commit c130cf3f9a81e01d99bcc8b371dc6b05b9eee5e8
Author: Dan Callaghan <dcallagh@redhat.com>
Date:   Thu Apr 17 16:17:00 2014 +1000

    Automatic commit of release 0.16.2

commit 9927f0f0687027708a6a866c41e0f17fd33bd35c
Author: Dan Callaghan <dcallagh@redhat.com>
Date:   Wed Apr 16 17:03:43 2014 +1000

    release notes for 0.16.2
    
    Change-Id: I75d06caff7e2a6a3719849609e1320fb368cedfd

commit 7183a29d6c3ed18c2f0068b9c42554ea1a336d92
Author: Dan Callaghan <dcallagh@redhat.com>
Date:   Wed Apr 16 14:15:42 2014 +1000

    tests: fix typo in error message in test_system_loan
    
    The typo was always here, but prior to commit 856aff0a the error
    messages were never actually checked so nobody noticed.
    
    Change-Id: I1f2ef1ce896b99fdb6eeb58ec705824685484bf0

commit 3534082658c38d99ff7d6d8bc95255c2df907b3a
Author: Amit Saha <asaha@redhat.com>
Date:   Tue Apr 8 11:01:17 2014 +1000

    Do not export 'secret' column in CSV export of systems
    
    Bug: 1085047
    Change-Id: I9d298dcf63e372c1f54e3fefa266eec804caaa4a

commit d25bd4af9045b09e0d4b3e57f2a1b951b14934a8
Author: Amit Saha <asaha@redhat.com>
Date:   Tue Apr 8 11:01:17 2014 +1000

    tests: fix test_csv_roundtrip to upload a valid CSV file
    
    Change-Id: Ic557e4f091e465e44ab5d01b5ab230eb60ea8372

commit 856aff0abe7aaedf27545ab6d6f14bc494ef5a0d
Author: Matt Jia <mjia@redhat.com>
Date:   Fri Apr 11 15:27:34 2014 +1000

    System loan management:display the actual error messages
    
    Bug: 999391
    Change-Id: I7af9dba866a86038914db6ebed760ebde75a5b3e

commit 72cffa7536594cbec9217cc7182e1f7ffec4e358
Author: Amit Saha <asaha@redhat.com>
Date:   Wed Apr 9 17:19:40 2014 +1000

    CSV import: Use an UTF-8 aware CSV importer
    
    This patch makes the CSV import handling code work with UTF-8 data
    rather than depending on the database layer to store them as such.
    
    Change-Id: Iea384b55df2562488abd6eb94dbc84b510e72c2f

commit 17d0f94c50af2823672c1729f9ba0383badf5628
Author: Amit Saha <asaha@redhat.com>
Date:   Thu Apr 10 12:06:05 2014 +1000

    bkr workflow-installer-test: Fix inconsistent output
    
    Bug: 1078610
    Change-Id: Iadeebd800686c75d127b402594f10ee795657e0e

commit 5fd20c73269bc14b9e79fc2adca5816643f6a16f
Author: Raymond Mancy <rmancy@redhat.com>
Date:   Wed Apr 2 23:38:04 2014 +1000

    Do not return removed lab controllers in 'bkr list-labcontrollers'
    
    Bug: 1083562
    Change-Id: Ib36e20175a4e5f011698933855a6e5c12d65f153

commit 5a1793e4702c15febae02ce5147275ed6fc4541b
Author: Dan Callaghan <dcallagh@redhat.com>
Date:   Mon Apr 7 11:31:38 2014 +1000

    look up packages from recipe directly instead of going through ksmeta
    
    The fact that the recipe packages were passed as a colon-separated
    string in ksmeta was just a relic from when that was necessary in
    Cobbler. Now we can grab the packages from the recipe object directly,
    avoiding the problem with colons being mangled.
    
    However, to avoid breaking compatibility we still need to accept
    'packages' is ksmeta. The current behaviour is that it replaces any
    packages coming from the recipe, so that behaviour is preserved in this
    patch.
    
    Bug: 952635
    Change-Id: I487abbb40b757d5490a538ddb71f1594f91d7c12

commit 54521b1d3a2b9208742f32c852aee71b327612c8
Author: Amit Saha <asaha@redhat.com>
Date:   Tue Apr 8 18:22:01 2014 +1000

    Report an error when an empty CSV file is imported
    
    Bug: 1085238
    Change-Id: Ibcfca94ea4deb56355ecc6cb797947148a796b04

commit eb04df779d8ff9b8693429ecbe15d851d3c40652
Author: Dan Callaghan <dcallagh@redhat.com>
Date:   Tue Apr 8 12:40:39 2014 +1000

    use a nested transaction each row when importing CSV
    
    This avoids leaking updates to the database in case a row fails to be
    imported for some reason.
    
