Reviewed:  https://review.opendev.org/677848
Committed: https://git.openstack.org/cgit/openstack/swift/commit/?id=6204a2bc571a51e91f97ad4342b8ae216aaf6f36
Submitter: Zuul
Branch:    feature/losf

commit ff5ea003b3a1b37d8417aa17d3521237768dfe62
Author: Tim Burke <email address hidden>
Date:   Tue Aug 20 22:20:44 2019 -0700

    ec: log durability of frags that fail to reconstruct
    
    Whether the frag is durable or non-durable greatly affects how much I
    care whether I can reconstruct it.
    
    Change-Id: Ie6f46267d4bb567ecc0cc195d1fd7ce55c8cb325

commit 94366fd00ec1e71b4f67ef8595b31162c6945c28
Author: Thiago da Silva <email address hidden>
Date:   Mon Aug 19 22:31:41 2019 +0200

    Add Dockerfile to build a py3 swift docker image
    
    This patch contains only a new Dockerfile and few changes to
    be able to build both py2 and py3 images.
    
    Next patch should contain changes to add the gate jobs to build
    a py3 docker image
    
    Change-Id: Ifdebde9597a787abcd553756e22261e2faaeedfc

commit e92191306587335ad0aab3a3b9dffef21765e17b
Author: Clay Gerrard <email address hidden>
Date:   Mon Aug 19 11:06:20 2019 -0500

    Rename symlink method
    
    Related-Change-Id: I179ea6180d31146bb947061c69b1807c59529ac8
    Change-Id: I770ea1be25e339e5ce4341b532a1fff9e1373152

commit 82e427a8b14e3efee2da99d622fad04ed6461a50
Author: Clay Gerrard <email address hidden>
Date:   Mon Aug 19 10:56:50 2019 -0500

    Fix symlink docstring
    
    Related-Change-Id: I179ea6180d31146bb947061c69b1807c59529ac8
    Change-Id: I41cbf445057be09a102ef5515a9ce6a4f14e7458

commit a21c0be70ca6c200076d274c54f6906e1cf92a61
Author: Tim Burke <email address hidden>
Date:   Wed Jun 5 12:02:07 2019 -0700

    dlo: Respond 200 on multi-range GETs
    
    We already had a unit test for this, but it was inadvertently retesting
    what test_get_suffix_range_many_segments tested.
    
    Change-Id: I43eee7029ca884268c3f2ad74300b94b299fd8d2
    Closes-Bug: #1831790

commit 7f42309a335370f567aefdf0236cfe01cbbfca5d
Author: Tim Burke <email address hidden>
Date:   Thu Mar 21 15:13:36 2019 -0700

    py3: Cover account/container func tests
    
    Change-Id: I9a754f1eb06debbe16800a05fc8e969af2f89043

commit 3e7752d8c80473e93d537c7ba30419ff9313ebb8
Author: Thiago da Silva <email address hidden>
Date:   Fri Aug 16 18:30:13 2019 +0200

    Update docker image to latest Alpine 3.10.1
    
    Change-Id: I888c51a965ee27175584f575560a1d50f4534663

commit 1abc9c4f9d4c2f9d8ac996cca489b0ec65660a05
Author: Tim Burke <email address hidden>
Date:   Thu Jan 24 00:23:01 2019 +0000

    Allow "static symlinks"
    
    ... by embedding something like `If-Match: <etag>` semantics in the
    symlink.
    
    When creating a symlink, users may now specify an optional
    X-Symlink-Target-Etag header. If present, the etag of the final object
    returned to the client will be checked; if it does not match the
    X-Symlink-Target-Etag header, a 409 Conflict error will be returned to
    the client.
    
    Note that, unlike "dynamic symlink" behavior, the target object must
    exist with the matching Etag for the "static symlink" to be created.
    
    Since we're validating the Etag anyway, we also set the content-type of
    the symlink to match if the client didn't otherwise specifiy and send
    the etag & content-length along to the container listing as well.
    
