summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
...
* | | | | | | Merge "Imported Translations from Transifex"OpenStack Jenkins2012-10-0121-3057/+3120
|\ \ \ \ \ \ \ | |_|_|_|_|_|/ |/| | | | | |
| * | | | | | Imported Translations from TransifexOpenStack Jenkins2012-10-0121-3057/+3120
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Change-Id: I92d4367972c8080d151b841c09098c835c44ea05
* | | | | | | Merge "Use self.flags() instead of manipulating FLAGS by hand"Jenkins2012-10-014-34/+18
|\ \ \ \ \ \ \
| * | | | | | | Use self.flags() instead of manipulating FLAGS by handJohannes Erdfelt2012-09-284-34/+18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Since these test cases subclass test.TestCase, they can use self.flags() to temporarily mutate FLAGS and have it cleaned automatically in the tearDown() method provided by test.TestCase Change-Id: I70ec3f8c947172816415190908d8a650d87d7b92
* | | | | | | | Merge "Use test.TestCase provided self.mox and self.stubs"Jenkins2012-10-012-106/+34
|\| | | | | | | | |/ / / / / / |/| | | | | |
| * | | | | | Use test.TestCase provided self.mox and self.stubsJohannes Erdfelt2012-09-282-106/+34
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Since these test cases subclass test.TestCase, it's unnecessary to create a new self._mox when self.mox already exists. Also, self.stubs already exists. Finally, it's unnecessary to call VerifyAll() since test.TestCase does that as well. Change-Id: Ib5a176d9d3448f3eb86d13ca6199d1ea557480d0
* | | | | | | Merge "xenapi: implement resume_state_on_host_boot"Jenkins2012-09-301-0/+5
|\ \ \ \ \ \ \
| * | | | | | | xenapi: implement resume_state_on_host_bootJohannes Erdfelt2012-09-281-0/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Implement the resume_state_on_host_boot() driver method for XenAPI. This makes it possible to automatically start instances when nova-compute starts up with the XenAPI driver. Change-Id: I68eb11fae57825898009aa1d9567d8112aad53b3
* | | | | | | | Merge "Imported Translations from Transifex"OpenStack Jenkins2012-09-3021-11173/+11989
|\ \ \ \ \ \ \ \
| * | | | | | | | Imported Translations from TransifexOpenStack Jenkins2012-09-3021-11173/+11989
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Change-Id: I73e460aa78cb6565f47caadff27eeadb0d0c0249
* | | | | | | | | Merge "Remove unnecessary setUp, tearDown and __init__ in tests"Jenkins2012-09-3016-74/+0
|\ \ \ \ \ \ \ \ \
| * | | | | | | | | Remove unnecessary setUp, tearDown and __init__ in testsJohannes Erdfelt2012-09-2816-74/+0
| | |/ / / / / / / | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | These methods simply call the same parent class method with the same arguments. This is the default behavior of Python and re-implementing that is unnecessary. Change-Id: I29503f6c5442e2aae645899dc157466e6201ba75
* | | | | | | | | Merge "libvirt: continue detach if instance not found"Jenkins2012-09-301-7/+14
|\ \ \ \ \ \ \ \ \
| * | | | | | | | | libvirt: continue detach if instance not foundVishvananda Ishaya2012-09-271-7/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently the libvirt code will always logout of the iscsi target during detach error, but the exception will cause the volume to remain in an attached state. This needs to be more consistent, so This patch checks specifically for not found and logs a warning and explicitly detaches. Any other exception will leave iscsi logged in and leave the volume in the attached state. Fixes bug 1057756 Change-Id: I23bb7f27e8a9b20fdbf62b0629ca8cc448f427c5
* | | | | | | | | | Merge "libvirt: allows attach and detach from all domains"Jenkins2012-09-302-16/+25
|\| | | | | | | | |
| * | | | | | | | | libvirt: allows attach and detach from all domainsVishvananda Ishaya2012-09-272-16/+25
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Previously attaching/detaching from a shutoff domain would fail. Fixes bug 1057730 Change-Id: I876872700da125cb078746401b1b80da265880ff
* | | | | | | | | | Merge "Fixes error message for flavor-create duplicate ID"Jenkins2012-09-304-4/+9
|\ \ \ \ \ \ \ \ \ \
| * | | | | | | | | | Fixes error message for flavor-create duplicate IDAndy McCrae2012-09-274-4/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fixes bug 1056918 Change-Id: I516153211504b67f6e8b8ebb138abcef4be4d73e
* | | | | | | | | | | Merge "Fix XenServer's ability to boot xen type images"Jenkins2012-09-303-1/+60
|\ \ \ \ \ \ \ \ \ \ \
| * | | | | | | | | | | Fix XenServer's ability to boot xen type imagesMate Lakat2012-09-253-1/+60
