk: run ceph-osd when --mark-init=none
    
    Signed-off-by: Loic Dachary <loic@dachary.org>
    (cherry picked from commit 0fcc081858fae4febbb6a613a93cfbbcedd5a320)

commit 6ea59a3a0c38c9629036442aff652489eecce51a
Author: Loic Dachary <loic@dachary.org>
Date:   Mon Dec 30 23:57:39 2013 +0100

    ceph-disk: implement --mark-init=none
    
    It is meant to be used when preparing and activating a directory that is
    not to be used with init. No file is created to identify the init
    system, no symbolic link is made to the directory in /var/lib/ceph
    and the init scripts are not called.
    
    Signed-off-by: Loic Dachary <loic@dachary.org>
    (cherry picked from commit e773b68f4c89ac56b425c710d7dcdc3d74a92926)

commit 3aae5b764c5698c665a708a3f6f2243dfee1a28d
Author: Loic Dachary <loic@dachary.org>
Date:   Wed Jan 1 22:07:57 2014 +0100

    ceph-disk: fsid is a known configuration option
    
    Use get_conf_with_default instead of get_conf because fsid is a known
    ceph configuration option. It allows overriding via CEPH_ARGS which is
    convenient for testing. Only options that are not found in config_opts.h
    are fetch via get_conf.
    
    Signed-off-by: Loic Dachary <loic@dachary.org>
    (cherry picked from commit b65eb377f5e93ea85644e4c0939365fd7ac36072)

commit a33ad6d150206232806ee466f301e41f15c631e9
Author: Loic Dachary <loic@dachary.org>
Date:   Wed Jan 1 22:06:29 2014 +0100

    ceph-disk: use CalledProcessError.returncode
    
    CalledProcessError has no errno data member
    
    Signed-off-by: Loic Dachary <loic@dachary.org>
    (cherry picked from commit 97f516a1ddfb2d014e1f7e762c4155e4b9bcb90b)

commit 6ae96d3ce757b05daeb7c1437778149b07248d47
Author: Loic Dachary <loic@dachary.org>
Date:   Wed Jan 1 22:13:15 2014 +0100

    ceph-disk: display the command output on OSD creation failure
    
    The string form of a CalledProcessError instance does not include the
    output datamember. Add it to the Error exception for debug purposes.
    
    Signed-off-by: Loic Dachary <loic@dachary.org>
    (cherry picked from commit d09af0fa50f322c9e59765f3badd497f5ca184d4)

commit deefb3a05788b308631812c652e9eccaeb999d7f
Author: Loic Dachary <loic@dachary.org>
Date:   Mon Dec 30 23:07:27 2013 +0100

    ceph-disk: which() uses PATH first
    
    Instead of relying on a hardcoded set of if paths. Although this has the
    potential of changing the location of the binary being used by ceph-disk
    on an existing installation, it is currently only used for sgdisk. It
    could be disruptive for someone using a modified version of sgdisk but
    the odds of this happening are very low.
    
    Signed-off-by: Loic Dachary <loic@dachary.org>
    (cherry picked from commit 2b935bbf60bafb6dd488c0eb30f156fce1b9d197)

commit 06bbea187293ac52aefbcf4e98c162bb3286341d
Author: Loic Dachary <loic@dachary.org>
Date:   Mon Dec 30 22:48:46 2013 +0100

    ceph-disk: add --prepend-to-path to control execution
    
    /usr/bin is hardcoded in front of some ceph programs which makes it
    impossible to control where they are located via the PATH.
    
    The hardcoded path cannot be removed altogether because it will most
    likely lead to unexpected and difficult to diagnose problems for
    existing installations where the PATH finds the program elsewhere.
    
    The --prepend-to-path flag is added and defaults to /usr/bin : it prepends
    to the PATH environment variable. The hardcoded path is removed
    and the PATH will be used: since /usr/bin is searched first, the
    legacy behavior will not change.
    
    Signed-off-by: Loic Dachary <loic@dachary.org>
    (cherry picked from commit ad6b4b4b08b6ef7ae8086f2be3a9ef521adaa88c)

commit ccf182241dc04c79e2377b6ba2f060dbfe6dddae
Author: Loic Dachary <loic@dachary.org>
Date:   Mon Dec 30 17:15:29 2013 +0100

    ceph-disk: make exception handling terse in main_activate_journal
    
    Signed-off-by: Loic Dachary <loic@dachary.org>
    (cherry picked from commit 908348b8047e8577ecf9133f2683f91423694416)

commit d60e952f3c00d0e551b04c0d6486b11d9761be35
Author: Loic Dachary <loic@dachary.org>
Date:   Mon Dec 30 17:13:47 2013 +0100

    ceph-disk: do not hide main_activate() exceptions
    
    Signed-off-by: Loic Dachary <loic@dachary.org>
    (cherry picked from commit 55ca7bb2da73f1be1293710a635cfea42abd7682)

commit 827edb93e3971fc7549fe831894776c6bf163346
Author: Loic Dachary <loic@dachary.org>
Date:   Mon Dec 30 17:11:30 2013 +0100

    ceph-disk: fix activate() indent
    
    Signed-off-by: Loic Dachary <loic@dachary.org>
    (cherry picked from commit 324804a81c37ff89f2488e2ba106033c0e6e119e)

commit 27988cf9ffedf3289099e51f1e7c3f5b8d79d667
Author: Loic Dachary <loic@dachary.org>
Date:   Mon Dec 30 17:09:35 2013 +0100

    ceph-disk: remove noop try:
    
    Signed-off-by: Loic Dachary <loic@dachary.org>
    (cherry picked from commit de0050596b5f56863c3486c1cd5e7ffea62e3d00)

commit 75bdb5430d9403bed8bd2ccd6a9c5a30d63a189e
Author: Loic Dachary <loic@dachary.org>
Date:   Mon Dec 30 16:57:36 2013 +0100

    ceph-disk: fix Error() messages formatting
    
    Mainly using % instead of ,
    
    Signed-off-by: Loic Dachary <loic@dachary.org>
    (cherry picked from commit b82ccfbfa786cd5436b48ec38276c5a48028ce1d)

commit 5df1eaa55bc7ebe68473831d06cf9819fa39baa6
Author: Loic Dachary <loic@dachary.org>
Date:   Mon Dec 30 12:26:20 2013 +0100

    ceph-disk: prepare --data-dir must not override files
    
    ceph-disk does nothing when given a device that is already prepared. If
    given a directory that already contains a successfully prepared OSD, it
    will however override it.
    