    Bonus goodness:
    
    - Tighten assertions on Content-Location
    - Get rid of swift.source-sniffing by making versioned_writes
      symlink-aware ('cause I'm going to want to make it symlink-aware
      later anyway)
    - Allow middlewares left of symlink to set their own
      Container-Update-Override-Etag when creating a symlink
    - Set dynamic symlink content type if client doesn't supply something
    
    Co-Authored-By: Clay Gerrard <email address hidden>
    Change-Id: I179ea6180d31146bb947061c69b1807c59529ac8

commit 0c16fd9536ac37c838c4229800abdb684a7bd4d4
Author: Alexandra Settle <email address hidden>
Date:   Fri Aug 16 12:44:13 2019 +0100

    Fixing broken links
    
    Small changes, but helpful, mostly.
    
    Backport: stein rocky
    
    Change-Id: Ic4b6524d7804d2f74b2973b6acdb9e2679209cd4

commit 27e7e80e922a51f3f4d21aa4d58f84fa9b5c6621
Author: Tim Burke <email address hidden>
Date:   Mon Jul 15 18:07:01 2019 -0700

    py3: fix up swift-orphans
    
    Change-Id: Id1280abd92e8bb02fcaa4701a0e9d211d9d6e33e

commit 9e9c363361bb3d8aaef977efaabea9f0ba52dbe3
Author: Thiago da Silva <email address hidden>
Date:   Wed Aug 14 18:00:50 2019 +0200

    fix ec overview doc formatting
    
    Change-Id: I3fee824e4fdc21aac21a6bb97f24a82d8ea3577d

commit 47c7058259b9f69b51d92ee806988db6ca37f28c
Author: Thiago da Silva <email address hidden>
Date:   Wed Aug 14 16:13:25 2019 +0200

    Add params to get_*_metadata internal client methods
    
    Change-Id: I1d2d063904447a444ecb506b93ee698a5181b6b9

commit cb308119160d5af7db9b65a7ab2225bb07edb614
Author: Samuel Merritt <email address hidden>
Date:   Thu Feb 1 12:03:49 2018 -0800

    Allow bulk delete of big SLO manifests
    
    If you set SLO's max_manifest_segments to a value larger than 10000,
    then clients are able to create manifests with that many segments, but
    unable to use "?multipart-manifest=delete" to delete them.
    
    This is because the SLO middleware has its very own bulk-deleter that
    it uses to handle such requests, and that bulk-deleter only allows
    10000 deletions per request by default. This commit removes the
    limitation so that any SLO manifest can be deleted along with its
    segments.
    
    I considered setting max-deletes-per-request to be equal to SLO's
    max_manifest_segments, but that only works if max_manifest_segments
    has never been decreased.
    
    Note that this commit does not increase max_manifest_segments. Clients
    cannot make SLOs any bigger than they could before. Also note that
    this commit does not affect user-initiated bulk deletes, i.e. POST
    requests with "?bulk-delete=true" set. Those requests are still
    limited in their size, and those limits are not changed.
    
    Change-Id: I6a35937e8418f4f2b8e29825fc9c40415e34742f
    Closes-Bug: 1746685

commit 227fe30251984feb45b761390a4ec1d8f5cc894d
Author: Tim Burke <email address hidden>
Date:   Fri Aug 9 15:59:59 2019 -0700

    py3: port test/functional/test_versioned_writes.py
    
    Change-Id: I8c2517599a6d1cdf68050dc7acb039f09875fd53

commit f05119c16fb33140a655b30e76734f481a479412
Author: Tim Burke <email address hidden>
Date:   Mon Aug 5 17:33:23 2019 -0700

    py3: mostly port s3 func tests
    
    test_bucket.py is proving somewhat problematic.
    
    Change-Id: I5b337ef66a23fc989762801dd6a5ba1ed903f57b

commit c90365f98d64a38aa6a76a7efcd803470be384bd
Author: Pete Zaitcev <email address hidden>
Date:   Fri Aug 9 15:06:47 2019 -0500

    Allow Connection: value to be case-insensitive
    
    Proxies and LBs parse the Connection: header and re-insert
    the value with random casing. This occurs when we run functional
    tests against realistic clusters.
    
