If two lists had differed because one had an extra element, both lists
would be included in the exception message. If the lists were large (for
example the list of extensions) it could be very difficult trying to
figure out what is different between the two lists.
This change will show the items that only exist in the expected and/or
result and not the other list.
Change-Id: I6102faea037b4f5e31361adc9db81e79592b7008
Use the default bit length of the underlying ssh-keygen command
(currently 2048) if no bit length is supplied, rather than defaulting to
1024 bits.
bug 1103130
Change-Id: Iba9d378d5bf9e28663e52180ed04c31c16d08aad
Signed-off-by: Zane Bitter <zbitter@redhat.com>
When volumes were removed from nova, api-paste.ini was not updated.
This removes the outdated sections of api-paste.ini.
Change-Id: I6dcc792ed87f6b02e5fdffb0af46a3fbf6b37404
Modified: doc/api_samples/os-flavor-manage/flavor-create-post-req.json
Deleted final comma for proper syntax
Partially fixes: bug 1088528 (part of the bug requires changes in other repo - done)
Change-Id: I5b0b29b4c720a95d5c4c5d957e77dc2e4c9cf8ee
Currently each time the trusted filter is called to check host_pass, it
polls Open Attestation Service to get the trusted level for the host. This
solution is not good on scalability.
With a cache for the host trust level, trusted filter don't need to
consult OAT service if the cache is still valid, thus improves the
scalability. The cache is setup when the first time the filter consulting
the OAT service.
The trusted_filters are reinitialized for each VM that is
scheduled, thus the cache is valid only for one VM instance scheduing.
Implements blueprint trusted-filter-cache
Change-Id: I613da53876e6cd548566e27ab1693287023e9861
Signed-off-by: Yunhong, Jiang <yunhong.jiang@intel.com>
Moved global variables to the top.
Removed once used variables
Used option type definitions in OPT_TYPES
Change-Id: Ia0bf39933bac84dbf432e6b4590ceb2527e6889c
Using the setup.py testr command in run_tests.sh will sync run_tests.sh
up with the way tox currently runs the tests and collects coverage.
Change-Id: I21e3965280b3242a860c6be49c4dca44f197b03a
Nova coverage was broken because additional coverage tests were added
that interfere with collecting coverage during unit testing. These new
tests were not properly excluded by the coverage unit test regex. Modify
the regex to exclude any tests of the form '.*test.*coverage.*'. This
allows `tox -ecover` and `run_tests.sh --coverage` to run properly.
Change-Id: Id35f50ec335707c9eadb0e7fb3aa0833661f7333
As in nova/availability_zones.py, the method get_host_availability_zone()
expects "context" as a module rather than a RequestContext object.
But for our preference, the argument "context" is better to be a
RequestContext object. So fix this by passing an context object into the
method.
* Fix the usage of argument "context"
* Add unittest for the module "availability_zones"
Fix bug lp:#1100598
Change-Id: I5886ba26131261d97be65f18e9492c40401c5a7f
because of testr, we can't assume the migrations tests run in
sequence, so they make run all at the same time. This is fine for
our in memory sqlite db, but when talking to a real external
database, not so good.
Since a second mysql test was added, we started racing on these
migrations them often hitting at the same time causing random fails
in gate.
Fix this by moving all the mysql tests into 1 test function, so we
won't race.
Fixes bug #1101874
Change-Id: I7627f3bd510266c99af018d355f960b352573759
Get Floating IP Pools is always responding 403 (Forbidden) cause its method
name, which is also being used as action name, doesn't match with the name in
policy.json file.
I opted for this approach cause it also fixes the different names between the
extension (which is named floating **ip** pools instead of floating pools) and
the backend.
Another good practice would be change the default_floating_pool option to
default_float_ip_pool, but that would require change the conf file and possibly
break some compatibility.
Fixes bug 1091668
Change-Id: I2627e8b0f45c5c39bbbf1a4a12b448e1992fa4b4