    Instead of overriding the files in the osd data directory, return
    immediately if the magic file exists. Make it so the magic file is
    created last to accurately reflect the success of the OSD preparation.
    
    Signed-off-by: Loic Dachary <loic@dachary.org>
    (cherry picked from commit 7dfe550ce18623cde4ae43a2416e31ef81381ab9)

commit 2e8496a6e0b74ff351bc2a4f138ddc665d8c30ca
Author: Loic Dachary <loic@dachary.org>
Date:   Sun Dec 15 15:34:17 2013 +0100

    ceph-disk: zap needs at least one device
    
    If given no argument, ceph-disk zap should display the usage instead of
    silently doing nothing. Silence can be confused with "I zapped all the
    disks".
    
    http://tracker.ceph.com/issues/6981 fixes #6981
    
    Signed-off-by: Loic Dachary <loic@dachary.org>
    (cherry picked from commit 07888ef3fd4440332c8287d0faa9f23a32cf141c)

commit 0fe2ac47a82d3caa9dca2a9009b68a14b577e35c
Author: Alfredo Deza <alfredo@deza.pe>
Date:   Fri Dec 13 12:06:25 2013 -0500

    use the new get_command helper in check_call
    
    Signed-off-by: Alfredo Deza <alfredo@deza.pe>
    (cherry picked from commit 897dfc113fe3b86f3dda53172933bfd4f8089869)

commit 8fc9f1b791248f08a311d619addc4465cac298f9
Author: Alfredo Deza <alfredo@deza.pe>
Date:   Thu Dec 12 11:16:38 2013 -0500

    use the absolute path for executables if found
    
    Signed-off-by: Alfredo Deza <alfredo@deza.pe>
    (cherry picked from commit a9334a1c8c6681305e76b361377864d0dd1e3d34)

commit a2de2f718c9e341c0b6ec2ebffdf03d0f9da0fbd
Author: Alfredo Deza <alfredo@deza.pe>
Date:   Thu Dec 12 10:26:05 2013 -0500

    remove trailing semicolon
    
    Signed-off-by: Alfredo Deza <alfredo@deza.pe>
    (cherry picked from commit 43561f791607f5fd6f03d5421e1f30a29fb4194e)

commit 2cd22b5d39760ac4b5baad0b05c676b833ae9213
Author: Alfredo Deza <alfredo@deza.pe>
Date:   Wed Dec 11 15:41:45 2013 -0500

    replace sgdisk subprocess calls with a helper
    
    Signed-off-by: Alfredo Deza <alfredo@deza.pe>
    (cherry picked from commit e19e38012bc4579054f63865e682c8c3a7829c7b)

commit e8cc27cc8eb981c71b8cafb58964317687ca5787
Author: Jonathan Davies <jonathan.davies@canonical.com>
Date:   Tue Dec 3 21:26:43 2013 +0000

    Call --mbrtogpt on journal run of sgdisk should the drive require a GPT table.
    
    Signed-off-by: Jonathan Davies <jonathan.davies@canonical.com>
    Reviewed-by: Sage Weil <sage@inktank.com>
    Reviewed-by: Loic Dachary <loic@dachary.org>
    (cherry picked from commit 35011e0b01d65e4c001876882d597451f2028345)

commit 3942a93fae3331b2d08f93181eb2c9cf90b50440
Author: Loic Dachary <loic@dachary.org>
Date:   Thu Nov 21 13:21:22 2013 +0100

    ceph-disk: blacklist /dev/fd0
    
    blkid -s TYPE /dev/fd0 has been verified to hang forever on a
    H8DMR-82 supermicro motherboard running
    
    3.8.0-33-generic #48~precise1-Ubuntu SMP Thu Oct 24 16:28:06 UTC 2013
    x86_64
    
    It is unlikely that ceph will ever be used on floppy disks, they
    can be blacklisted.
    
    http://tracker.ceph.com/issues/6827 fixes: #6827
    
    Signed-off-by: Loic Dachary <loic@dachary.org>
    (cherry picked from commit 65701978715237ff5a4c68573c0696fd9d438e4f)

commit ccc6b0c5e6d47eefef5cf1fa17bf69c47c9459c4
Author: git-harry <git-harry@live.co.uk>
Date:   Thu Oct 3 10:32:50 2013 +0100

    Make fsid comparison case-insensitive
    
    get_fsid and find_cluster_by_uuid are modified so ceph-disk activate and
    ceph-disk activate-all will work if the fsid uses uppercase characters.
    
    Signed-off-by: Harry Harrington <git-harry@live.co.uk>
    (cherry picked from commit 22f8325dbfce7ef2e97bf015c0f8bba53e75dfe9)

commit 367cf1bbf86233eb20ff2304e7d6caab77b84fcc
Author: Josh Durgin <josh.durgin@inktank.com>
Date:   Tue Feb 11 10:14:36 2014 -0800

    librbd: remove limit on number of objects in the cache
    
    The number of objects is not a significant indicated of when data
    should be written out for rbd. Use the highest possible value for
    number of objects and just rely on the dirty data limits to trigger
    flushing. When the number of objects is low, and many start being
    flushed before they accumulate many requests, it hurts average request
    size and performance for many concurrent sequential writes.
    
