openstack/python-dracclient
Code Issues Proposed changes

Merge "Add realtime support to drive conversion"

Browse Source
This commit is contained in:
Zuul 2019-05-10 17:40:28 +00:00 committed by Gerrit Code Review
commit a34705b1bb
3 changed files with 149 additions and 154 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

43
dracclient/client.py
View File

@ -1043,32 +1043,33 @@ class DRACClient(object):
def change_physical_disk_state(self, mode,
controllers_to_physical_disk_ids=None):
"""Convert disks RAID status and return a list of controller IDs
"""Convert disks RAID status
Builds a list of controller ids that have had disks converted to the
specified RAID status by:
- Examining all the disks in the system and filtering out any that are
not attached to a RAID/BOSS controller.
- Inspect the controllers' disks to see if there are any that need to
be converted, if so convert them. If a disk is already in the desired
status the disk is ignored. Also check for failed or unknown disk
statuses and raise an exception where appropriate.
- Return a list of controller IDs for controllers whom have had any of
their disks converted, and whether a reboot is required.
This method intelligently converts the requested physical disks from
RAID to JBOD or vice versa. It does this by only converting the
disks that are not already in the correct state.
The caller typically should then create a config job for the list of
controllers returned to finalize the RAID configuration.
:param mode: constants.RaidStatus enumeration used to determine what
raid status to check for.
:param mode: constants.RaidStatus enumeration that indicates the mode
to change the disks to.
:param controllers_to_physical_disk_ids: Dictionary of controllers and
corresponding disk ids we are inspecting and creating jobs for
when needed.
:returns: a dict containing the following key/values:
corresponding disk ids to convert to the requested mode.
:returns: a dictionary containing:
- conversion_results, a dictionary that maps controller ids
to the conversion results for that controller. The
conversion results are a dict that contains:
- The is_commit_required key with the value always set to
True indicating that a config job must be created to
complete disk conversion.
- The is_reboot_required key with a RebootRequired
enumerated value indicating whether the server must be
rebooted to complete disk conversion.
Also contained in the main dict are the following key/values,
which are deprecated, should not be used, and will be removed
in a future release:
- is_reboot_required, a boolean stating whether a reboot is
required or not.
required or not.
- commit_required_ids, a list of controller ids that will
need to commit their pending RAID changes via a config job.
need to commit their pending RAID changes via a config job.
:raises: DRACOperationFailed on error reported back by the DRAC and the
exception message does not contain NOT_SUPPORTED_MSG constant.
:raises: Exception on unknown error.

106
dracclient/resources/raid.py
View File

@ -12,6 +12,7 @@
# under the License.
import collections
import copy
import logging
from dracclient import constants
@ -557,15 +558,17 @@ class RAIDManagement(object):
:param mode: constants.RaidStatus enumeration used to
determine what raid status to check for.
:param physical_disks: all physical disks
:param controllers_to_physical_disk_ids: Dictionary of controllers
we are inspecting and creating jobs for when needed. If
needed modify this dict so that only drives that need to
be changed to RAID or JBOD are in the list of disk keys
for corresponding controller.
:param controllers_to_physical_disk_ids: Dictionary of controllers and
corresponding disk ids to convert to the requested mode.
:returns: a dictionary mapping controller FQDDs to the list of
physical disks that need to be converted for that controller.
:raises: ValueError: Exception message will list failed drives and
drives whose state cannot be changed at this time, drive
state is not "ready" or "non-RAID".
"""
controllers_to_physical_disk_ids = copy.deepcopy(
controllers_to_physical_disk_ids)
p_disk_id_to_status = {}
for physical_disk in physical_disks:
p_disk_id_to_status[physical_disk.id] = physical_disk.raid_status
@ -624,34 +627,37 @@ class RAIDManagement(object):
raise ValueError(error_msg)
return controllers_to_physical_disk_ids
def change_physical_disk_state(self, mode,
controllers_to_physical_disk_ids=None):
"""Convert disks RAID status and return a list of controller IDs
"""Convert disks RAID status
Builds a list of controller ids that have had disks converted to the
specified RAID status by:
- Examining all the disks in the system and filtering out any that are
not attached to a RAID/BOSS controller.
- Inspect the controllers' disks to see if there are any that need to
be converted, if so convert them. If a disk is already in the desired
status the disk is ignored. Also check for failed or unknown disk
statuses and raise an exception where appropriate.
