Reviewed:  https://review.openstack.org/244249
Committed: https://git.openstack.org/cgit/openstack/swift/commit/?id=19226e4449bca399981cde3a1e9882735fed74d5
Submitter: Jenkins
Branch:    feature/crypto

commit 2d85a3f6997fbadac210207f841c4ec25ff50cc4
Author: Tushar Gohad <email address hidden>
Date:   Fri Oct 23 13:16:33 2015 +0000

    EC: Use best available ec_type in unittests
    
    To minimize external library dependencies for Swift unit
    tests and SAIO, PyECLib 1.1.1 introduces a native backend
    'liberasurecode_rs_vand.'  This patch is to migrate over
    the unit tests to the new ec_type when available.
    
    This change will work with current pyeclib requirements
    (==1.0.7) and also future requirements (>=1.0.7).
    
    When we're able to raise *our* requirements to >=1.1.1 we
    should remove jerasure from the list of preferred backends.
    Related SAIO doc and example config changes should be
    included with that patch.
    
    Co-Authored-By: Clay Gerrard <email address hidden>
    Change-Id: Idf657f0acf0479bc8158972e568a29dbc08eaf3b

commit 66dc1eebb1db51c81891b1f7bcc3e85aef9b8c1d
Author: Bill Huber <email address hidden>
Date:   Fri Oct 16 11:27:34 2015 -0500

    ObjectControllers return application errors as 499 on bad read
    
    In the _transfer_data method, we translate all (Exception, Timeout)
    into a 499 whereas we should consider translating them to 500 on
    particular returning error scenarios.
    
    This affects both ReplicatedObjectController and ECObjectControllear.
    
    Change-Id: I571bbc5b1451243907b094a5718c8735fd824268
    Closes-Bug: 1504299

commit 705642db4c27c34ac8eb840d40b16470b85418c8
Author: venkatamahesh <email address hidden>
Date:   Tue Nov 10 19:46:31 2015 +0530

    Change stackforge repo to openstack repo
    
    The projects which are moved to openstack
    are corrected by replacing stackforge with openstack
    
    Change-Id: I65b794a7f10df617bc2a4caf2c4010477a82dbc2

commit d755f5b52024e255787d180bbca60c09192a8850
Author: John Dickinson <email address hidden>
Date:   Mon Nov 9 22:03:34 2015 -0800

    suppress warning output in a unit test
    
    test_write_builder_after_device_removal() wasn't setting a
    default min_part_hours so a warnign was printed. Explicitly
    adding a min_part_hours suppresses the warning
    
    Change-Id: I6f234b72c34e066abb91f28e6eacf50e29be8842

commit 37337d5fcb0a29344381f8de7d810a9e76e10928
Author: Pradeep Kumar Singh <email address hidden>
Date:   Sun May 17 13:35:58 2015 +0530

    Read the response body, if response status is greater than 300.
    
    internal_client was not reading response if response status is not 200.
    So proxy server treats this as client disconnect and logs 499 in log file.
    This patch fixes that by reading response if response status is greater
    than or equal to 300 and in acceptable statuses.
    
    Closes-Bug: #1364752
    
    Change-Id: I0512a25895da583956f76031e3c5de5c970bce01

commit c81f202f714c36390a25b24fa96d1b277788ade9
Author: Kota Tsuyuzaki <email address hidden>
Date:   Thu Nov 5 17:37:30 2015 -0800

    Add missing docs for ring.builder.rebalance
    
    commit 71993d84e88cc1d5f7742182905cace21c7e88cb added
    a new 'remove_dev' column to the
    swift.common.ring.builder.rebalance return value.
    
    This patch adds the docs for that and clean up a bit to
    the variable name to be easy to read.
    
    Change-Id: Idfd46e47b9f6894cbafc8b7701a4c7414212f79f

commit 331382e1922add01959b88e58bc8c1d2029f4550
Author: Paul Dardeau <email address hidden>
Date:   Thu Oct 29 21:09:01 2015 +0000

    Add unit tests to cover print_item_locations
    
    Add unit tests to cover all code paths in print_item_locations
    function in cli/info.py.
    
    Update comment to match what's tested for invalid/missing policy.
    
    Update tests to verify output of print_item_locations
    
    Corrected PEP8 compliance violations.
    