    Fixes: #7385
    Backport: emperor, dumpling
    Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
    (cherry picked from commit 0559d31db29ea83bdb6cec72b830d16b44e3cd35)

commit 4856fc2950442fafc208c967af5131f7874f5441
Author: Josh Durgin <josh.durgin@inktank.com>
Date:   Tue Feb 11 11:53:00 2014 -0800

    ObjectCacher: use uint64_t for target and max values
    
    All the options are uint64_t, but the ObjectCacher was converting them
    to int64_t. There's never any reason for these to be negative, so
    change the type.
    
    Adjust a few conditionals so that they only convert known-positive
    signed values to uint64_t before comparing with the target and max
    values. Leave the actual stats accounting as loff_t for now, since
    bugs in accounting will have bad effects if negative values wrap
    around.
    
    Backport: emperor, dumpling
    Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
    (cherry picked from commit db034acf546a72739ff6543241543f3bd651f3ae)

commit a572aafe178dfa3ab4d82aa5c7f4a72a63590ccf
Author: Josh Durgin <josh.durgin@inktank.com>
Date:   Tue Feb 11 10:35:14 2014 -0800

    ObjectCacher: remove max_bytes and max_ob arguments to trim()
    
    These are never passed, so replace them with the defaults.
    
    Backport: emperor, dumpling
    Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
    (cherry picked from commit bf8cf2d6d21a204a099347f3dcd5b48100b8c445)

commit 4889234f80dbd95e5b2012ff9bfc2d87960034e2
Merge: 6af657b 30a604d
Author: Sage Weil <sage@inktank.com>
Date:   Tue Feb 11 08:32:37 2014 -0800

    Merge pull request #1210 from dachary/dumpling
    
    common: admin socket fallback to json-pretty format (dumpling)
    
    Reviewed-by: Sage Weil <sage@inktank.com>

commit 30a604d2d6a5de7bcbfa0136e9a766cb34f96c0a
Author: Loic Dachary <loic@dachary.org>
Date:   Mon Feb 10 23:42:38 2014 +0100

    common: admin socket fallback to json-pretty format
    
    If the format argument to a command sent to the admin socket is not
    among the supported formats ( json, json-pretty, xml, xml-pretty ) the
    new_formatter function will return null and the AdminSocketHook::call
    function must fall back to a sensible default.
    
    The CephContextHook::call and HelpHook::call failed to do that and a
    malformed format argument would cause the mon to crash. A check is added
    to each of them and fallback to json-pretty if the format is not
    recognized.
    
    To further protect AdminSocketHook::call implementations from similar
    problems the format argument is checked immediately after accepting the
    command in AdminSocket::do_accept and replaced with json-pretty if it is
    not known.
    
    A test case is added for both CephContextHook::call and HelpHook::call
    to demonstrate the problem exists and is fixed by the patch.
    
    Three other instances of unsafe calls to new_formatter were found and
    a fallback to json-pretty was added. All other calls have been audited
    and appear to be safe.
    
    http://tracker.ceph.com/issues/7378 fixes #7378
    
    Signed-off-by: Loic Dachary <loic@dachary.org>
    (cherry picked from commit 165e76d4d03ffcc490fd3c2ba60fb37372990d0a)

commit 6af657bd4cf87f6f7afe84257941e7e2366183d9
Author: Josh Durgin <josh.durgin@inktank.com>
Date:   Wed Feb 5 17:26:02 2014 -0800

    qa: add script for testing rados client timeout options
    
    Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
    (cherry picked from commit 9e62beb80b6c92a97ec36c0db5ea39e417661b35)

commit 21a0b67440eaa07fa051d26321a88654ee9e8ae7
Author: Josh Durgin <josh.durgin@inktank.com>
Date:   Wed Feb 5 17:25:24 2014 -0800

    rados: check return values for commands that can now fail
    
    A few places were not checking the return values of commands, since
    they could not fail before timeouts were added.
    
    Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
    (cherry picked from commit 79c1874346ff55e2dc74ef860db16ce70242fd00)

commit 2c42dd26b49d01922bd3df3ba26af70da0387a99
Author: Josh Durgin <josh.durgin@inktank.com>
Date:   Wed Feb 5 17:24:16 2014 -0800

    librados: check and return on error so timeouts work
    
    Some functions could not previously return errors, but they had an
    int return value, which can now receive ETIMEDOUT.
    
    Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
    (cherry picked from commit 8e9459e897b1bc2f97d52ee07701fd22069efcf3)

commit 28fe564cc298b96c063e6cca44304046a86942cd
Author: Josh Durgin <josh.durgin@inktank.com>
Date:   Wed Feb 5 17:22:14 2014 -0800

    msg/Pipe: add option to restrict delay injection to specific msg type
    
    This makes it possible to test timeouts reliably by delaying certain
    messages effectively forever, but still being able to e.g. connect and
    authenticate to the monitors.
    
    Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
    (cherry picked from commit d389e617c1019e44848330bf9570138ac7b0e5d4)

commit 1d2ea9efb844cb902d282c3a752acc45fefeb04c
Author: Josh Durgin <josh.durgin@inktank.com>
Date:   Mon Feb 3 18:30:00 2014 -0800

    MonClient: add a timeout on commands for librados
    
    Just use the conf option directly, since librados is the only caller.
    
    Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
    (cherry picked from commit 671a76d64bc50e4f15f4c2804d99887e22dcdb69)

commit 30dafacd0b54bb98b01284851e0d5abf76324e95
Author: Josh Durgin <josh.durgin@inktank.com>
Date:   Mon Feb 3 17:59:21 2014 -0800

    Objecter: implement mon and osd operation timeouts
    
    This captures almost all operations from librados other than mon_commands().
    