    Change-Id: Icb0d3e3708949127e9750f2022de8c74ce51ebe9

commit 996aa4547f5248c49e47478f186f3104e1c6c5da
Author: Clay Gerrard <email address hidden>
Date:   Thu Aug 8 14:16:00 2019 -0500

    Consolidate Container-Update-Override headers
    
    Related-Change-Id: I179ea6180d31146bb947061c69b1807c59529ac8
    Related-Change-Id: I056edc68aee8c0db2a2c4a5b9e3d242a895975b3
    
    Change-Id: I84bd29ae48ff1b0826794a8fdf9aa87670ad4aa4

commit 3ee6de408e8507c8501a9abb17fbf9e08fd65c17
Author: Tim Burke <email address hidden>
Date:   Thu Aug 8 14:05:06 2019 -0700

    slo: Add X-Manifest-Etag to responses
    
    This matches the ETag of the underlying swift object, as opposed to the
    MD5-of-MD5s that is the large object's ETag.
    
    Change-Id: Ifab726f63739f62aeef495c970939410341694d1

commit e095bf3dabad005d48985f06f6bfe14683621a59
Author: Tim Burke <email address hidden>
Date:   Thu Aug 8 08:54:49 2019 -0700

    Bring our IPv6 setup more in line with devstack's
    
    Change-Id: I4a32239eec37eed4085fbc2a283def4b69c816ac

commit 7c74dbd255eac241e70745f269ba9383c124821a
Author: Tim Burke <email address hidden>
Date:   Wed Apr 17 10:15:26 2019 -0700

    py3: port dlo func tests
    
    Drive-by: fix an issue where tf.config wasn't initialized when we decide
    whether to skip test_dlo_referer_on_segment_container.
    
    Change-Id: Iad8b3ab40dccb23847cde5c09f55627da0963fe0

commit 7bb7da6d2dc4267f8279170b6d326e81777d3c06
Author: Tim Burke <email address hidden>
Date:   Tue Aug 6 11:11:52 2019 -0700

    s3api: Translate 503 to S3-style 503s
    
    ...instead of logging tracebacks about unexpected status codes.
    
    Change-Id: Iadb0a40092b8347eb5c04785cc14d1324cc9396f

commit 2c727f65ffae3d925fb825e20616615ab319d1d2
Author: Tim Burke <email address hidden>
Date:   Tue Aug 6 10:00:41 2019 -0700

    py3: Add swift-tox-func-encryption-py37 gate job
    
    Change-Id: Ia9ae0fc226dfc9b40157faebac100c10a9180c62

commit a10d936ee57c5eb331b56965fa79779e77978c84
Author: Tim Burke <email address hidden>
Date:   Tue Aug 6 10:14:47 2019 -0700

    Use `is` to compare against sentinel object
    
    ...since `==` can have repercussions by way of __eq__()
    
    Change-Id: I668b11251a3d1192d42dfe914010d089df413da7

commit 2e4c8a924e2c11e0cbabfd1353de49602d1aaef1
Author: Tim Burke <email address hidden>
Date:   Mon Aug 5 14:47:51 2019 -0700

    py3: port RBAC func tests
    
    Yes, it really is just the import.
    
    Change-Id: Id7e2891b6600af102176eaa2e87f81b528adde8d

commit ab192415347223cfb49af7a62e124ef6f35779a6
Author: Clay Gerrard <email address hidden>
Date:   Mon Aug 5 15:40:58 2019 -0500

    Increase node_timeout for real sever unittests
    
    I saw an encryption test fail with 10s node timeout when the py36
    unittests ran in the gate.  I believe there is an outsized potential for
    a beneficial increase in test reliability when run in the gate compared
    to relatively small chance of any negative side-effect in test failure
    responsiveness.
    
    Change-Id: Ia31912828d416d84c39782222e4636a97a8bfe44

commit 25aeb0ca4905756501e62210a5bc2699c731a1f4
Author: Clay Gerrard <email address hidden>
Date:   Thu Dec 6 16:36:16 2018 -0800

    Make GreenAsyncPile not hang
    
    It's probably weird that StreamingPile has this interfaces that swallows
    exceptions, but this seems better than hanging.
    
