summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* Merge "Check that all po/pot files are valid"2014.1.rc1Jenkins2014-04-011-0/+4
|\
| * Check that all po/pot files are validAndreas Jaeger2014-03-311-0/+4
| | | | | | | | | | | | | | | | | | Perform minimal checks on po and pot files so that no broken files can get imported. Change-Id: I328fc11c688c16737ae88984e4778483ce5db5ab Co-Authored-By: Clark Boylan <clark.boylan@gmail.com> Related-Bug: #1298645
* | Manually update all translated stringsDevananda van der Veen2014-03-3161-8934/+10404
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Git has been updating the localization (.po) files, and as a result, the 'tx pull -a' command, run by jenkins in the periodic translation jobs, has been failing to pull new strings from transifex. This patch is a manual pull from transifex to update all the strings for inclusion in Icehouse RC1. A longer-term fix has been proposed to -infra here: https://review.openstack.org/#/c/84191/ Change-Id: I9fba04d2369b36fe4902e388774506cd8e6ea377 Closes-bug: #1298645
* | Merge "ironic_host_manager was missing two stats"Jenkins2014-03-311-0/+5
|\ \
| * | ironic_host_manager was missing two statsDevananda van der Veen2014-03-281-0/+5
| | | | | | | | | | | | | | | | | | | | | ironic_host_manager.IronicNodeState was not setting either self.total_usable_disk_gb or self.updated. Change-Id: I0d055dba584cf16043f80e96a16c30e42aaf6bb0
* | | Merge "If no swap is specified default to 1MB"Jenkins2014-03-312-2/+15
|\ \ \ | |_|/ |/| |
| * | If no swap is specified default to 1MBLucas Alvares Gomes2014-03-312-2/+15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | After a244d1bc7b the Nova Ironic Driver will add the pxe_swap_mb parameter using the value in the flavor (which defaults to 0 if not specified). In the Ironic code we had something like get('pxe_swap_mb', 1) to prevent the swap size for being 0, but that get() won't work anymore because the pxe_swap_mb is set. So this patch is taking another approach to force the the pxe_swap_mb parameter to be > 0. Related-Bug: #1297937 Change-Id: Ie124f58681fc184950918cde1bc8c565c1387fc8
* | | Merge "Adds exact match filters to nova scheduler"Jenkins2014-03-314-0/+156
|\ \ \ | |/ / |/| |
| * | Adds exact match filters to nova schedulerRohan Kanade2014-03-284-0/+156
| |/ | | | | | | | | | | | | | | | | | | | | | | Add exact match ram, disk, core filters which will match exact values of requested metrics, this is required for Baremetal/Ironic instance filtering. Unit tests have been proposed to nova at https://review.openstack.org/#/c/83728 Change-Id: I20faa1c17f4121429bbbea80e59b168095909b48 Closes-Bug: #1291396
* | Merge "Remove obsolete po entries - they break translation jobs"Jenkins2014-03-2965-15234/+0
|\ \
| * | Remove obsolete po entries - they break translation jobsDevananda van der Veen2014-03-2865-15234/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | The ironic-propose-translation-update job has been failing to update from transifex because our .po files contain duplicates of obsolete entries which were causing errors. Change-Id: I41a056cbe10096c07e7682d8d79ff8f8c58ad3e2 Related-bug: #1298645
* | | Add note to ssh about impact on ci testingChris Krelle2014-03-281-0/+4
|/ / | | | | | | | | | | | | This patch adds a note to the virsh commands section about how changes will impact CI testing. Change-Id: I2e7ac23a69a19ef718458ec9c4974b6e79e6a4bf
* | Merge "Remove some obsolete settings from DevStack doc"Jenkins2014-03-281-5/+0
|\ \
| * | Remove some obsolete settings from DevStack docAdam Gandelman2014-03-271-5/+0
| | | | | | | | | | | | | | | | | | | | | | | | The BM_* variables are no longer used by Devstack for Ironic deployment and the new defaults should work fine. Removes references to the old settings. Change-Id: Ia19dccaf389273c5a224b804eb270c08d31a1811
* | | Merge "Imported Translations from Transifex"OpenStack Jenkins2014-03-2866-8449/+11209
|\ \ \ | |_|/ |/| |
| * | Imported Translations from TransifexOpenStack Jenkins2014-03-2866-8449/+11209
| | | | | | | | | | | | Change-Id: Ieacb32654feb0350b1c417fb362a8445ef3bacc9
* | | Merge "Fix seamicro validate() method definition"Jenkins2014-03-282-3/+3
|\ \ \ | |/ / |/| |
| * | Fix seamicro validate() method definitionDevananda van der Veen2014-03-272-3/+3
| |/ | | | | | | | | | | | | | | | | | | | | | | | | | | | | The base class definition of vendor_passthru.validate() was changed in 48a4223f so that it also required a 'task' parameter, but the seamicro.vendor_passthru driver had not landed yet. Between that change and the time the seamicro driver was landed, another change caused us to stop running unit tests on third-party drivers. So, we didn't notice that the seamicro.vendor_passthru.validate() method definition was incompatibly broken. Closes-bug: #1298648 Change-Id: I9eab4c1d1be0546a0cdbf40759e88bb22f5015d2