    Get the values for the timeouts from the Objecter constructor, so only
    librados uses them.
    
    Add C_Cancel_*_Op, finish_*_op(), and *_op_cancel() for each type of
    operation, to mirror those for Op. Create a callback and schedule it
    in the existing timer thread if the timeouts are specified.
    
    Fixes: #6507
    Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
    (cherry picked from commit 3e1f7bbb4217d322f4e0ece16e676cd30ee42a20)
    
    Conflicts:
    	src/osd/OSD.cc
    	src/osd/ReplicatedPG.cc
    	src/osdc/Objecter.cc
    	src/osdc/Objecter.h

commit 2aa0d7589283fe5879a966601fbfe690816cc4a1
Author: Josh Durgin <josh.durgin@inktank.com>
Date:   Mon Feb 3 12:53:15 2014 -0800

    librados: add timeout to wait_for_osdmap()
    
    This is used by several pool operations independent of the objecter,
    including rados_ioctx_create() to look up the pool id in the first
    osdmap.
    
    Unfortunately we can't just rely on WaitInterval returning ETIMEDOUT,
    since it may also get interrupted by a signal, so we can't avoid
    keeping track of time explicitly here.
    
    Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
    (cherry picked from commit 1829d2c9fd13f2cbae4e192c9feb553047dad42c)
    
    Conflicts:
    	src/librados/RadosClient.cc

commit 55d254d5ea113d8be29afe18dc9940f14d93ebb4
Author: Josh Durgin <josh.durgin@inktank.com>
Date:   Mon Feb 3 12:09:34 2014 -0800

    conf: add options for librados timeouts
    
    These will be implemented in subsequent patches.
    
    Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
    (cherry picked from commit 0dcceff1378d85ca6d81d102d201890b8a71af6b)

commit ccd9b5eb9bcfe2360e874b306d4decd8815ecf46
Author: Sage Weil <sage@inktank.com>
Date:   Sun Feb 9 20:53:00 2014 -0800

    test_striper: fix warning
    
    Signed-off-by: Sage Weil <sage@inktank.com>

commit bb44c0d5e06c2b699c81a76373a0edde1bf90445
Author: Sage Weil <sage@inktank.com>
Date:   Sat Feb 8 12:21:26 2014 -0800

    crushtool: add cli test for off-by-one tries vs retries bug
    
    See bug #7370.  This passes on dumpling and breaks prior to the #7370 fix.
    
    Backport: emperor, dumpling
    Signed-off-by: Sage Weil <sage@inktank.com>
    (cherry picked from commit ed32c4002fb5cb1dd546331651eaf7de1a017471)

commit c301ec689caac5bbcd21e02226045d3454cc82fd
Author: Sage Weil <sage@inktank.com>
Date:   Mon Feb 3 08:54:14 2014 -0800

    client: use 64-bit value in sync read eof logic
    
    The file size can jump to a value that is very much larger than our current
    position (for example, it could be a disk image file that gets a sparse
    write at a large offset).  Use a 64-bit value so that 'some' doesn't
    overflow.
    
    Signed-off-by: Sage Weil <sage@inktank.com>
    Reviewed-by: John Spray <john.spray@inktank.com>
    (cherry picked from commit 7ff2b541c24d1c81c3bcfbcb347694c2097993d7)

commit 183deb899bc6b1b7b2a1ec639425e45786e56b01
Author: Sage Weil <sage@inktank.com>
Date:   Sat Jan 4 22:39:35 2014 -0800

    osd: do not send peering messages during init
    
    Do not send any peering messages while we are still working our way
    through init().
    
    Fixes: #7093
    Signed-off-by: Sage Weil <sage@inktank.com>
    (cherry picked from commit 35da8f9d80e0c6c33fb6c6e00f0bf38f1eb87d0e)
    Signed-off-by: Greg Farnum <greg@inktank.com>

commit 3a1fb17401897b0717cf1d264c6e6fdf99fc703f
Author: Sage Weil <sage@inktank.com>
Date:   Tue Jan 28 18:46:37 2014 -0800

    OSDMap: fix deepish_copy_from
    
    Start with a shallow copy!
    
    Signed-off-by: Sage Weil <sage@inktank.com>
    (cherry picked from commit d0f13f54146694a197535795da15b8832ef4b56f)
    
    Conflicts:
    
    	src/osd/OSDMap.h

commit f01bd8a94adc4ce8e7131f74282de5b434a7316a
Author: Yehuda Sadeh <yehuda@inktank.com>
Date:   Thu Jan 23 13:48:28 2014 -0800

    rgw: fix listing of multipart upload parts
    
    Fixes: #7169
    
    There are two issues here. One is that we may return more entries than
    we should (as specified by max_parts). Second issue is that the
    NextPartNumberMarker is set incorrectly. Both of these issues mainly
    affect uploads with > 1000 parts, although can be triggered with less
    than that.
    
    Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>

commit 6b657992d2397e399eacf554357ed807fda399fa
Author: Yehuda Sadeh <yehuda@inktank.com>
Date:   Thu Nov 21 06:52:48 2013 -0800

    rgw: initialize RGWUserAdminOpState::system_specified
    
    Fixes: #6829
    Backport: dumpling, emperor
    We didn't init this member variable, which might cause that when
    modifying user info that has this flag set the 'system' flag might
    inadvertently reset.
    
    Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
    (cherry picked from commit 561e7b0b287e65e90b80699e45a52ae44e94684f)

commit 585e0e7eec1bbee60fe352166b593d53476003f8
Author: Robin H. Johnson <robbat2@gentoo.org>
Date:   Sun Dec 15 12:26:19 2013 -0800

    rgw: Fix CORS allow-headers validation
    
    This fix is needed because Ceph presently validates CORS headers in a
    case-sensitive manner. Keeps a local cache of lowercased allowed headers
    to avoid converting the allowed headers to lowercase each time.
    