    Change-Id: I8fe45c0f0d291efc84f3edf5d6b7cd116b5c7835

commit 3189410f9d5be3a845dbd47fdd77584f4b76da8a
Author: Tim Burke <email address hidden>
Date:   Mon Jul 22 12:38:30 2019 -0700

    Ignore 404s from handoffs for objects when calculating quorum
    
    We previously realized we needed to do that for accounts and containers
    where the consequences of treating the 404 as authoritative were more
    obvious: we'd cache the non-existence which prevented writes until it
    fell out of cache.
    
    The same basic logic applies for objects, though: if we see
    
        (Timeout, Timeout, Timeout, 404, 404, 404)
    
    on a triple-replica policy, we don't really have any reason to think
    that a 404 is appropriate. In fact, it seems reasonably likely that
    there's a thundering-herd problem where there are too many concurrent
    requests for data that *definitely is there*. By responding with a 503,
    we apply some back-pressure to clients, who hopefully have some
    exponential backoff in their retries.
    
    The situation gets a bit more complicated with erasure-coded data, but
    the same basic principle applies. We're just more likely to have
    confirmation that there *is* data out there, we just can't reconstruct
    it (right now).
    
    Note that we *still want to check* those handoffs, of course. Our
    fail-in-place strategy has us replicate (and, more recently,
    reconstruct) to handoffs to maintain durability; it'd be silly *not* to
    look.
    
    UpgradeImpact:
    --------------
    Be aware that this may cause an increase in 503 Service Unavailable
    responses served by proxy-servers. However, this should more accurately
    reflect the state of the system.
    
    Co-Authored-By: Thiago da Silva <email address hidden>
    Change-Id: Ia832e9bab13167948f01bc50aa8a61974ce189fb
    Closes-Bug: #1837819
    Related-Bug: #1833612
    Related-Change: I53ed04b5de20c261ddd79c98c629580472e09961
    Related-Change: Ief44ed39d97f65e4270bf73051da9a2dd0ddbaec

commit 7678e94a2c3907df13d271b0d4c82054bbb0bf8a
Author: Clay Gerrard <email address hidden>
Date:   Thu Aug 1 09:31:56 2019 -0500

    Negative test for non-empty chunked put symlink
    
    Change-Id: I0c3a235b8905e02d2e202215f7f0b43db295b79b
    Related-Change-Id: I66f2f16d7cf0b9cb0888e3b9b1a17b9112233d21

commit c19d09f60b2d501698b741a123cc6b8203969016
Author: Tim Burke <email address hidden>
Date:   Mon Jun 17 15:30:40 2019 -0700

    Give ECAppIter greenthreads a chance to wrap up
    
    Otherwise, we can hit a "generator already executing" error in
    test/functional/test_slo.py:TestSlo.test_slo_multi_ranged_get
    
    Also, set a *slightly* newer lower-bound for greenlet (from 2011 instead
    of 2010) so the sleep() doesn't cause us to lose exception state.
    
    Change-Id: Id458fc0a43fd71aab9f6edbacc1f0b370c9f7537

commit a1455b3a3566a1a0d5680e540d19ba6dc417b1db
Author: Tim Burke <email address hidden>
Date:   Fri Jul 26 21:29:42 2019 -0700

    symlink: Allow symlinks to be created via chunk-encoded PUTs
    
    ... provided they're still zero bytes.
    
    Change-Id: I66f2f16d7cf0b9cb0888e3b9b1a17b9112233d21

commit 8635e8cc867a5f44f6269b1f85b2a3d360a35561
Author: Tim Burke <email address hidden>
Date:   Fri Jul 26 15:59:35 2019 -0700

    Make backport releases conform to existing CHANGELOG formatting
    
    Change-Id: Ib121d6c52601488647fd16dfd4f49bb4492c7a2f

commit 02cfb6c02779d9bfe27b05cea0fd6dee7e9d9351
Author: Ghanshyam Mann <email address hidden>
Date:   Thu Jul 18 19:11:39 2019 +0000

    Add integrated gate tempest and grenade job
    
    Swift gate does not run integrated-gate jobs which are
    tempest-full and grenade-py3 job. Tempest-full job run
    all services tests. But all services tests are not needed
    to run on swift gate because it does not depends on all services.
    