    Change-Id: I84958cb70205ee8d7ea246826dd56201fa642da9

commit e0430fc74a564a0556ef5eab1accd110566aee31
Author: Ondřej Nový <email address hidden>
Date:   Sun Nov 1 18:10:06 2015 +0100

    Compare Swift config checksum in swift-recon --all
    
    Change-Id: I796fe0895f4e5ddeb04c0d79a73579ce8bb9aa40

commit 609095b23f6c5734c6a7aed80c488b96b31137e8
Author: Victor Stinner <email address hidden>
Date:   Thu Aug 27 00:51:58 2015 +0200

    On py3, use dnspython3 dependency, not dnspython
    
    Update requirements to install dnspython3 on Python 3 and dnspython on
    Python 2. dnspython is not compatible with Python 3, and dnspython3
    is not compatible with Python 2.
    
    Add a requirement on pbr to ensure that pbr 1.6 or newer is
    installed. pbr 1.0 or newer is required to support environment
    markers in requirements which are now used to select dnspython or
    dnspython3 depending on the Python major version.
    
    Note: pbr 1.0 is enough, but OpenStack global requirements asks to
    use "pbr>=1.6".
    
    Change-Id: Ie5d437cd396972d4143ffffb96ec3c289e221b08

commit 6858510b59c23ebef527a92c3258dc3c5c6046f5
Author: Alistair Coles <email address hidden>
Date:   Thu Sep 3 16:13:17 2015 +0100

    Re-organise ssync tests
    
    We have some tests that exercise both the sender and receiver,
    but are spread across test_ssync_sender.py and test_ssync_receiver.py.
    This creates a new module test_ssync.py and moves the end-to-end tests
    into there.
    
    Change-Id: Iea3e9932734924453f7241432afda90abbc75c06

commit 48421f902e2e293a3c9ef672dbec5971ca530fd6
Author: Venkateswarlu Pallamala <email address hidden>
Date:   Wed Nov 4 00:53:05 2015 -0800

    Optimize the code performance
    
    Usually, the number of characters in FORBIDDEN_CHARS are fewer than
    the number of characters in the req.path - so it's better to use
    smaller count for the outer loop.
    
    Change-Id: I78b7e0ff4cae1cc90e8d29d7bddcec7fe0ba4bd4

commit faef717cd3fda6aeb3a0799f25d0483bc9bdce7a
Author: Bill Huber <email address hidden>
Date:   Thu Oct 22 10:14:29 2015 -0500

    Add unit tests for swift.proxy.controllers.base
    
    This patch adds more unit tests to diminish missing pieces
    of the coverage in the proxy_controllers_base unit test.
    
    Change-Id: I85ba1955c681cc9d5b2a70ac31155678d2e5b6fd

commit febdd6c1b4035a3b63b5deb7690b2a41014e5321
Author: Peter Lisák <email address hidden>
Date:   Fri Oct 23 12:40:15 2015 +0200

    Device marked to be removed in info about the ring.
    
    Showing devices with 'DEL' mark if will be removed next rebalance.
    
    Change-Id: I171aa8658b1c4ac1689ab9532fe65d114567baa7

commit a4a5fcad8f1ec9059cfab0c63e81106355b0dbd5
Author: Kota Tsuyuzaki <email address hidden>
Date:   Wed Oct 7 02:14:49 2015 -0700

    Add unit tests for utils.validate_hash_path
    
    This patch add some unit tests to prevent regression and to describe
    validate_hash_path behavior, I found in review with
    
    https://review.openstack.org/#/c/231864/
    
    *bonus*
    - Fix test_hash_path to use "with" syntax instead of "try/finally"
      for assigning a testing value into a global variable.
    
    Change-Id: I948999a8fb8addb9a378dbf8bee853b205aeafad

commit e31ecb24b6684da692fbca8b7fe34c1dd4047a3a
Author: Samuel Merritt <email address hidden>
Date:   Wed Oct 21 13:05:37 2015 -0700

    Get rid of contextlib.nested() for py3
    
    contextlib.nested() is missing completely in Python 3.
    
    Since 2.7, we can use multiple context managers in a 'with' statement,
    like so:
    
        with thing1() as t1, thing2() as t2:
            do_stuff()
    
    Now, if we had some code that needed to nest an arbitrary number of
    context managers, there's stuff we could do with contextlib.ExitStack
    and such... but we don't. We only use contextlib.nested() in tests to
    set up bunches of mocks without crazy-deep indentation, and all that
    stuff fits perfectly into multiple-context-manager 'with' statements.
    