    CORS 6.2.6: If any of the header field-names is not a ASCII
    case-insensitive match for any of the values in list of headers do not
    set any additional headers and terminate this set of steps.
    
    Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
    Reviewed-by: Yehuda Sadeh <yehuda@inktank.com>
    (cherry picked from commit 31b60bfd9347a386ff12b4e4f1812d664bcfff01)

commit da2267a87e37afa28385e915a566de26fd784b70
Author: Robin H. Johnson <robbat2@gentoo.org>
Date:   Sun Dec 15 11:40:31 2013 -0800

    rgw: Clarify naming of case-change functions
    
    It is not clear that the lowercase_http_attr & uppercase_http_attr
    functions replace dashes with underscores. Rename them to match the
    pattern established by the camelcase_dash_http_attr function in
    preperation for more case-change functions as needed by later fixes.
    
    Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
    Reviewed-by: Yehuda Sadeh <yehuda@inktank.com>
    (cherry picked from commit 6a7edab2673423c53c6a422a10cb65fe07f9b235)

commit fdea76ef6f7d7c4ee89c30d430f5495072023957
Author: Robin H. Johnson <robbat2@gentoo.org>
Date:   Sun Dec 15 11:27:49 2013 -0800

    rgw: Look at correct header about headers for CORS
    
    The CORS standard dictates that preflight requests are made with the
    Access-Control-Request-Headers header containing the headers of the
    author request. The Access-Control-Allow-Headers header is sent in the
    response.
    
    The present code looks for Access-Control-Allow-Headers in request, so
    fix it to look at Access-Control-Request-Headers instead.
    
    Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
    Reviewed-by: Yehuda Sadeh <yehuda@inktank.com>
    (cherry picked from commit 2abacd9678ae04cefac457882ba718a454948915)

commit a5739e3903f8c30c8168199413818922b4a29bad
Author: Yehuda Sadeh <yehuda@inktank.com>
Date:   Fri Dec 6 11:07:09 2013 -0800

    rgw: fix reading bucket policy in RGWBucket::get_policy()
    
    Fixes: 6940
    Backport: dumpling, emperor
    
    We changed the way we keep the bucket policy, and we shouldn't try to
    access the bucket object directly. This had changed when we added the
    bucket instance object around dumpling.
    
    Reported-by: Gao, Wei M <wei.m.gao@intel.com>
    Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
    (cherry picked from commit 7a9a088d82d04f6105d72f6347673724ac16c9f8)

commit ba792cdc8d0c006e260e6a02a33ca1491b35901c
Author: Yehuda Sadeh <yehuda@inktank.com>
Date:   Thu Jan 16 11:45:27 2014 -0800

    rgw: handle racing object puts when object doesn't exist
    
    If the object didn't exist before and now we have multiple puts coming
    in concurrently, we need to make sure that we behave correctly. Only one
    needs to win, the other one can fail silently. We do that by setting
    exclusive flag on the object creation and handling the error correctly.
    Note that we still want to return -EEXIST in some cases (when the
    exclusive flag is passed to put_obj_meta(), e.g., on bucket creation).
    
    Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
    (cherry picked from commit 86c15480fc3e33b9a3b84d0af68d8398fc732bae)

commit 1e2984ac4a4ed0508e1b5efc8b69ae4da0cac824
Author: Yehuda Sadeh <yehuda@inktank.com>
Date:   Thu Jan 16 11:33:49 2014 -0800

    rgw: don't return -ENOENT in put_obj_meta()
    
    Fixes: #7168
    
    An object put may race with the same object's delete. In this case just
    ignore the error, same behavior as if object was created and then
    removed.
    
    Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
    (cherry picked from commit 5c24a7ea905587fd4077e3b0cfc0f5ad2b178c29)

commit 5f1eea2ccf1ee4296530fc754bfd5dfc1571cd62
Author: Robin H. Johnson <robbat2@gentoo.org>
Date:   Sat Jan 18 18:01:20 2014 -0800

    rgw: Use correct secret key for POST authn
    
    The POST authentication by signature validation looked up a user based
    on the access key, then used the first secret key for the user. If the
    access key used was not the first access key, then the expected
    signature would be wrong, and the POST would be rejected.
    
    Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
    (cherry picked from commit 586ad1f69e6260ef298805647911b3b49f8be7cf)

commit 14e933eb2e8adf3e7df2aef823d863c593067928
Author: Joao Eduardo Luis <joao.luis@inktank.com>
Date:   Tue Jan 28 15:54:33 2014 +0000

    osd: OSDMonitor: ignore pgtemps from removed pool
    
    There's a window in-between receiving an MOSDPGTemp message from an OSD
    and actually handling it that may lead to the pool the pg temps refer to
    no longer existing. This may happen if the MOSDPGTemp message is queued
    pending dispatching due to an on-going proposal (maybe even the pool
    removal).
    
    This patch fixes such behavior in two steps:
    
    1. Check if the pool exists in the osdmap upon preprocessing
     - if pool does not exist in the osdmap, then the pool must have been
       removed prior to handling the message, but after the osd sent it.
     - safe to ignore the pg update
    2. If all pg updates in the message have been ignored, ignore the whole
       message.  Otherwise, let prepare handle the rest.
    
    3. Recheck if pool exists in the osdmap upon prepare
     - We may have ignored this pg back in preprocess, but other pgs in the
       message may have led the message to be passed on to prepare; ignore
       pg update once more.
    4. Check if pool is pending removal and ignore pg update if so.
    