    Tempest has prepared the new jobs for swift gate which will run
    only Swift, Cinder and Glance related tests.
    
    New template 'integrated-gate-object-storage'(Depends-On) includes
    the tempest-integrated-object-storage and grenade-py3 jobs
    
    This commit adds the new integrated-gate template in swift gate.
    
    Depends-On: https://review.opendev.org/#/c/670353/
    Change-Id: Ia0543dca729c3a2aa67ce9f560c1ce5976c00abf

commit 097a62cad35b90966b6d2e1e40714eb67f5bdc5a
Author: Andreas Jaeger <email address hidden>
Date:   Fri Jul 26 11:49:15 2019 +0200

    Fix broken API-ref link
    
    The URL is object-store, not object-storage. Fix one occurence.
    
    Change-Id: I08ed3b19cbe0024fd67a6fb7a7a6db0d9084976b

commit 71f83963c2b701bdc6ec67e97429dac661e8ffc6
Author: Tim Burke <email address hidden>
Date:   Wed Jul 24 16:59:08 2019 -0700

    py3: fix non-ascii metadata handling in account-server
    
    Previously, we were storing the WSGI-style UTF-8-bytes-decoded-as-Latin-1
    strings in the JSON field, and sending them back to eventlet directly.
    If running in a mixed py2/py3 cluster, replication would eventually get
    that back to the py2 server, and worse, the native-string version would
    get back to the py3 server! Then on GET or HEAD, eventlet would barf
    if any characters were outside the Latin-1 range.
    
    Closes-Bug: #1837805
    Change-Id: I31d942e72fd7bfbb1db4dbb1dd522dff69969e5d

commit 60c27d3aef6e7412b41e04445d72aa62ea9883d3
Author: Tim Burke <email address hidden>
Date:   Thu Jul 25 10:56:53 2019 -0700

    py3: Fix title-casing in HeaderKeyDict
    
    Change-Id: I1152c47c52f6482ec877142c96845b00bf6dcc5b
    Related-Change: I130ba5014b7eff458d87ab29eb42fe45607c9a12

commit 73ec8f15df9a1f112f981f15a3ddc323eb1fb723
Author: Andreas Jaeger <email address hidden>
Date:   Mon Jul 22 18:53:40 2019 +0200

    Update api-ref location
    
    The api documentation is now published on docs.openstack.org instead
    of developer.openstack.org. Update all links that are changed to the
    new location.
    
    Note that the Swift API lives at /object-store and not /object-storage.
    
    Note that redirects will be set up as well but let's point now to the
    new location.
    
    For details, see:
    http://lists.openstack.org/pipermail/openstack-discuss/2019-July/007828.html
    
    Change-Id: Ie38357e4c278335c35d186708573bb6bdabaa012

commit a48104c5454a216464313cd3473297c113069450
Author: Tim Burke <email address hidden>
Date:   Thu May 16 14:37:42 2019 -0700

    py3: port SLO func tests
    
    Drive-by: remove some py2/py3 code branches where encoding/decoding
    should always succeed.
    
    Change-Id: Iabfa157f2b20e6fd650a441e9c0f6163478ce147

commit ef5a37c2bf5a32b0fa6404974aa76332a88c0841
Author: Tim Burke <email address hidden>
Date:   Fri Jul 19 21:24:47 2019 -0700

    slo: Better handle non-manifest responses when refetching manifest
    
    Previously, we never checked whether the response we get when refetching
    is even successful, much less whether it's still coming from an SLO.
    
    Now, if the refetched data is newer, act on it. If it's older, 503.
    