- Return a list of controller IDs for controllers whom have had any of
their disks converted, and whether a reboot is required.
This method intelligently converts the requested physical disks from
RAID to JBOD or vice versa. It does this by only converting the
disks that are not already in the correct state.
The caller typically should then create a config job for the list of
controllers returned to finalize the RAID configuration.
:param mode: constants.RaidStatus enumeration used to determine what
raid status to check for.
:param mode: constants.RaidStatus enumeration that indicates the mode
to change the disks to.
:param controllers_to_physical_disk_ids: Dictionary of controllers and
corresponding disk ids we are inspecting and creating jobs for
when needed.
:returns: a dict containing the following key/values:
corresponding disk ids to convert to the requested mode.
:returns: a dictionary containing:
- conversion_results, a dictionary that maps controller ids
to the conversion results for that controller. The
conversion results are a dict that contains:
- The is_commit_required key with the value always set to
True indicating that a config job must be created to
complete disk conversion.
- The is_reboot_required key with a RebootRequired
enumerated value indicating whether the server must be
rebooted to complete disk conversion.
Also contained in the main dict are the following key/values,
which are deprecated, should not be used, and will be removed
in a future release:
- is_reboot_required, a boolean stating whether a reboot is
required or not.
required or not.
- commit_required_ids, a list of controller ids that will
need to commit their pending RAID changes via a config job.
need to commit their pending RAID changes via a config job.
:raises: DRACOperationFailed on error reported back by the DRAC and the
exception message does not contain NOT_SUPPORTED_MSG constant.
:raises: Exception on unknown error.
@ -678,13 +684,14 @@ class RAIDManagement(object):
Raise exception if there are any failed drives or
drives not in status 'ready' or 'non-RAID'
'''
self._check_disks_status(mode, physical_disks,
controllers_to_physical_disk_ids)
final_ctls_to_phys_disk_ids = self._check_disks_status(
mode, physical_disks, controllers_to_physical_disk_ids)
is_reboot_required = False
controllers = []
controllers_to_results = {}
for controller, physical_disk_ids \
in controllers_to_physical_disk_ids.items():
in final_ctls_to_phys_disk_ids.items():
if physical_disk_ids:
LOG.debug("Converting the following disks to {} on RAID "
"controller {}: {}".format(
@ -697,22 +704,39 @@ class RAIDManagement(object):
if constants.NOT_SUPPORTED_MSG in str(ex):
LOG.debug("Controller {} does not support "
"JBOD mode".format(controller))
pass
controllers_to_results[controller] = \
utils.build_return_dict(
doc=None,
resource_uri=None,
is_commit_required_value=False,
is_reboot_required_value=constants.
RebootRequired.false)
else:
raise
else:
if conversion_results:
reboot_true = constants.RebootRequired.true
reboot_optional = constants.RebootRequired.optional
_is_reboot_required = \
conversion_results["is_reboot_required"]
is_reboot_required = is_reboot_required \
or (_is_reboot_required
in [reboot_true, reboot_optional])
if conversion_results["is_commit_required"]:
controllers.append(controller)
controllers_to_results[controller] = conversion_results
return {'is_reboot_required': is_reboot_required,
# Remove the code below when is_reboot_required and
# commit_required_ids are deprecated
reboot_true = constants.RebootRequired.true
reboot_optional = constants.RebootRequired.optional
_is_reboot_required = \
conversion_results["is_reboot_required"]
is_reboot_required = is_reboot_required \
or (_is_reboot_required
in [reboot_true, reboot_optional])
controllers.append(controller)
else:
controllers_to_results[controller] = \
utils.build_return_dict(
doc=None,
resource_uri=None,
is_commit_required_value=False,
is_reboot_required_value=constants.