    Change-Id: Id472958b007948f05dbd4c7fb8cf3ffab58e2681

commit 73e032049ffb9b371f30d5230472b920446d08f4
Author: Paul Dardeau <email address hidden>
Date:   Fri Oct 23 18:20:25 2015 +0000

    Update admin guide with region.
    
    Added region prefix to example commands for adding devices to ring.
    Also updates description to include region prefix.
    
    Change-Id: Ie6d6485b497cea973e37909b5b19b44946c8aa89

commit 5d8d2e98ee8004cb009f4bd419fbd8cd5757aaa2
Author: Alistair Coles <email address hidden>
Date:   Fri Oct 23 16:38:24 2015 +0100

    Minor improvement to updater unit tests
    
    Follow up for change I7c57bba053711a27d3802efe6f2a0bf53483a54f
    
    The current tests won't prevent this kind of regression in the
    updaters' __init__ methods:
    
      self.node_timeout = int(float(conf.get('node_timeout', 10)))
    
    so this makes them a little tighter.
    
    Change-Id: I60bd79ac392deb1e8a108357d48474dddd9028c1

commit 1a2b54fc0a22713ff0ca575e0a22834682670c55
Author: Alistair Coles <email address hidden>
Date:   Mon Oct 19 13:55:02 2015 +0100

    Fix missing *-replicator conf sections in deployment guide
    
    The doc for these sections was missing because of an rst error - the
    source is there in rst file but didn't make it into the html output.
    
    Add doc for per_diff and max_diffs in account and container doc sections.
    
    Also, fix a bunch of other sphinx build errors and most of the warnings.
    
    Change-Id: If9ed2619b2f92c6c65a94f41d8819db8726d3893

commit 8f78c1a2b066e816cfd643d048ebbe3b654fbe87
Author: OpenStack Proposal Bot <email address hidden>
Date:   Fri Oct 23 06:34:15 2015 +0000

    Imported Translations from Zanata
    
    For more information about this automatic import see:
    https://wiki.openstack.org/wiki/Translations/Infrastructure
    
    Change-Id: Ifebb8f1b50ac4c2c0f43df48c2414ce0f3742e42

commit 59849a9923d7e214725afc14d0d229194884fc54
Author: Zack M. Davis <email address hidden>
Date:   Thu Oct 22 15:13:09 2015 -0700

    SLO docs(tring): manifest params in table, code block; copyedits
    
    It's æsthetically pleasing to display the segment description parameters
    in a table (using the reStructuredText "simple" table markup) and to use
    a code block (which receives syntax hightlighting in the rendered
    document!) for the example JSON fragment. Also, a few very minor copy
    changes are made (a comma after an introductory phrase clarifies; JSON
    and ETag receieve their canonical capitalizations; the elements of the
    manifest JSON aren't "files" exactly; the inclusive nature of the range
    is noted for the benefit of harried readers who won't read the "Range
    Specification" section later in the document).
    
    Change-Id: I5296fc04c6fe5410ad14bba7de0a7f8ab060b126

commit b3a0acd5c77201161dcd169850df589ede331106
Author: Ganesh Maharaj Mahalingam <email address hidden>
Date:   Tue Oct 20 15:16:39 2015 +0000

    DiskFileNoSpace check during object delete
    
    On a full disk, a call to delete an object will fail when it tries to
    write tombstones. Handling DiskFileNoSpace exception raised by
    swift.common.utils.
    
    Change-Id: I8f0cfcc4159ee154fcd3e7ca90c422aa5aadf0b3
    Signed-off-by: Ganesh Maharaj Mahalingam <email address hidden>
    Closes-Bug: 1491675

commit b376e9dc342a2ea4427467b7cd276f335298134d
Author: Kota Tsuyuzaki <email address hidden>
Date:   Wed Oct 21 05:33:04 2015 -0700

    Add verification socket close causes StopIteration
    
    This patch add a expositive test that closing connection
    (fd and socket) causes StopIteration on reader at proxy server
    on PUT object case.
    
    And then, if the connection closed and proxy detects a lack of
    data length, it will result in HTTPClientDisconnected.
    