    Bug: 1085149
    Change-Id: I626fbfca3c3c8a352384e5b9b8eae235d4a218fb

commit 33b6c18ae3673f449c1c62a125d8e61becda084f
Author: Matt Jia <mjia@redhat.com>
Date:   Tue Apr 8 11:36:27 2014 +1000

    access policy editor: encode group name in the url
    
    bug: 1085028
    Change-Id: I8ae45efe995d8373c460e692a4004eb65b3d3263

commit be4da184c85f214dc6f4dafb15f3995c592e116e
Author: Matt Jia <mjia@redhat.com>
Date:   Fri Mar 28 09:21:45 2014 +1000

    access policy editor: make group name linkable
    
    Currently the group controller only handles group id.
    With this patch it also handles group name which can be used to link
    to the edit page.
    
    Bug: 1073767
    Change-Id: I4fc6d454b13eafea438fd17c215d29d4e3707556

commit 8112839c1a5ff4406ca5d1e6961f3a9f37ba9361
Author: Amit Saha <asaha@redhat.com>
Date:   Mon Mar 31 16:25:54 2014 +1000

    Allow disabling IPv6 in Beah
    
    In addition to beah_beaker.conf, write the IPV6_DISABLED
    configuration option in beah.conf as well.
    This is a second commit to implement the beaker side of the patch.
    (First commit: c24b88f)
    
    Bug: 1065811

commit 2d876707620e216f346d13de482c14f36270840c
Author: Dan Callaghan <dcallagh@redhat.com>
Date:   Mon Mar 31 13:15:22 2014 +1000

    beaker-init: add admin user as owner of admin group
    
    Bug: 966339
    Change-Id: I798868d064e3ae541758bad84dc23d2536a54a57

commit 8aa3e9760bb42bd41de4c8879f496ae84442bb41
Author: Amit Saha <asaha@redhat.com>
Date:   Fri Mar 28 13:07:31 2014 +1000

    Use 'systemctl' instead of 'chkconfig' on distros using systemd
    
    This patch also adds a new global "systemd" to the kickstart generation
    templates so that it can be directly used when such a check is desired.
    
    Change-Id: I040c8608415130edbf6636ab8da0e0e8842e02fe

commit 5c8b17c8d274e3820566be86c7cfcdac681ca161
Author: Amit Saha <asaha@redhat.com>
Date:   Fri Mar 28 15:59:13 2014 +1000

    Replace "Fedora" by "Fedora20" as the OS Major
    
    Fedora is not really a correct OS Major, so replacing it by a correct one.

commit 865856267136297e3871848be98e6e8ec35e6dca
Author: Matt Jia <mjia@redhat.com>
Date:   Thu Mar 13 10:58:56 2014 +1000

    Run pylint when building a RPM
    
    Change-Id: I5573272690353703e8ea719f180126830582917a

commit 3de4c484cdcadd210aa9966cb9e7e26f70f75280
Author: Dan Callaghan <dcallagh@redhat.com>
Date:   Tue Mar 25 13:09:03 2014 +1000

    pylint: suppress some additional pylint false positives
    
    Change-Id: I22848653924c69f3b57dee034009c5c51415beeb

commit 3503222b1acec325a619e8a3686aa0b8b25498d4
Author: Dan Callaghan <dcallagh@redhat.com>
Date:   Tue Mar 25 13:08:01 2014 +1000

    pylint: fix minor errors identified by pylint
    
    Change-Id: I9b0af80ce4b2f44f6469c9185a33c1ba039b7d22

commit ff7a1f59bf3b195eb1e86d938b694ccc91ad8236
Author: Dan Callaghan <dcallagh@redhat.com>
Date:   Tue Mar 25 13:03:32 2014 +1000

    pylint: decorate SQL expression hybrid methods with @classmethod
    
    The SQL expression half of a hybrid method really is a "classmethod" in
    the sense that it receives a class and not an instance as its first
    argument. So it makes sense to decorate it with @classmethod. That way
    we also keep pylint happy.
    
    Change-Id: I2d98d8c56597831f5f965ab2496b31903af50001

commit 39c1602cc20e6a34688003182178898805c7999d
Author: Dan Callaghan <dcallagh@redhat.com>
Date:   Tue Mar 25 11:43:18 2014 +1000

    pylint: rename importer mix-in classes to have Mixin in their name
    
    This way pylint will ignore missing members in the methods of these
    classes.
    