    We delegate checking the pending value to prepare_pgtemp() because in this
    case we should only ignore the update IFF the pending value is in fact
    committed.  Otherwise we should retry the message.  prepare_pgtemp() is
    the appropriate place to do so.
    
    Fixes: 7116
    
    Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
    (cherry picked from commit f513f66f48383a07c70ca18a4dba6c2449ea9860)

commit 3672fc49bea5efcef2069673c88fd8e526eeb7e5
Author: Sage Weil <sage@inktank.com>
Date:   Tue Jan 28 11:00:34 2014 -0800

    OSDMonitor: use deepish_copy_from for remove_down_pg_temp
    
    This is a backport of 368852f6c0a884b8fdc80a5cd6f9ab72e814d412.
    
    Make a deep copy of the OSDMap to avoid clobbering the in-memory copy with
    the call to apply_incremental.
    
    Fixes: #7060
    Signed-off-by: Sage Weil <sage@inktank.com>

commit e6fef68ee51a2e3be6537086246ce860819788eb
Author: Sage Weil <sage@inktank.com>
Date:   Fri Jan 24 11:03:26 2014 -0800

    OSDMap: deepish_copy_from()
    
    Make a deep(ish) copy of another OSDMap.  Unfortunatley we can't make the
    compiler-generated copy operator/constructors private until c++11.  :(
    
    Signed-off-by: Sage Weil <sage@inktank.com>
    (cherry picked from commit bd54b9841b9255406e56cdc7269bddb419453304)

commit 2528186c0dd327607b1e76a51581df72aeea52f2
Author: Sage Weil <sage@inktank.com>
Date:   Tue Jan 28 10:26:12 2014 -0800

    buffer: make 0-length splice() a no-op
    
    This was causing a problem in the Striper, but fixing it here will avoid
    corner cases all over the tree.  Note that we have to bail out before
    the end-of-buffer check to avoid hitting that check when the bufferlist is
    also empty.
    
    Signed-off-by: Sage Weil <sage@inktank.com>
    Reviewed-by: Greg Farnum <greg@inktank.com>
    (cherry picked from commit ff5abfbdae07ae8a56fa83ebaa92000896f793c2)

commit fe9539e118a830aa71bc5846ad2693543213fc59
Author: Sage Weil <sage@inktank.com>
Date:   Tue Jan 28 10:09:17 2014 -0800

    osdc/Striper: test zero-length add_partial_result
    
    If we add a partial result that is 0-length, we used to hit an assert in
    buffer::list::splice().  Add a unit test to verify the fix.
    
    Signed-off-by: Sage Weil <sage@inktank.com>
    Reviewed-by: Greg Farnum <greg@inktank.com>
    (cherry picked from commit 28c7388d320a47657c2e12c46907f1bf40672b08)

commit 9801718e6e7a335d1aeffb7f3d958b012d8dd79c
Author: Derek Yarnell <derek@umiacs.umd.edu>
Date:   Mon Jan 27 12:27:51 2014 -0700

    packaging: apply udev hack rule to RHEL
    
    In the RPM spec file there is a test to deploy the uuid hack udev rules
    for older udev operating systems. This includes CentOS and RHEL, but the
    check currently only is for CentOS, causing RHEL clients to get a bogus
    osd rules file.
    
    Adjust the conditional to apply to RHEL as well as CentOS. (The %{rhel}
    macro is defined in both platforms' redhat-rpm-config package.)
    
    Fixes http://tracker.ceph.com/issues/7245
    
    Signed-off-by: Ken Dreyer <ken.dreyer@inktank.com>
    (cherry picked from commit 64a0b4fa563795bc22753940aa3a4a2946113109)

commit 5817078ba9b2aa38f39e1f62d8d08e943646c0bb
Merge: 9fa8d2d abf6441
Author: Sage Weil <sage@inktank.com>
Date:   Thu Jan 23 07:12:37 2014 -0800

    Merge pull request #1129 from ceph/wip-dumpling-backport-6620
    
    mds: MDSMap: adjust buffer size for uint64 values with more than 5 chars

commit abf64417bbc63b006c53a07262a5774d0d40550e
Author: Joao Eduardo Luis <jecluis@gmail.com>
Date:   Thu Jan 23 11:36:16 2014 +0000

    mds: MDSMap: adjust buffer size for uint64 values with more than 5 chars
    
    Fixes: #6620
    
    Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
    (cherry picked from commit 0e8182edd850f061421777988974efbaa3575b9f)

commit 9fa8d2d1bd4a8cf56fdebae484bee35c3930fb62
Author: Sage Weil <sage@inktank.com>
Date:   Tue Jan 21 11:29:56 2014 -0800

    mon/MDSMonitor: do not generate mdsmaps from already-laggy mds
    
    There is one path where a mds that is not sending its beacon (e.g.,
    because it is not running at all) will lead to proposal of new mdsmaps.
    Fix it.
    
    Backport: emperor, dumpling
    Signed-off-by: Sage Weil <sage@inktank.com>
    (cherry picked from commit 584c2dd6bea3fe1a3c7f306874c054ce0cf0d2b5)

commit a2e34b8d322e571600d6c0a3ff363f9910196f28
Author: John Spray <john.spray@inktank.com>
Date:   Mon Jan 20 11:08:27 2014 +0000

    Fix #7187: Include all summary items in JSON health output
    
    Signed-off-by: John Spray <john.spray@inktank.com>
    (cherry picked from commit fdf3b5520d150f14d90bdfc569b70c07b0579b38)

commit ee9fc590a556354b54a174a4790c4142949d7409
Author: Yehuda Sadeh <yehuda@inktank.com>
Date:   Mon Jan 6 18:32:42 2014 -0800

    rgw: convert bucket info if needed
    
    Fixes: #7110
    In dumpling, the bucket info was separated into bucket entry point and
    bucket instance objects. When setting bucket attrs we only ended up
    updating the bucket instance object. However, pre-dumpling buckets still
    keep everything at the entry-point object, so acl changes didn't affect
    anything (because we never updated the entry point). This change just
    converts the bucket info into the new format.
    
    Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
    (cherry picked from commit a5f8cc7ec9ec8bef4fbc656066b4d3a08e5b215b)

commit 69a99e6edd67aadaea77fb5ebcf46190dbfcdbc6
Author: Sage Weil <sage@inktank.com>
Date:   Sat Jan 4 22:40:43 2014 -0800

    osd: ignore OSDMap messages while we are initializing
    
    The mon may occasionally send OSDMap messages to random OSDs, but is not
    very descriminating in that we may not have authenticated yet.  Ignore any
    messages if that is the case; we will reqeust whatever we need during the
    BOOTING state.
    
    Fixes: #7093
    Signed-off-by: Sage Weil <sage@inktank.com>
    (cherry picked from commit f68de9f352d53e431b1108774e4a23adb003fe3f)

commit 1545ddbd3b6dfac897d483bd58a4335d68b9b62a
Author: Sage Weil <sage@inktank.com>
Date:   Sat Jan 4 22:43:26 2014 -0800

    mon: only send messages to current OSDs
    
    When choosing a random OSD to send a message to, verify not only that
    the OSD id is up but that the session is for the same instance of that OSD
    by checking that the address matches.
    
    Fixes: #7093
    Backport: emperor, dumpling
    Signed-off-by: Sage Weil <sage@inktank.com>
    (cherry picked from commit 98ed9ac5fed6eddf68f163086df72faabd9edcde)

commit 4d88dd10bfab4e5fb45632245be5f79eeba73a30
Author: Sage Weil <sage@inktank.com>
Date:   Mon Aug 26 13:58:47 2013 -0700

    osd: discriminate based on connection messenger, not peer type
    
    Replace ->get_source().is_osd() checks and instead see if it is the
    cluster_messenger so that we do not confuse ourselves when we get
    legit requests from other OSDs on our public interface.
    
    NOTE: backporting this because a mixed cluster may send OSD requests
    via the client interface, even though dumpling doesn't do this.
    
    Signed-off-by: Sage Weil <sage@inktank.com>
    (cherry picked from commit a6b04c5d8ba043727a2e39a62e9d4126485bcfeb)
    
    Conflicts:
    
    	src/osd/OSD.cc

commit 2354b957b7924af3e81f559d5c975262848c3bf6
Author: Loic Dachary <loic@dachary.org>
Date:   Sun Dec 15 22:59:51 2013 +0100

    qa: test for error when ceph osd rm is EBUSY
    
    http://tracker.ceph.com/issues/6824 fixes #6824
    
    Signed-off-by: Loic Dachary <loic@dachary.org>
    (cherry picked from commit 31507c90f0161c4569a2cc634c0b5f671179440a)

commit 10ad2e01f832ca9293a1bda1c82f7d318e8c3589
Author: Loic Dachary <loic@dachary.org>
Date:   Sun Dec 15 16:27:02 2013 +0100

    mon: set ceph osd (down|out|in|rm) error code on failure
    
    Instead of always returning true, the error code is set if at least one
    operation fails.
    
    EINVAL if the OSD id is invalid (osd.foobar for instance).
    EBUSY if trying to remove and OSD that is up.
    
    When used with the ceph command line, it looks like this:
    
        ceph -c ceph.conf osd rm osd.0
        Error EBUSY: osd.0 is still up; must be down before removal.
        kill PID_OF_osd.0
        ceph -c ceph.conf osd down osd.0
        marked down osd.0.
        ceph -c ceph.conf osd rm osd.0 osd.1
        Error EBUSY: removed osd.0, osd.1 is still up; must be down before removal.
    
    http://tracker.ceph.com/issues/6824 fixes #6824
    
    Signed-off-by: Loic Dachary <loic@dachary.org>
    (cherry picked from commit 15b8616b13a327701c5d48c6cb7aeab8fcc4cafc)

commit db82bfec50349783c3bb708c65cf89bceff7b2ca
Author: Joao Eduardo Luis <joao.luis@inktank.com>
Date:   Tue Oct 29 20:30:37 2013 +0000

    mon: OSDMonitor: fix some annoying whitespace
    
    Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
    (cherry picked from commit 42c4137cbfacad5654f02c6608cc0e81b45c06be)

commit 1abb16988727a21dff61edca48bfcb05c60d26f2
Author: Josh Durgin <josh.durgin@inktank.com>
Date:   Thu Dec 26 17:38:52 2013 -0800

    librbd: call user completion after incrementing perfcounters
    
    The perfcounters (and the ictx) are only valid while the image is
    still open.  If the librbd user gets the callback for its last I/O,
    then closes the image, the ictx and its perfcounters will be
    invalid. If the AioCompletion object is has not run the rest of its
    complete() method yet, it will access these now-invalid addresses,
    possibly leading to a crash.
    
    The AioCompletion object is independent of the ictx and does not
    access it again after incrementing perfcounters, so avoid this race by
    calling the user's callback after this step. The AioCompletion object
    will be cleaned up by the rest of complete_request(), independent of
    the ImageCtx.
    
    Fixes: #5426
    Backport: dumpling, emperor
    Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
    (cherry picked from commit 4cea7895da7331b84d8c6079851fdc0ff2f4afb1)

commit 4dae50d8a0ffe3f3491eae62950fe46229aaa0e6
Author: Josh Durgin <josh.durgin@inktank.com>
Date:   Fri Dec 6 16:03:20 2013 -0800

    objecter: don't take extra throttle budget for resent ops
    
    These ops have already taken their budget in the original op_submit().
    It will be returned via put_op_budget() when they complete.
    If there were many localized reads of missing objects from replicas,
    or cache pool redirects, this would cause the objecter to use up all
    of its op throttle budget and hang.
    
    Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
    (cherry picked from commit 8d0180b1b7b48662daef199931efc7f2a6a1c431)
    
    Conflicts:
    
    	src/osdc/Objecter.cc

commit 4944618fe74e1875e37a5de9a5bfe3922f906942
Author: Josh Durgin <josh.durgin@inktank.com>
Date:   Thu Dec 5 17:44:37 2013 -0800

    rbd: check write return code during bench-write
    
    This is allows rbd-bench to detect http://tracker.ceph.com/issues/6938
    when combined with rapidly changing the mon osd full ratio.
    
    Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
    (cherry picked from commit 3caf3effcb113f843b54e06099099909eb335453)

commit 3a7a0353d7c8521a2c3c25778801d45f0ca10957
Author: Josh Durgin <josh.durgin@inktank.com>
Date:   Thu Dec 5 17:36:33 2013 -0800

    objecter: resend all writes after osdmap loses the full flag
    
    Now that the osd does not respond if it gets a map with the full flag
    set first, clients need to resend all writes.
    
    Clients talking to old osds are still subject to the race condition,
    so both sides must be upgraded to avoid it.
    
    Refs: #6938
    Backport: dumpling, emperor
    Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
    (cherry picked from commit e32874fc5aa6f59494766b7bbeb2b6ec3d8f190e)
    
    Conflicts:
    
    	src/osdc/Objecter.h

commit f4d0fc8afac1bcbea76f46744397a8462025f373
Author: Josh Durgin <josh.durgin@inktank.com>
Date:   Thu Dec 5 17:34:38 2013 -0800

    osd: drop writes when full instead of returning an error
    
    There's a race between the client and osd with a newly marked full
    osdmap.  If the client gets the new map first, it blocks writes and
    everything works as expected, with no errors from the osd.
    
    If the osd gets the map first, however, it will respond to any writes
    with -ENOSPC. Clients will pass this up the stack, and not retry these
    writes later.  -ENOSPC isn't handled well by all clients. RBD, for
    example, may pass it on to qemu or kernel rbd which will both
    interpret it as EIO.  Filesystems on top of rbd will not behave well
    when they receive EIOs like this, especially if the cluster oscillates
    between full and not full, so some writes succeed.
    
    To fix this, never return ENOSPC from the osd because of a map marked
    full, and rely on the client to retry all writes when the map is no
    longer marked full.
    
    Old clients talking to osds with this fix will hang instead of
    propagating an error, but only if they run into this race
    condition. ceph-fuse and rbd with caching enabled are not affected,
    since the ObjectCacher will retry writes that return errors.
    
    Refs: #6938
    Backport: dumpling, emperor
    Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
    (cherry picked from commit 4111729dda7437c23f59e7100b3c4a9ec4101dd0)

commit 2b109803b9434d2d495e65d6a045838ee3ebf169
Author: Yehuda Sadeh <yehuda@inktank.com>
Date:   Wed Nov 6 16:55:52 2013 -0800

    objecter: clean pause / unpause logic
    
    op->paused holds now whether operation should be paused or not, and it's
    being updated when scanning requests. No need to do a second scan.
    
    Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
    (cherry picked from commit 5fe3dc647bf936df8e1eb2892b53f44f68f19821)

commit 7b229fca346b7ab60131b229591d35f63f64a6ad
Author: Yehuda Sadeh <yehuda@inktank.com>
Date:   Wed Nov 6 16:15:47 2013 -0800

    objecter: set op->paused in recalc_op_target(), resend in not paused
    
    When going through scan_requests() in handle_osd_map() we need to make
    sure that if an op should not be paused anymore then set it on the op
    itself, and return NEED_RESEND. Otherwise we're going to miss reset of
    the full flag.
    Also in handle_osd_map(), make sure that op shouldn't be paused before
    sending it. There's a lot of cleanup around that area that we should
    probably be doing, make the code much more tight.
    
    Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
    (cherry picked from commit 98ab7d64a191371fa39d840c5f8e91cbaaa1d7b7)

commit a3fdfd77c953a8f9ace4991630599e60f56abc4a
Author: Josh Durgin <josh.durgin@inktank.com>
Date:   Wed Nov 6 10:46:37 2013 +0800

    objecter: don't resend paused ops
    
    Paused ops are meant to block on the client side until a new map that
    unpauses them is recieved. If we send paused writes when the FULL flag
    is set, we'll get -ENOSPC from the osds, which is not what Objecter
    users expect. This may cause rbd without caching to produce an I/O
    error instead of waiting for the cluster to have capacity.
    
    Fixes: #6725
    Backport: dumpling
    Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
    (cherry picked from commit c5c399d327cfc0d232d9ec7d49ababa914d0b21a)

commit 78037886ae6ae278d8fb2db52d7767b8612313fe
Merge: a60ac91 4e47a00
Author: Sage Weil <sage@inktank.com>
Date:   Sun Dec 29 21:10:30 2013 -0800

    Merge pull request #808 from ceph/wip-6152-dumpling
    
    rgw: Fix S3 auth when using response-* query string params

commit 4e47a003bb3adcee690067d4037c248a93023c20
Author: Sylvain Munaut <s.munaut@whatever-company.com>
Date:   Thu Aug 29 16:17:30 2013 +0200

    rgw: Fix S3 auth when using response-* query string params
    
    Reviewed-by: Yehuda Sadeh <yehuda@inktank.com>
    Signed-off-by: Sylvain Munaut <s.munaut@whatever-company.com>
    (cherry picked from commit 7a7361d7e798d94796d4985e2e4b35af22463ae2)