| | |_|_|_|_|_|_|/ / / | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fixes bug 1055431. As the scheduler wants to boot a vm_mode=xen type of image, the host's "supported_instances" capability is used for finding a good candidate. In the Xapi case, this field was not populated. This fix modifies the xapi xenhost plugin, so the Xen host capabilities are returned back to the compute node, as "host_capabilities". On the compute side, the mentioned information is used, to extract the "supported_instances" information. Change-Id: I2da11ab81f74b5b52e2c30832a694470978e21b0
* | | | | | | | | | | Merge "all_extensions api_samples testing for server actions"Jenkins2012-09-3033-0/+381
|\ \ \ \ \ \ \ \ \ \ \
| * | | | | | | | | | | all_extensions api_samples testing for server actionsSean Dague2012-09-2533-0/+381
| |/ / / / / / / / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | server actions sometimes return servers objects, which means their output is different when extensions are enabled. This adds api samples for those cases to give a more realistic view to the user of what they might expect. Change-Id: Iaa9a4c3a373dee41931c5dc4f1b805f09a05299f
* | | | | | | | | | | Merge "Add servers/ips api_samples tests"Jenkins2012-09-309-0/+76
|\ \ \ \ \ \ \ \ \ \ \
| * | | | | | | | | | | Add servers/ips api_samples testsDan Smith2012-09-259-0/+76
| | |_|_|_|_|_|_|_|/ / | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Change-Id: I65f8b53b4126458e1522f0bc47559dc1318a473a
* | | | | | | | | | | Merge "Deleting security group does not mark rules as deleted"Jenkins2012-09-301-0/+6
|\ \ \ \ \ \ \ \ \ \ \
| * | | | | | | | | | | Deleting security group does not mark rules as deletedAaron Rosen2012-09-251-0/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Adds query to also delete rules for parent_group_id Fixes bug 1056380 Change-Id: I3d041b49703bf298a80d2ebe19c1298eecf2d98d
* | | | | | | | | | | | Merge "libvirt: Improve the idempotency of iscsi detach"Jenkins2012-09-301-2/+2
|\ \ \ \ \ \ \ \ \ \ \ \
| * | | | | | | | | | | | libvirt: Improve the idempotency of iscsi detachVishvananda Ishaya2012-09-271-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When detaching an iscsi volume it is possible for the iscsi commands to run concurrently, causing a target to be deleted by one greenthread while the other is continuing. When removing the iscsi connection, we should always ignore exit code 21 because that means that the target has already been removed. Fixes bug 1057719 Change-Id: I0c9f2623f85a817e2be506f9a6d523d45c76848a
* | | | | | | | | | | | | Merge "Revert "Add full test environment.""Jenkins2012-09-303-7/+2
|\ \ \ \ \ \ \ \ \ \ \ \ \
| * | | | | | | | | | | | | Revert "Add full test environment."James E. Blair2012-09-283-7/+2
| | |_|_|_|_|_|/ / / / / / | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This reverts commit 30e513309fe25af358b4d9ed47da2c7574fc0f06. The python mysql module was removed from test-requires in favor of test-options, but the work to actually use that was never completed. Revert the change so that mysql testing is actually performed again. Change-Id: Iaf35811cc748272229260f2dab95520995a69b44
* | | | | | | | | | | | | Merge "Max I/O ops per host scheduler filter"Jenkins2012-09-302-0/+60
|\ \ \ \ \ \ \ \ \ \ \ \ \ | |_|_|_|_|_|_|_|/ / / / / |/| | | | | | | | | | | |
| * | | | | | | | | | | | Max I/O ops per host scheduler filterBrian Elliott2012-09-282-0/+60
| |/ / / / / / / / / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Provide ability to filter hosts based on the number of concurrent build, resize, and image snapshot operations in progress. FLAGS.max_io_ops_per_host is used to control the number of concurrent permitted operations. Change-Id: I77feca00a6b6b248de8423cef7fb8da2cf704930
* | | | | | | | | | | | Merge "Add api samples to cloudpipe extension"Jenkins2012-09-2913-0/+131
|\ \ \ \ \ \ \ \ \ \ \ \
| * | | | | | | | | | | | Add api samples to cloudpipe extensionMauro S. M. Rodrigues2012-09-2813-0/+131
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Change-Id: I0c9d5ac191c9c9dc6c74f955857aebf7c1952b94
* | | | | | | | | | | | | Merge "Fix CloudPipe extension XML serialization"Jenkins2012-09-292-10/+15
|\| | | | | | | | | | | |
| * | | | | | | | | | | | Fix CloudPipe extension XML serializationDan Smith2012-09-282-10/+15
| | |_|_|_|_|_|/ / / / / | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | It is completely broken right now and this fixes it to properly traverse a list of cloudpipe instance dictionaries, resulting in XML output that matches the API spec. Fixes bug: 1056242 Change-Id: Ic768afeaa76d776fd55a4f618d14fa41ed4c8a63
* | | | | | | | | | | | Merge "Collect more accurate bandwidth data for XenServer"Jenkins2012-09-2914-76/+330