    Change-Id: I4c9968e7f3ce2ee7da02962cbaa9101ea4df35a9

commit de07e06cb93bab49f2dbfaaa6a5cdd9c163a693a
Author: Dan Callaghan <dcallagh@redhat.com>
Date:   Tue Mar 25 11:28:44 2014 +1000

    pylint: qpid support was removed from client and LC in commit 27b6f19a
    
    Change-Id: I88c2934fbef4730f9bebc06ab330e64528ce2b54

commit 9772453dc214abd6c90e09f74ef3a6bdbbd17bc4
Author: Dan Callaghan <dcallagh@redhat.com>
Date:   Tue Mar 25 10:55:40 2014 +1000

    pylint: _parse_response method does not exist in Python 2.7
    
    This method was removed in favour of parse_response which just uses the
    buffered socket instead of the raw socket. We can call parse_response
    unconditionally since it is defined on earlier Python versions too.
    
    Change-Id: Ia18877cbb468e41b800b2b7ac717c2bd5b9d93b1

commit 161a8dca9295acbfe3535c158cfb30981e61bc59
Author: Dan Callaghan <dcallagh@redhat.com>
Date:   Tue Mar 25 10:54:42 2014 +1000

    pylint: convert mix-in to helper functions in test_schema.py
    
    Pylint didn't like the mix-in methods referencing instance attributes
    which are not defined on the mix-in.
    
    Change-Id: I24e4d5f095b27b281ced8868224732e473851a24

commit 56339b6aa77145dc220057c94d381139f4cd57fb
Author: Dan Callaghan <dcallagh@redhat.com>
Date:   Tue Mar 25 10:53:02 2014 +1000

    pylint: subtask watching was "temporarily" disabled in commit 5eee47b1f
    
    Change-Id: I4af824dd2179d8d725b1f44517887e2b7fc7360b

commit ddcac992d6d028717ac943ecacd7fc4ebb36008e
Author: Dan Callaghan <dcallagh@redhat.com>
Date:   Tue Mar 25 14:03:18 2014 +1000

    also run Common and LabController unit tests during build
    
    Change-Id: Id21f0f36e736be934382b96a69b4209e17b29cf0

commit 2ad10547a733428713a0f617c5e02fd77b150391
Author: Dan Callaghan <dcallagh@redhat.com>
Date:   Tue Mar 25 14:03:52 2014 +1000

    tests: greenlet failures were not being reported in test_async
    
    If a test function being run inside a greenlet failed, it was printed to
    stderr but was never raised in the main context so the failure was
    ignored.
    
    Also ensured that all MonitoredSubprocess instances in the tests have
    a timeout set. This is a workaround to ensure that there is always at
    least one other greenlet alive when we block on the test greenlet.
    Gevent is not smart enough to know that we have a SIGCHLD handler which
    will spawn new greenlets at a future time, therefore when it sees only
    one live greenlet waiting on an event it thinks we have deadlocked. The
    proper fix for this would be to use the built-in child watching in
    Gevent 1.0 final.
    
    Change-Id: I938aa7303ee7764cd60b91f4893871c8bed6a1d5

commit e77da55a782a5c1d01135177ebeb23f13a577640
Author: Dan Callaghan <dcallagh@redhat.com>
Date:   Tue Mar 25 14:48:43 2014 +1000

    tests: reduce subprocess count to 600 in test_reaper_race
    
    800 seems to be still too high to reliably pass when run in mock or in
    a user session with the default ulimit of 1024, although it seems to me
    the numbers don't really add up...
    
    Change-Id: I9f411b338156ef6a75ecfd117ada90e5215eb2e3

commit 51a633c35c193cba05d4ebbc7e14f10b6accc0c5
Author: Amit Saha <asaha@redhat.com>
Date:   Thu Mar 27 12:48:51 2014 +1000

    beaker-wizard: Add beakerlib dictionary to existing dictionaries
    
    Bug: 1077555

commit c24b88f630c42307ee4e618ee46169eae668f85f
Author: Amit Saha <asaha@redhat.com>
Date:   Tue Mar 18 23:44:15 2014 +1000

    Allow disabling IPv6 in Beah
    
    This patch implements the Beaker side of the support by adding a new
    ``ks_meta`` option.
    
    Bug: 1065811
    Change-Id: I511148ba46caadf4dff2108bba61413406c29440

commit 8358fead9f118909ccc7779c3083a981b482d729
Author: Dan Callaghan <dcallagh@redhat.com>
Date:   Tue Mar 25 15:41:41 2014 +1000

    tests: need to manually expire osmajor.install_options_by_arch
    
    On newer SQLAlchemy, which is more conservative about only expiring
    things from the session when necessary, this test was failing. That's
    because the OSMajorInstallOptions.lazy_create call completely bypasses
    the ORM session, so SQLAlchemy was not refreshing the
    osmajor.install_options_by_arch associationproxy.
    
    The simplest fix is to manually mark it as expired after the lazy_create
    call (since really, we are invalidating it behind SQLAlchemy's back).
    
    Change-Id: If2929854fcdc6be70225f10cf848758b77a3fd76