RebootRequired.false)
return {'conversion_results': controllers_to_results,
'is_reboot_required': is_reboot_required,
'commit_required_ids': controllers}
def is_realtime_supported(self, raid_controller_fqdd):

154
dracclient/tests/test_raid.py
View File

@ -991,13 +991,13 @@ class ClientRAIDManagementTestCase(base.BaseTest):
raid_mgt._check_disks_status,
mode,
physical_disks,
self.controllers_to_physical_disk_ids.copy())
self.controllers_to_physical_disk_ids)
mode = constants.RaidStatus.jbod
self.assertRaises(ValueError,
raid_mgt._check_disks_status,
mode,
physical_disks,
self.controllers_to_physical_disk_ids.copy())
self.controllers_to_physical_disk_ids)
def test_check_disks_status_fail(self, mock_requests):
mode = constants.RaidStatus.raid
@ -1009,13 +1009,13 @@ class ClientRAIDManagementTestCase(base.BaseTest):
raid_mgt._check_disks_status,
mode,
physical_disks,
self.controllers_to_physical_disk_ids.copy())
self.controllers_to_physical_disk_ids)
mode = constants.RaidStatus.jbod
self.assertRaises(ValueError,
raid_mgt._check_disks_status,
mode,
physical_disks,
self.controllers_to_physical_disk_ids.copy())
self.controllers_to_physical_disk_ids)
def test_check_disks_status_no_change(self, mock_requests):
raid_mgt = self.drac_client._raid_mgmt
@ -1023,11 +1023,8 @@ class ClientRAIDManagementTestCase(base.BaseTest):
physical_disks = [self.disk_1, self.disk_2,
self.disk_3, self.disk_4]
raid_cntl_to_phys_disk_ids = (self.controllers_to_physical_disk_ids.
copy())
raid_mgt._check_disks_status(mode, physical_disks,
raid_cntl_to_phys_disk_ids)
raid_cntl_to_phys_disk_ids = raid_mgt._check_disks_status(
mode, physical_disks, self.controllers_to_physical_disk_ids)
raid_len = len(raid_cntl_to_phys_disk_ids['RAID.Integrated.1-1'])
self.assertEqual(raid_len, 0)
@ -1037,10 +1034,8 @@ class ClientRAIDManagementTestCase(base.BaseTest):
physical_disks = [disk_1_non_raid, disk_2_non_raid,
self.disk_3, self.disk_4]
jbod_cntl_to_phys_disk_ids = (self.controllers_to_physical_disk_ids.
copy())
raid_mgt._check_disks_status(mode, physical_disks,
jbod_cntl_to_phys_disk_ids)
jbod_cntl_to_phys_disk_ids = raid_mgt._check_disks_status(
mode, physical_disks, self.controllers_to_physical_disk_ids)
jbod_len = len(jbod_cntl_to_phys_disk_ids['RAID.Integrated.1-1'])
self.assertEqual(jbod_len, 0)
@ -1049,11 +1044,8 @@ class ClientRAIDManagementTestCase(base.BaseTest):
mode = constants.RaidStatus.jbod
physical_disks = [self.disk_1, self.disk_2, self.disk_3, self.disk_4]
jbod_cntl_to_phys_disk_ids = (self.controllers_to_physical_disk_ids.
copy())
raid_mgt._check_disks_status(mode, physical_disks,
jbod_cntl_to_phys_disk_ids)
jbod_cntl_to_phys_disk_ids = raid_mgt._check_disks_status(
mode, physical_disks, self.controllers_to_physical_disk_ids)
jbod_len = len(jbod_cntl_to_phys_disk_ids['RAID.Integrated.1-1'])
self.assertEqual(jbod_len, 2)
@ -1062,10 +1054,8 @@ class ClientRAIDManagementTestCase(base.BaseTest):
disk_2_non_raid = self.disk_2._replace(raid_status='non-RAID')
physical_disks = [disk_1_non_raid, disk_2_non_raid,
self.disk_3, self.disk_4]
raid_cntl_to_phys_disk_ids = (self.controllers_to_physical_disk_ids.
copy())
raid_mgt._check_disks_status(mode, physical_disks,
raid_cntl_to_phys_disk_ids)
raid_cntl_to_phys_disk_ids = raid_mgt._check_disks_status(
mode, physical_disks, self.controllers_to_physical_disk_ids)
raid_len = len(raid_cntl_to_phys_disk_ids['RAID.Integrated.1-1'])
self.assertEqual(raid_len, 2)
@ -1080,13 +1070,13 @@ class ClientRAIDManagementTestCase(base.BaseTest):
raid_mgt._check_disks_status,
mode,
physical_disks,
self.controllers_to_physical_disk_ids.copy())
self.controllers_to_physical_disk_ids)
mode = constants.RaidStatus.jbod
self.assertRaises(ValueError,
raid_mgt._check_disks_status,
mode,
physical_disks,
self.controllers_to_physical_disk_ids.copy())
self.controllers_to_physical_disk_ids)
@mock.patch.object(dracclient.client.WSManClient,
'wait_until_idrac_is_ready', spec_set=True,
@ -1102,15 +1092,21 @@ class ClientRAIDManagementTestCase(base.BaseTest):
mock_requests.post(
'https://1.2.3.4:443/wsman',
text=test_utils.RAIDEnumerations[uris.DCIM_PhysicalDiskView]['ok'])
mock_convert_physical_disks.return_value = {'is_commit_required': True,
'is_reboot_required':
constants.RebootRequired
.true}
cntl_to_phys_d_ids = self.controllers_to_physical_disk_ids
cvt_phys_disks_return_value = {'is_commit_required': True,
'is_reboot_required': constants.