* | Merge "Process public API list as regular expressions"Jenkins2014-03-284-9/+38
|\ \
| * | Process public API list as regular expressionsRoman Prykhodchenko2014-03-144-9/+38
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In order to allow flexible patterns for public API specification and to speed up the process of checking whether some request is performed against the public API, it's reasonable to interpret the list of public endpoints as regular expressions. This patch changes the logic of checking whether a request requires authorisation from searching the endpoint in the set to checking the endpoint against a list of regular expressions. Closes-bug: #1251880 Change-Id: I638ca0e20fa7e44fbeeae0d1e4c2f4188fb597a5
* | | Merge "Replace fixtures with mock in test_keystone.py"Jenkins2014-03-281-60/+18
|\ \ \
| * | | Replace fixtures with mock in test_keystone.pyLucas Alvares Gomes2014-03-251-60/+18
| | | | | | | | | | | | | | | | Change-Id: I83b5035c6919ced43cb4b39ae233976a3a06fa76
* | | | Merge "Raise unexpected exceptions during destroy()"Jenkins2014-03-271-1/+3
|\ \ \ \
| * | | | Raise unexpected exceptions during destroy()Adam Gandelman2014-03-271-1/+3
| | |_|/ | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Do not match exception by __name__ When an API request throws an unexpected exception. Avoids nova-compute AttributeError traceback and provides better debug info on the nova side. Change-Id: I2886df814e8641717c57f77626f92917d1ff744c Closes-bug: #1297549
* | | | Merge "Nova Ironic driver to set pxe_swap_mb in Ironic"Jenkins2014-03-271-0/+4
|\ \ \ \ | |/ / / |/| | |
| * | | Nova Ironic driver to set pxe_swap_mb in IronicLucas Alvares Gomes2014-03-261-0/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch fix the Nova Ironic driver to make it look at the swap field in the flavor and map it to Ironic as driver_info/pxe_swap_mb. Change-Id: Ic6091fd6b6c8e1a32a33969ce5cdd20f6c243167 Closes-Bug: #1297937
* | | | Merge "Start using oslosphinx theme for docs"Jenkins2014-03-2713-1291/+5
|\ \ \ \
| * | | | Start using oslosphinx theme for docsSergey Lukjanov2014-03-2713-1291/+5
| |/ / / | | | | | | | | | | | | | | | | | | | | | | | | Use the new oslosphinx version of the OpenStack doc theme instead of maintaining a copy of it in the Ironic repository. Change-Id: I1d949d1e2734b10e38dea0067db9f3cfd7109c11
* | | | Merge "Fix strings post landing for c63e1d9f6"Jenkins2014-03-271-4/+4
|\ \ \ \
| * | | | Fix strings post landing for c63e1d9f6Yuriy Zveryanskyy2014-03-261-4/+4
| |/ / / | | | | | | | | | | | | | | | | | | | | | | | | Based on comments https://review.openstack.org/#/c/77939/9/ironic/common/neutron.py Change-Id: I965eb94d4c76d7024620792bcd926bce20476d1c
* | | | Merge "Do not use __builtin__ in python3"Jenkins2014-03-273-4/+3
|\ \ \ \
| * | | | Do not use __builtin__ in python3yangxurong2014-03-133-4/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | __builtin__ does not exist in Python 3, use six.moves.builtins instead. Change-Id: I79b61b5ab67f491ce88015f799ff7d66c9eeb7ee closes-bug: #1290234
* | | | | Provide a new ComputeManager for IronicRobert Collins2014-03-273-15/+88
| |/ / / |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Ironic is internally clustered and that causes some friction with Nova which assumes that each n-cpu instance is the only one responsible for any given VM. While this is going to be addressed in the long term, at the moment we need to make it possible to run in an HA setup for Ironic. The short term solution proposed is to run 2+ nova-compute's each of which reports the same hostname (e.g. ironic). This works but has some caveats - one of which is that _init_instance will now race between different nova-compute instances starting up at the same time. A custom ComputeManager permits us to address that without prejuidice to future long term solutions. Relatedly, TripleO needs Ironic to permit service startup before keystone is initialised, which the removal of API calls during init_host permits - and as there are no API calls needed for correct behaviour of the Ironic driver, this is straight forward :). See https://etherpad.openstack.org/p/ironic-nova-friction for more discussion. Change-Id: I68d46c4da8715df03c3a88393b55665dc57045a3 Closes-Bug: #1295503
* | | | Merge "Run periodic_task in a with a dynamic timer"Jenkins2014-03-251-3/+4
|\ \ \ \