|\ \ \ \ \ \ \ \ \ \ \ \ | |_|_|_|_|_|_|_|/ / / / |/| | | | | | | | | | |
| * | | | | | | | | | | Collect more accurate bandwidth data for XenServerMonsyne Dragon2012-09-2514-76/+330
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This changes the method used to poll xenserver for bandwidth data. The reccomended way of collecting such data from xenserver (namely the RRD files provided by the hosts) do not seem to be reliable, they will sometimes be correct, often will be signifigantly under (> 10%), and occasionally will show artifacts, such as phantom 4gb bandwidth 'spikes'. This patch changes that to use the much simpler method of simply polling the byte counters on the VIF network devices on the host. (We have old non-nova code that does that on xenserver, and that method is known to work). This should also make it much easier for other hypervisors other than xenserver to implement bandwidth polling, as polling the counters is a rather more universal method. Fixes bug 1055737 Change-Id: I6a280d8bbfcc74914f888b11bc09349a270a5f58
* | | | | | | | | | | | Synchronize docstring with actual implementationJohannes Erdfelt2012-09-281-2/+1
| |_|/ / / / / / / / / |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This method no longer calls eventlet.sleep(0) since it's implicit any time a database connection is checked out (which is a few times each loop in this method). Change-Id: Iff3c2e0fb2f7a11092b4d25c03a20beb3b24c91e
* | | | | | | | | | | Merge "xenapi: make it easier to recover from failed migrations"Jenkins2012-09-281-6/+6
|\ \ \ \ \ \ \ \ \ \ \
| * | | | | | | | | | | xenapi: make it easier to recover from failed migrationsJohannes Erdfelt2012-09-261-6/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If a migration/resize fails, the source VM may or may not have it's name changed in xapi. This makes recovering from a failed migration more difficult than it needs to be. Moving the rename to be done first ensures that it always be renamed. Change-Id: I9296aa67af01d4a832bc63bcb9b71ee20b1221cc
* | | | | | | | | | | | Merge "Added compute node stats to HostState"Jenkins2012-09-284-11/+160
|\ \ \ \ \ \ \ \ \ \ \ \
| * | | | | | | | | | | | Added compute node stats to HostStateBrian Elliott2012-09-274-11/+160
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Added compute node statistics to the HostState class for use by scheduler filters. These are statistics pre-computed on the compute side of the house and are of general interest for filter authors. Change-Id: I6d979b1a9f26b8390c20b4ebbde3c9cb8f000396
* | | | | | | | | | | | | Merge "optimize slightly device lookup with LXC umounts"Jenkins2012-09-281-1/+2
|\ \ \ \ \ \ \ \ \ \ \ \ \
| * | | | | | | | | | | | | optimize slightly device lookup with LXC umountsPádraig Brady2012-09-261-1/+2
| |/ / / / / / / / / / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Only canonicalize the path once, rather than for each entry in /proc/mounts. Change-Id: I6e582b44f171a1c4cc151d615e9dcfae1e590a53 Reported-by: Kevin L. Mitchell
* | | | | | | | | | | | | Fixes csv list required for qemu-img create.Joshua Harlow2012-09-271-1/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Makes the list sent to qemu-img create be formatted correctly. Fixes bug 1057806. Change-Id: I7f0ec3617754f9251d582f4265eb771859cb6e15
* | | | | | | | | | | | | Merge "Pass block_device_info to destroy in revert_resize"Jenkins2012-09-271-1/+5
|\ \ \ \ \ \ \ \ \ \ \ \ \ | |_|_|_|/ / / / / / / / / |/| | | | | | | | | | | |
| * | | | | | | | | | | | Pass block_device_info to destroy in revert_resizeMatt Odden2012-09-271-1/+5
| | |_|_|_|_|_|_|/ / / / | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fixes bug #1056285 Passing block_device_info allows the driver to clean up block device mappings (detaching) properly on the newly resized instance Change-Id: I2527f883883c741dbf4d20be0ca33dbef6f9f36a
* | | | | | | | | | | | Merge "Enable list with no dict objects to be sorted in api samples"Jenkins2012-09-271-4/+7
|\ \ \ \ \ \ \ \ \ \ \ \
| * | | | | | | | | | | | Enable list with no dict objects to be sorted in api samplesMauro S. M. Rodrigues2012-09-271-4/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | So far we had only dicts to be sorted. Although Aggregates extension has a response case which a list of strings is returned and the current key attribute didn't checked if the instance was a dict or another object which caused a break in the use case above. This fix this, and use None as key to compare if the element is not a dictionary instance. Change-Id: Id5d31d1d2ad88fb538d8a9637a1969b179dd93ab