RebootRequired.true}
mock_convert_physical_disks.return_value = cvt_phys_disks_return_value
expected_return_value = {'RAID.Integrated.1-1':
cvt_phys_disks_return_value,
'AHCI.Integrated.1-1':
cvt_phys_disks_return_value}
results = self.drac_client.change_physical_disk_state(
mode, cntl_to_phys_d_ids)
mode, self.controllers_to_physical_disk_ids)
self.assertTrue(results["is_reboot_required"])
self.assertEqual(len(results["commit_required_ids"]), 2)
self.assertEqual(results['conversion_results'],
expected_return_value)
@mock.patch.object(dracclient.resources.raid.RAIDManagement,
'list_physical_disks', spec_set=True,
@ -1128,38 +1124,44 @@ class ClientRAIDManagementTestCase(base.BaseTest):
physical_disks = [disk_1_non_raid, disk_2_non_raid,
self.disk_3, self.disk_4]
mock_list_physical_disks.return_value = physical_disks
mock_convert_physical_disks.return_value = {'is_commit_required': True,
'is_reboot_required':
constants.RebootRequired
.true}
cntl_to_phys_d_ids = self.controllers_to_physical_disk_ids
boss_return_value = {'is_commit_required': False,
'is_reboot_required':
constants.RebootRequired.false}
raid_return_value = {'is_commit_required': True,
'is_reboot_required':
constants.RebootRequired.true}
mock_convert_physical_disks.return_value = raid_return_value
results = self.drac_client.change_physical_disk_state(
mode, cntl_to_phys_d_ids)
mode, self.controllers_to_physical_disk_ids)
self.assertTrue(results["is_reboot_required"])
self.assertEqual(len(results["commit_required_ids"]), 1)
self.assertEqual(len(results['conversion_results']), 2)
self.assertEqual(results['conversion_results']['AHCI.Integrated.1-1'],
boss_return_value)
self.assertEqual(results['conversion_results']['RAID.Integrated.1-1'],
raid_return_value)
@mock.patch.object(dracclient.resources.raid.RAIDManagement,
'list_physical_disks', spec_set=True,
autospec=True)
@mock.patch.object(dracclient.resources.raid.RAIDManagement,
'convert_physical_disks', spec_set=True,
autospec=True)
def test_change_physical_disk_state_none(
self, mock_requests,
mock_convert_physical_disks,
mock_list_physical_disks):
mode = constants.RaidStatus.raid
physical_disks = [self.disk_1, self.disk_2, self.disk_3, self.disk_4]
mock_convert_physical_disks.return_value = {'is_commit_required': True,
'is_reboot_required':
constants.RebootRequired
.true}
mock_list_physical_disks.return_value = physical_disks
cntl_to_phys_d_ids = self.controllers_to_physical_disk_ids
expected_return_value = {'is_commit_required': False,
'is_reboot_required':
constants.RebootRequired.false}
results = self.drac_client.change_physical_disk_state(
mode, cntl_to_phys_d_ids)
mode, self.controllers_to_physical_disk_ids)
self.assertFalse(results["is_reboot_required"])
self.assertEqual(len(results["commit_required_ids"]), 0)
self.assertEqual(results['conversion_results']['RAID.Integrated.1-1'],
expected_return_value)
self.assertEqual(results['conversion_results']['AHCI.Integrated.1-1'],
expected_return_value)
@mock.patch.object(dracclient.resources.raid.RAIDManagement,
'list_physical_disks', spec_set=True,
@ -1179,11 +1181,17 @@ class ClientRAIDManagementTestCase(base.BaseTest):
physical_disks = [disk_1_non_raid, disk_2_non_raid,
self.disk_3, self.disk_4]
mock_list_physical_disks.return_value = physical_disks
cntl_to_phys_d_ids = self.controllers_to_physical_disk_ids
expected_return_value = {'is_commit_required': False,
'is_reboot_required':
constants.RebootRequired.false}
results = self.drac_client.change_physical_disk_state(