| * | | | Run periodic_task in a with a dynamic timerJay Faulkner2014-03-251-3/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In ironic.common.service, Ironic was running the periodic task loop every periodic_interval (60s by default) seconds. This means that any periodic task configured with an interval shorter than periodic_interval was only getting run every periodic_interval seconds. I reconfigured periodic tasks to run in a dynamic loop, where now the periodic task loop will run every min(periodic_interval, next_run) seconds. Change-Id: I17a270ecbb5fd53149a89816f4be1c9a0c27b541 Closes-Bug: 1295874
* | | | | Merge "Speed up ironic tests"Jenkins2014-03-253-2/+6
|\ \ \ \ \
| * | | | | Speed up ironic testsJim Rollenhagen2014-03-243-2/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This change speeds up ironic tests by 60% by: * Mocking eventlet.greenthread.sleep in a handful of tests * Changing a time.sleep(2) to time.sleep(0.2) in a test where time.sleep is used intentionally for testing Change-Id: I0cd2ae06549844ec3b76e7c91e1f49665ebc9717
* | | | | | Merge "Remove logging of exceptions from controller's methods"Jenkins2014-03-254-38/+8
|\ \ \ \ \ \
| * | | | | | Remove logging of exceptions from controller's methodsYuriy Zveryanskyy2014-03-254-38/+8
| | |_|_|/ / | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | WSME does exceptions logging in wsmeext.pecan via format_exception() function. Logging exceptions in Ironic controllers code causes duplicate messages in log. Change-Id: Ie5a9856508127557714166d838f0e762e37a611d
* | | | | | Merge "Run ipmi power status less aggressively"Jenkins2014-03-254-61/+75
|\ \ \ \ \ \
| * | | | | | Run ipmi power status less aggressivelyDan Prince2014-03-254-61/+75
| |/ / / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Ironic runs IPMI 'power status' in a loop until the desired state is reached. Running repeated IPMI commands in a tight loop (1 second sleep) seems to occasionally cause 'insufficient resources' errors with some hardware types. In this commit we update the IPMI retry timer to use a DynamicLoopingCall which can wait longer for each retry of the power status command. We increase the wait time exponentially between each call, until the total time would exceed the max retry time on the next iteration. This patch also removes the spurious initial 'power status' call which preceeded the 'power on' or 'power off' call without any gap. Since Ironic has a periodic task to sync the power state, it is safe to assume that the hardware's current power state is already known and does not need to be verified inside the set_power_state method. As part of this change we also increase the default IPMI retry_timeout config to 60 seconds. Co-authored-by: Dan Prince <dprince@redhat.com> Change-Id: Ie0e29239592bb64c7a10ecedae299df2c5053359 Closes-bug: #1291991
* | | | | | Merge "Imported Translations from Transifex"OpenStack Jenkins2014-03-2566-3223/+2914
|\ \ \ \ \ \
| * | | | | | Imported Translations from TransifexOpenStack Jenkins2014-03-2566-3223/+2914
| |/ / / / / | | | | | | | | | | | | | | | | | | Change-Id: If69eb05fc679aa1466bd185d5308338e6673415c
* | | | | | Merge "Update SeaMicro to use MixinVendorInterface"Jenkins2014-03-253-38/+6
|\ \ \ \ \ \
| * | | | | | Update SeaMicro to use MixinVendorInterfaceGhe Rivero2014-03-253-38/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | SeaMicro driver is now using the generic MixinVendorInterface for the VendorPassthru driver interface Change-Id: Iae8288d10fb9454a513a12b0c28add8b7504efd6
* | | | | | | Merge "Avoid API root controller dependency on v1 dir"Jenkins2014-03-2510-17/+17
|\ \ \ \ \ \ \
| * | | | | | | Avoid API root controller dependency on v1 dirRamakrishnan G2014-03-2510-17/+17
| | |/ / / / / | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This commit fixes the issue that pecan root controller is dependent on files in v1 directory. Change-Id: Id26db27b61094356ee02dbc7c5baa7e551a3cc5c Partial-Bug: #1255954
* | | | | | | Merge "Adds max retry limit to sync_power_state task"Jenkins2014-03-253-52/+138
|\ \ \ \ \ \ \
| * | | | | | | Adds max retry limit to sync_power_state taskRohan Kanade2014-03-243-52/+138
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * sync_power_state periodic task tries to set node's actual power state. In case of failure to sync power state, Ironic will keep trying to sync power state, Hence introduce a max retry per node sync power state limit * Set Node's DB power state as the actual Node power state if max retries exceeded * Set the Node's last_error msg after max retries exceed and set the node in maintainence mode Change-Id: Idbd3afab746f81b9097e97afb1d12a81686a4564 Closes-Bug: #1285806
* | | | | | | | Merge "Decrease running time of SeaMicro driver tests"Jenkins2014-03-251-6/+8
|\ \ \ \ \ \ \ \
View Lorry Controller Status