    The only reason why I leave test_client_disconnect as current
    is Bill is now working for the change at
    
    https://review.openstack.org/#/c/236007/
    
    Change-Id: If68f752a2b0dd6cfad4d007ca5d155e6ffb378e5

commit 71993d84e88cc1d5f7742182905cace21c7e88cb
Author: Lisak, Peter <email address hidden>
Date:   Fri Oct 9 16:13:55 2015 +0200

    swift-ring-builder can't remove a device with zero weight
    
    If a device with 0 weight is tried to remove, the following rebalance
    does not write changes into builder file.
    
    Scenario:
    $ swift-ring-builder object.builder set_weight --id 1 0.00
    $ swift-ring-builder object.builder rebalance
    Wait for moving files out of the device id=1.
    $ swift-ring-builder object.builder remove --id 1
    $ swift-ring-builder object.builder rebalance
    In fact, the device id=1 is not removed after rebalance (must be --force used).
    
    Change-Id: Iad5a444023eae9882a3addd7f119ff4d18559ddd

commit 2135841e6e3fa6705562a72030c31f0c69fd4470
Author: asettle <email address hidden>
Date:   Wed Oct 7 13:55:45 2015 +1000

    Documenting the ability to specify ranges for SLO segments
    
    Change-Id: Ied854f6385f3b28975808f4707e531581bf6c18e
    Closes-bug: 1499152

commit b6b7578190bc5de266d08c22984622d5566b8456
Author: Lisak, Peter <email address hidden>
Date:   Tue Oct 13 17:00:41 2015 +0200

    node_timeout as float in configs
    
    It is more convenient to use float node_timeout for fine tunning latency.
    
    Change-Id: I7c57bba053711a27d3802efe6f2a0bf53483a54f

commit 71e573f4ba8a45dd5962f9d180c6236ce85627ca
Author: Victor Stinner <email address hidden>
Date:   Mon Oct 19 16:59:14 2015 +0200

    Port swift.common.utils.backward() to Python 3
    
    backward() is written to handle binary files:
    
    * Replace literal native strings to literal byte strings: add b'...'
      prefix
    * Update unit tests: use byte strings
    * TemporaryFile(): use the default mode 'w+b', instead of using 'r+w'
      ('r+w' mode creates a Unicode file on Python 3).
    
    Change-Id: Ic91f7e6c605db0b888763080d49f0f501029837f

commit 4a13dcc4a8d7c890ff322190ca80fbb460b55de4
Author: Matthew Oliver <email address hidden>
Date:   Fri Oct 9 17:25:08 2015 +1100

    Make db_replicator usync smaller containers
    
    The current rule inside the db_replicator is to rsync+merge
    containers during replication if the difference between rowids
    differ by more than 50%:
    
      # if the difference in rowids between the two differs by
      # more than 50%, rsync then do a remote merge.
      if rinfo['max_row'] / float(info['max_row']) < 0.5:
    
    This mean on smaller containers, that only have few rows, and differ
    by a small number still rsync+merge rather then copying rows.
    
    This change adds a new condition, the difference in the rowids must
    be greater than the defined per_diff otherwise usync will be used:
    
      # if the difference in rowids between the two differs by
      # more than 50% and the difference is greater than per_diff,
      # rsync then do a remote merge.
      # NOTE: difference > per_diff stops us from dropping to rsync
      # on smaller containers, who have only a few rows to sync.
      if rinfo['max_row'] / float(info['max_row']) < 0.5 and \
              info['max_row'] - rinfo['max_row'] > self.per_diff:
    
    Change-Id: I9e779f71bf37714919a525404565dd075762b0d4
    Closes-bug: #1019712

commit a9ddc2d9ea402eaac7ccd8992387f77855968ab5
Author: Mahati Chamarthy <email address hidden>
Date:   Fri Oct 16 18:18:33 2015 +0530

    Hyperlink fix to first contribution doc
    
    Change-Id: I19fc1abc89f888233b80a57c68a152c1c1758640

commit 83a1151d13e096b480aefe6ec18259f2d7d021db
Author: Pete Zaitcev <email address hidden>
Date:   Fri Oct 9 16:45:20 2015 -0600

    Interpolate the explanation string not whole HTML body
    
    The only reason this exists is that I promised to do it.
    But in our case, there's no big advantage, and here's why.
    
    The general thinking goes that strings must be interpolated
    first because the body may contain a syntax that confuses the
    interpolation. So this patch makes the code "more correct".
    However, our HTML template is tightly controlled. It's not
    like it contains additional percents.
    
    So I'll just leave this here for now while I'm asking if
    the content type is set correctly.
    
