Joe Gordon [Tue, 29 Apr 2014 22:22:20 +0000 (15:22 -0700)]
Don't print duplicate messages on SystemExit
Don't log the error using LOG.error or LOG.exception and then
passing the same string to SystemExit since this will result in the
error being logged twice. Instead log the error and raise SystemExit(1).
In case the network is without a subnet, calling the validation
logic during the report phase leads to an error because the LSN
would have been already allocated during the migration phase.
Bypass the issue by calling the plugin directly, which is what
the validation logic does in the first place.
Henry Gessau [Fri, 25 Apr 2014 19:46:58 +0000 (15:46 -0400)]
Allow ML2 plugin test cases to be run independently
Some test cases in neutron.tests.unit.ml2.test_ml2_plugin override the
network_vlan_ranges option in the ml2_type_vlan config group. However,
the test module does not import the ML2 type driver module where this
config group's options are registered. When unit tests are run by
tox/testr the module is imported during test discovery. Running the
tests independently via nose or testtools requires the option to be
explicitly imported.
Henry Gessau [Thu, 13 Feb 2014 16:58:47 +0000 (11:58 -0500)]
Cisco APIC ML2 mechanism driver, part 1
This set of changes introduces a mechanism driver for the
Cisco APIC. Please see the blueprint for more information.
The review is submitted in two parts:
- Part 1 (this one)
o APIC REST Client
o APIC data model and migration script
o APIC configurations
- Part 2 (dependent on part 1)
o APIC mechanism driver
o APIC manager
Ann Kamyshnikova [Thu, 24 Apr 2014 11:34:45 +0000 (15:34 +0400)]
Fix incorrect usage of sa.String() type
In downgrade of 4eca4a84f08a_remove_ml2_cisco_cred_db migartion
there is a mistake in usage SQLAlchemy String type.
Used sa.string() instead of sa.String()
Jakub Libosvar [Fri, 18 Apr 2014 13:31:55 +0000 (15:31 +0200)]
Fix H302 violations in extensions package
H302 violation is reported by flake8 when importing separated objects from
modules instead of importing the whole module.
e.g. from package.module import function
function()
is changed to
from package import module
module.function()
5b7e61c Dispose db connections pool on disconnect d1988b9 Set sql_mode callback on connect instead of checkout a1a8280 Fix excessive logging from db.sqlalchemy.session 9933bdd Get mysql_sql_mode parameter from config 96a2217 Prevent incorrect usage of _wrap_db_error() 20a7510 Add from_config() method to EngineFacade fea119e Drop special case for MySQL traditional mode, update unit tests dda24eb Introduce mysql_sql_mode option, remove old warning 0b5af67 Introduce a method to set any MySQL session SQL mode 8dccc7b Handle ibm_db_sa DBDuplicateEntry integrity errors 5b9e9f4 Fix doc build errors in db.sqlalchemy ac84a40 Update log translation domains 86707cd Remove None for dict.get() 0545121 Fix duplicating of SQL queries in logs fcf517d Update oslo log messages with translation domains 630d395 Don't use cfg.CONF in oslo.db ce69e7f Don't store engine instances in oslo.db
Earlier Cisco N1kv plugin was using list events api to
poll policies from VSM.
It was inefficient and caused delay in processing.
So, now Cisco N1kv plugin switched to list profiles to
poll policies from VSM.
If one quries NSX doing GET /ws.v1/lswitch/LS_UUID/lport and LS_UUID
is a UUID that does not exist in NSX. NSX raises a 404. If LS_UUID is *
NSX returns an empty result string. This patch fixes the fake_api_client
so that it's behavior is correct.
Carl Baldwin [Wed, 2 Apr 2014 16:53:33 +0000 (16:53 +0000)]
Replace loopingcall in notifier with a delayed send
The loopingcall thread here was started before processes fork and so
the thread stops working after the fork call. This is a problem that
will probably need to be worked out in the long run.
To ensure that this notifier works correctly in all processes, this
change replaces the persistent loopingcall thread with a thread
created on demand to delay and batch up notifications. The first
notification will trigger spawning the thread to wait to send it. Any
notifications that come in the meantime will notice that there is
already a thread waiting to send and will return without spawning.
jasonrad [Fri, 24 Jan 2014 20:43:24 +0000 (15:43 -0500)]
Add 2-leg configuration to Radware LBaaS Driver
Radware's load balancer currently supports three network interfaces.
Prior to this code addition, the configuration of those three
interfaces was: one interface for management, one for high
availability backup, and one for load balancing servers where the
IP that the outside world sees (VIP) and the IP that the servers
respond to (PIP) are identical.
This change adds a new "2-legged" configuration: one interface for
management, one interface for a VIP network, and one interface for the
PIP and the high availability backup network.
During VIP creation, the LBaaS driver detects if the incoming
configuration is "2-legged" by comparing VIP and PIP networks. If they
are different it allocates a new port on the PIP network in order to
obtain an IP address for the PIP and names that port after its
associated VIP ID. vDirect is then directed to create a load balancer
with the PIP network also used as the HA network instead of using the
vDirect container's default HA network configuration.
During VIP deletion, the LBaas driver detects if a port for the PIP
had been previously allocated by searching for and deleting port names
that map to the VIP ID.
In migration 1341ed32cc1e_nvp_netbinding_update Enum type had been
changed incorrectly from ('flat', 'vlan', 'stt', 'gre')
to ('flat', 'vlan', 'stt', 'gre', 'l3_ext') for PostgeSQL.
The same problem is taken place for vlan_type in migrations
38fc1f6789f8_cisco_n1kv_overlay from ('vlan', 'vxlan', 'trunk',
'multi-segment') to ('vlan', 'overlay', 'trunk', 'multi-segment')
and in 46a0efbd8f0_cisco_n1kv_multisegm from ('vlan', 'vxlan') to
('vlan', 'vxlan', 'trunk', 'multi-segment').
In this change request was added separate method for changing Enum
type for PostgreSQL.
allow delete_port to work when there are multiple floating ips
It is possible to associate multiple floating ip addresses with a single
port through the use of multiple *fixed* ip addresses, e.g.:
nova boot ... --nic net-id=my-net-id myserver
nova add-fixed-ip myserver my-net-id
nova add-floating-ip --fixed-address x.x.x.1 myserver y.y.y.y.1
nova add-floating-ip --fixed-address x.x.x.2 myserver y.y.y.y.2
In this situation, neutron.db.l3_db.disassociate_floatingips would throw
an exception:
Exception: Multiple floating IPs found for port <port-id>
This would prevent someone from deleting an instance that was associated
with multiple floating ips.
This patch corrects disassociate_floatingips so that it will
disassociate all floating ips associated with the port, allowing the
delete operation to proceed correctly.
Paul Ward [Fri, 18 Apr 2014 19:57:38 +0000 (14:57 -0500)]
Add nova_ca_certificates_file option to neutron
Previously, neutron had no way to pass a certificates file to the
novaclient. This change is to add that ability, similar to the way
you can pass a certificates file to neutronclient in nova.conf via
neturon_ca_certificates_file.
Code Review / openstack-build / neutron-build.git / log