    Closes-Bug: #1837270
    Change-Id: I106b94c77da220c762869aa800c31b87c3dffeeb

commit 12a7b420621ae45a40cff3af01ec22078ad1756b
Author: Thomas Goirand <email address hidden>
Date:   Mon Jul 15 19:53:14 2019 +0200

    Fix test_parse_get_node_args
    
    Looks like xattr_supported_check was missing ERANGE
    
    Change-Id: I82263e48e836f38f77d81593c8435f64a4728b5d

commit 2926fc96a36a573380eca18d60440a2337442015
Author: Tim Burke <email address hidden>
Date:   Fri Mar 22 14:48:10 2019 -0700

    py3: Bring functional/test_object.py under test; add func-ec-py37 job
    
    Change-Id: I2929ce1f8e08ae5b4059605d09189c8521852423

commit a65350e887a53460ac18cc6c1c4fbb445aeee7e7
Author: Tim Burke <email address hidden>
Date:   Wed Jul 17 12:10:05 2019 -0700

    Make rolling-upgrade job voting again
    
    Now that os-testr has had a release, it passes again.
    
    Change-Id: Ieed88646d97ab2a1c7323a147ee97dae6fc86a0f

commit 424a2603d8d4a454f269c9df5b4c7dc0d389b9da
Author: Tim Burke <email address hidden>
Date:   Tue Jul 16 16:59:38 2019 -0700

    Fix up errno checking
    
    Change-Id: I196b4e886942eccd2cb3aff3819da1558ddb19d4

commit ebf561026f400a8e1c5e3984cb3034fc095fffa6
Author: Tim Burke <email address hidden>
Date:   Mon Jul 15 18:07:51 2019 -0700

    py3: fix object-replicator rsync output parsing
    
    Change-Id: I07707e99e4232b2cbe14986b958ce43c003a1ab6

commit a48ac28efe5c133b439ab0e525d315271bf370da
Author: Tim Burke <email address hidden>
Date:   Fri Jun 28 14:58:42 2019 -0700

    Authors/changelog for 2.22.0
    
    Change-Id: I09c099339d02943baa0944b5a208263536d1a547

commit bc5f4c061122aee5965cfb055beb3cb079073536
Author: Tim Burke <email address hidden>
Date:   Thu Apr 25 12:57:16 2019 -0700

    sharding: better handle get_shard_ranges failures
    
    The contract for ReplConnection.replicate() is that if we can get a
    response, we return it, and if we can't (because of a timeout, or a
    socket error, or some other http_client error like BadStatusLine), we
    return None to indicate the error.
    
    Previously, _fetch_and_merge_shard_ranges() always assumed the
    response would have a `status` attribute and raise an AttributeError
    when response was None. Now it will treat that case like other
    get_shard_range failures.
    
    Change-Id: I023b8a46c06e9a2755b5aa890a7992ef9633cba9

commit 3532df0fd3dc91af5534b1cb3d4de4bfec101b8f
Author: Corey Bryant <email address hidden>
Date:   Fri Jul 5 16:22:56 2019 -0400

    Add Python 3 Train unit tests
    
    This is a mechanically generated patch to ensure unit testing is in place
    for all of the Tested Runtimes for Train.
    
    See the Train python3-updates goal document for details:
    https://governance.openstack.org/tc/goals/train/python3-updates.html
    
    Change-Id: I29f2951053fa6714de46ed757116d99dd5c896cf
    Story: #2005924
    Task: #34249

commit 0ebfeddf65adb473fda2adfd1705534b6308cd2b
Author: Tim Burke <email address hidden>
Date:   Mon Jun 17 13:22:04 2019 -0700

    Bump up our minimum eventlet version
    
    We need 0.25.0 for py3 support, and increasing it for both is simplest.
    
    Also drag forward dnspython, as newer eventlet requires a newer
    dnspython. Note that we still don't explicitly list it as a dependency
    for py3; rather, it's been getting installed as a requirement for
    eventlet.
    
    Change-Id: If933b713bdd1dfc348959771d60c0756a9a94477

commit fbf6d847966059db8fb41eaa7952b7f802f16cc6
Author: Tim Burke <email address hidden>
Date:   Fri May 3 13:36:55 2019 -0700

    Make py36 job voting
    
    Change-Id: I4768fe41d6347bcae44774730350f4be583bdf15