mode, cntl_to_phys_d_ids)
mode, self.controllers_to_physical_disk_ids)
self.assertFalse(results["is_reboot_required"])
self.assertEqual(len(results["commit_required_ids"]), 0)
self.assertEqual(results['conversion_results']['RAID.Integrated.1-1'],
expected_return_value)
self.assertEqual(results['conversion_results']['AHCI.Integrated.1-1'],
expected_return_value)
@mock.patch.object(dracclient.resources.raid.RAIDManagement,
'list_physical_disks', spec_set=True,
@ -1203,12 +1211,12 @@ class ClientRAIDManagementTestCase(base.BaseTest):
physical_disks = [disk_1_non_raid, disk_2_non_raid,
self.disk_3, self.disk_4]
mock_list_physical_disks.return_value = physical_disks
cntl_to_phys_d_ids = self.controllers_to_physical_disk_ids
self.assertRaisesRegexp(
exceptions.DRACOperationFailed,
"OTHER_MESSAGE",
self.drac_client.change_physical_disk_state,
mode, cntl_to_phys_d_ids)
mode,
self.controllers_to_physical_disk_ids)
@mock.patch.object(dracclient.resources.raid.RAIDManagement,
'list_physical_disks', spec_set=True,
@ -1227,11 +1235,12 @@ class ClientRAIDManagementTestCase(base.BaseTest):
physical_disks = [disk_1_non_raid, disk_2_non_raid,
self.disk_3, self.disk_4]
mock_list_physical_disks.return_value = physical_disks
cntl_to_phys_d_ids = self.controllers_to_physical_disk_ids
self.assertRaisesRegexp(
Exception, "SOMETHING_BAD_HAPPENED",
Exception,
"SOMETHING_BAD_HAPPENED",
self.drac_client.change_physical_disk_state,
mode, cntl_to_phys_d_ids)
mode,
self.controllers_to_physical_disk_ids)
@mock.patch.object(dracclient.client.WSManClient,
'wait_until_idrac_is_ready', spec_set=True,
@ -1285,45 +1294,6 @@ class ClientRAIDManagementTestCase(base.BaseTest):
self.assertFalse(results["is_reboot_required"])
self.assertEqual(len(results["commit_required_ids"]), 0)
@mock.patch.object(dracclient.client.WSManClient,
'wait_until_idrac_is_ready', spec_set=True,
autospec=True)
@mock.patch.object(dracclient.resources.raid.RAIDManagement,
'list_physical_disks', spec_set=True,
autospec=True)
@mock.patch.object(dracclient.resources.raid.RAIDManagement,
'convert_physical_disks', spec_set=True,
autospec=True)
def test_change_physical_disk_state_conversion_return_values(
self, mock_requests,
mock_convert_physical_disks,
mock_list_physical_disks,
mock_wait_until_idrac_is_ready):
mock_requests.post(
'https://1.2.3.4:443/wsman',
text=test_utils.RAIDEnumerations[uris.DCIM_ControllerView]['ok'])
mode = constants.RaidStatus.jbod
physical_disks = [self.disk_1, self.disk_2, self.disk_3, self.disk_4]
'''Test all logic branches for 100% coverage, it is unlikely
convert_physical_disks() will return empty dict but we do check
for this case in change_physical_disk_state()'''
mock_convert_physical_disks.return_value = {}
mock_list_physical_disks.return_value = physical_disks
results = self.drac_client.change_physical_disk_state(mode)
self.assertFalse(results["is_reboot_required"])
self.assertEqual(len(results["commit_required_ids"]), 0)
'''Where convert_physical_disks() does not require a commit after
executing, unlikely case but provides 100% code coverage of all
logic branches.'''
mock_convert_physical_disks.return_value = {'is_commit_required':
False,
'is_reboot_required':
constants.RebootRequired
.false}
results = self.drac_client.change_physical_disk_state(mode)
self.assertFalse(results["is_reboot_required"])
self.assertEqual(len(results["commit_required_ids"]), 0)
@mock.patch.object(dracclient.client.WSManClient,
'wait_until_idrac_is_ready', spec_set=True,
autospec=True)