    Change-Id: Ia18aeb0f94ef389f8b95450986a566e5fa06aa10

commit 43cdc29669741568a898dc87eee3e09280887bba
Author: Kota Tsuyuzaki <email address hidden>
Date:   Tue Oct 13 00:22:31 2015 -0700

    Add backend subrequest tests for staticweb
    
    This is an followup for
    https://review.openstack.org/#/c/227204
    
    Above patch changes staticweb to use just make_env instead of
    pre authed env. The tests included in this patch assert
    "backend requests generated by staticweb *NEVER* include pre
    authed information (i.e. authorize_override=True)".
    
    Change-Id: Ifbbc43d3b5ac114bc3d5e4fcab9cacdf9a3f1a5c

commit 9993c1f124bcb6d645254196fe43d8c4f9feae6e
Author: Alistair Coles <email address hidden>
Date:   Wed Oct 14 15:49:35 2015 +0100

    Add test to verify staticweb uses auth'd subrequests
    
    Follow on for change Icf159d7e567ac5481e710c5910db686bdcba6336
    
    Related-Bug: 1489749
    
    Change-Id: I96920645d50322c5232bbcd6bb38b0f26795b92d

commit 4eb796b29ddee52f5010655e0d5477f2d7923854
Author: Carlos Cavanna <email address hidden>
Date:   Thu Oct 8 13:40:20 2015 -0400

    bug/1270114. Missing documentation for Bulk Middleware.
    
    Closes-Bug: #1270114
    
    This patch addresses the missing bulk middleware documentation for
    the following bug:
    
    https://bugs.launchpad.net/swift/+bug/1270114
    
    And feature:
    
    https://review.openstack.org/#/c/176161/
    
    The documentation will show under:
    
    http://docs.openstack.org/developer/swift/middleware.html
    
    Change-Id: I73c101e09830d7120b5e5c1333ae8d570a9e2303

commit 384b91eb824376659989b904f9396cbf2e02d2bd
Author: asettle <email address hidden>
Date:   Thu Sep 3 15:11:46 2015 +1000

    Moving DLO functionality doc to the middleware code
    
    This change moves the RST DLO documentation from
    statically inside overview_large_objects.rst and moves it
    to middleware/dlo.py.
    This is where all middleware RST documentation is defined.
    
    The overview_large_objects.rst is still the main page
    for information on large objects, so now dynamically
    points to both the DLO and SLO middleware RST
    documentation and the relevant middleware.rst page
    simply points to it.
    
    Change-Id: I40d918c8b7bc608ab945805d69fe359521df038a
    Closes-bug: #1276852

commit a7af802497f3b86d33cbee852500df456c316685
Author: Christian Schwede <email address hidden>
Date:   Sat Aug 29 16:03:07 2015 +0000

    Do not use pre-authenticated requests in staticweb
    
    staticweb middleware uses make_pre_authed_env, this makes it possible to
    anonymously list containers without any read acl set if the metadata
    "web-listings: true" is set on a container. Using make_env enforces correct
    read_acl validation; however it is now also required to add ".rlistings" to the
    read acl.
    
    Also, if the staticweb middleware is put in the proxy pipeline before an
    authentication middleware, it broke authenticated GET and HEAD requests. This
    has some side effects in clients, because a html response is sent that might be
    parsed wrongly by the client. In case of python-swiftclient this was shown as an
    empty container without any ACL or web-listings:true meta set. This might lead
    to information leaks, because a user trusts the output from python-swiftclient
    and assumes an empty, private container even if the container contains public
    readable data. staticweb now checks if "swift.authorize" is included in the
    environ and skips itself if not.
    
    Closes-Bug: 1489749
    Change-Id: Icf159d7e567ac5481e710c5910db686bdcba6336
    Depends-On: Ie24bb995023c377e49796910ad80a256b00daa03

commit 2996974e5d48b4efaa1b271b8fbd0387bced7242
Author: Ondřej Nový <email address hidden>
Date:   Sat Oct 10 14:56:30 2015 +0200

    Script for running unit, func and probe tests at once
    
    When developing Swift it's often needed to run all tests.
    This script makes it much simpler.
    
    Change-Id: I67e6f7cc05ebd0475001c1b56e8f6fd09c8c644f

commit c2182fd4163050a5f76eb3dedb7703dc821fa83d
Author: janonymous <email address hidden>
Date:   Fri Jul 17 20:20:15 2015 +0530

    Python3: do not use im_self/im_func/func_closure
    
    Use __self__, __func__ and __closure__ instead, as they work
    with both Python 2 and 3.
    
