[ARVADOS] updated: 8ea68a13a5cba8e6967cf32e1ae7b02e16dacf12

[git at public.curoverse.com]

Summary of changes:
 apps/workbench/test/test_helper.rb  | 135 +++++++++++++++---------------------
 sdk/go/arvadostest/run_servers.go   |   4 +-
 sdk/python/tests/run_test_server.py | 104 +++++++++++++--------------
 3 files changed, 110 insertions(+), 133 deletions(-)

  discards  3ffc4fd391454639fe3b5f08e5beaadd159863bf (commit)
  discards  c3086805dca3f5f7fef4fbc907aeed0b0ea53f09 (commit)
  discards  68f0e2bea864bede8c653ecdde6c7fdec68d3670 (commit)
       via  8ea68a13a5cba8e6967cf32e1ae7b02e16dacf12 (commit)
       via  a1ad1f1d7ce5edf62945340c34ba5e22a36f1157 (commit)
       via  13181107ecaeaa92e5d96b05270e56b2d807af39 (commit)

This update added new revisions after undoing existing revisions.  That is
to say, the old revision is not a strict subset of the new revision.  This
situation occurs when you --force push a change and generate a repository
containing something like this:

 * -- * -- B -- O -- O -- O (3ffc4fd391454639fe3b5f08e5beaadd159863bf)
            \
             N -- N -- N (8ea68a13a5cba8e6967cf32e1ae7b02e16dacf12)

When this happens we assume that you've already had alert emails for all
of the O revisions, and so we here report only the revisions in the N
branch from the common base, B.

Those revisions listed above that are new to this repository have
not appeared on any other notification email; so we list those
revisions in full, below.


commit 8ea68a13a5cba8e6967cf32e1ae7b02e16dacf12
Author: Tom Clegg <tom at curoverse.com>
Date:   Mon Feb 2 03:04:03 2015 -0500

    3021: Use run_test_server.py instead of the special workbench way of starting an API server.

commit a1ad1f1d7ce5edf62945340c34ba5e22a36f1157
Author: Tom Clegg <tom at curoverse.com>
Date:   Sun Feb 1 20:13:48 2015 -0500

    3021: Round up various implementations of "run test servers" into an arvadostest module. Update to work with new run_test_server.py.
    
    * Get the API host, port, and admin token from the run_test_server.py
      --auth output, instead of guessing localhost:3000 etc.
    
    * Fix race condition in proxy TestPutAskGetForbidden: set up the
      client first, then switch back to a real token for the server to
      use, instead of hoping the server grabs its token before sleep(0.1)
      returns and we sabotage the environment with a bogus one.

commit 13181107ecaeaa92e5d96b05270e56b2d807af39
Author: Tom Clegg <tom at curoverse.com>
Date:   Sun Feb 1 19:48:09 2015 -0500

    3021: Refactor run_test_server.
    
    * Always start API server with both https:// and wss:// enabled. Use
      mocks to test the websocket-not-available case.
    
    * Leave the API server running between tests; just hit reset() when
      needed. Wait until exiting the process to stop the server.
    
    * Do not use rake (tmp:cache:clear, db:test:load, db:fixtures:load) to
      reset the API server's database before starting up a new
      server. Instead, call /database/reset after starting it. (If that
      doesn't work, crash instead of proceeding with undefined results.)
    
    * If a parent process has already started an API server and put its
      endpoint in ARVADOS_TEST_API_HOST, use that instead of starting a
      new one, and leave it running at exit.
    
    * Start servers on random port numbers, so [some] test suites can run
      concurrently with others on a single host (assuming they're
      configured to use different databases).

-----------------------------------------------------------------------


hooks/post-receive
--