    Modifying usage of __func__ in codebase.
    
    Change-Id: I57e907c28c1d4646605e70194ea3650806730b83

commit c0866ceaac2f69ae01345a795520141f59ec64f5
Author: Samuel Merritt <email address hidden>
Date:   Fri Sep 25 17:26:37 2015 -0700

    Improve SLO PUT error checking
    
    This commit tries to give the user a reason that their SLO manifest
    was invalid instead of just saying "Invalid SLO Manifest File". It
    doesn't get every error condition, but it's better than before.
    
    Examples of things that now have real error messages include:
     * bad keys in manifest (e.g. using "name" instead of "path")
     * bogus range (e.g. "bytes=123-taco")
     * multiple ranges (e.g. "bytes=10-20,30-40")
     * bad JSON structure (i.e. not a list of objects)
     * non-integer size_bytes
    
    Also fixed an annoyance with unspecified-size segments that are too
    small. Previously, if you uploaded a segment reference with
    '{"size_bytes": null, ...}' in it and the referenced segment was less
    than 1 MiB, you'd get a response that looked like this:
    
        HTTP/1.1 400 Bad Request
        Content-Length: 62
        Content-Type: text/html; charset=UTF-8
        X-Trans-Id: txd9ee3b25896642098e4d9-0055dd095a
        Date: Wed, 26 Aug 2015 00:33:30 GMT
    
        Each segment, except the last, must be at least 1048576 bytes.
    
    This is true, but not particularly helpful, since it doesn't tell you
    which of your segments violated the rule.
    
    Now you get something more like this:
    
        HTTP/1.1 400 Bad Request
        Content-Length: 49
        Content-Type: text/plain
        X-Trans-Id: tx586e52580bac4956ad8e2-0055dd09c2
        Date: Wed, 26 Aug 2015 00:35:14 GMT
    
        Errors:
        /segs/small, Too Small; each segment, except the last...
    
    It's not exactly a tutorial on SLO manifests, but at least it names
    the problematic segment.
    
    This also changes the status code for a self-referential manifest from
    409 to 400. The rest of the error machinery was using 400, and
    special-casing self-reference would be really annoying. Besides, now
    that we're showing more than one error message at a time, what would
    the right status code be for a manifest with a self-referential
    segment *and* a segment with a bad range? 400? 409? 404.5? It's much
    more consistent to just say invalid manifest --> 400.
    
    Change-Id: I2275683230b36bc273319254e37c16b9e9b9d69c

commit 84f0a544452fefc65f6755fdd04cc2cc1eb11c88
Author: Victor Stinner <email address hidden>
Date:   Thu Oct 8 15:42:27 2015 +0200

    py3: Replace basestring with six.string_types
    
    The builtin basestring type was removed in Python 3. Replace it with
    six.string_types which works on Python 2 and Python 3.
    
    Change-Id: Ib92a729682322cc65b41050ae169167be2899e2c

commit 2447e0cf6ed502375a80e33e6aa4a5110305fa6f
Author: Victor Stinner <email address hidden>
Date:   Thu Oct 8 15:41:02 2015 +0200

    Replace itertools.ifilter with six.moves.filter
    
    Replace itertools.ifilter() with six.moves.filter(). The
    itertools.ifilter was removed in Python 3. This change makes modified
    code compatible with Python 2 and Python 3.
    
    Replace itertools.izip() with six.moves.zip().
    
    The patch was generated by the itertools operation of the sixer tool.
    
    Change-Id: Ie7f787cc6d66edfceb8fa2c1a906351a8c8c5fed

commit 752ceb266b3e72361242250e50396671df8b9595
Author: Clay Gerrard <email address hidden>
Date:   Fri Oct 2 16:40:28 2015 -0700

    Close ECAppIter's sub-generators before propagating GeneratorExit
    
    ... which ensures no Timeouts remain pending after the parent generator
    is closed when a client disconnects before being able to read the entire
    body.
    
    Also tighten up a few tests that may have left some open ECAppIter
    generators lying about after the tests themselves had finished.  This
    has the side effect of preventing the extraneous printing of the Timeout
    errors being raised by the eventlet hub in the background while our
    unittests are running.
    
    Change-Id: I156d873c72c19623bcfbf39bf120c98800b3cada