| SridarK | i will let u know if i see some other issue | 00:00 |
|---|---|---|
| yushiro | thank you so much | 00:06 |
| *** SridarK has quit IRC | 00:57 | |
| *** vishwanathj has quit IRC | 01:02 | |
| *** vishwanathj has joined #openstack-fwaas | 01:02 | |
| *** yamamoto has quit IRC | 01:15 | |
| *** yamamoto has joined #openstack-fwaas | 01:21 | |
| *** vishwanathj has quit IRC | 02:03 | |
| *** vishwanathj has joined #openstack-fwaas | 02:03 | |
| *** yamamoto has quit IRC | 02:05 | |
| *** vishwanathj has quit IRC | 02:55 | |
| *** vishwanathj has joined #openstack-fwaas | 02:55 | |
| *** vishwanathj has quit IRC | 03:04 | |
| *** vishwanathj has joined #openstack-fwaas | 03:04 | |
| *** yamamoto has joined #openstack-fwaas | 03:05 | |
| *** vishwanathj has quit IRC | 03:08 | |
| *** vishwanathj has joined #openstack-fwaas | 03:08 | |
| *** yamamoto has quit IRC | 03:10 | |
| *** vishwanathj is now known as vishwanathj_zzz | 03:19 | |
| *** SarathMekala has joined #openstack-fwaas | 03:19 | |
| *** _SarathMekala_ has joined #openstack-fwaas | 03:23 | |
| *** mickeys has quit IRC | 03:28 | |
| *** chandanc_ has joined #openstack-fwaas | 03:28 | |
| *** vishwanathj_zzz has quit IRC | 03:29 | |
| *** vishwanathj_zzz has joined #openstack-fwaas | 03:29 | |
| *** yamamoto has joined #openstack-fwaas | 03:39 | |
| *** yushiro is now known as yushiro_lunch | 03:43 | |
| *** vishwanathj_zzz has quit IRC | 03:44 | |
| *** vishwanathj_zzz has joined #openstack-fwaas | 03:44 | |
| *** yamamoto has quit IRC | 03:45 | |
| *** SridarK_ has joined #openstack-fwaas | 03:48 | |
| *** vishwanathj_zzz has quit IRC | 03:53 | |
| *** vishwanathj_zzz has joined #openstack-fwaas | 03:53 | |
| *** vishwanathj_zzz has quit IRC | 03:55 | |
| *** vishwanathj_zzz has joined #openstack-fwaas | 03:55 | |
| *** chandanc_ has quit IRC | 04:00 | |
| *** yamamoto has joined #openstack-fwaas | 04:04 | |
| *** yamamoto has quit IRC | 04:18 | |
| *** chandanc_ has joined #openstack-fwaas | 04:19 | |
| chandanc_ | Hello All | 04:19 |
| SridarK_ | chandanc_: Hi | 04:24 |
| chandanc_ | Hello SridarK_ | 04:25 |
| SridarK_ | chandanc_: quick check can u tell me the version of distutils and setuptools on ur setup | 04:26 |
| chandanc_ | Can I pull the new PS and start testing ? | 04:26 |
| chandanc_ | sure | 04:26 |
| SridarK_ | chandanc_: yes i think u can | 04:26 |
| SridarK_ | chandanc_: there is one other thing | 04:26 |
| SridarK_ | was discussing with njohnston on the fwaas l3 ext patch | 04:27 |
| chandanc_ | ubuntu@devstack4:~$ pip list|grep setup | 04:27 |
| chandanc_ | setuptools (26.0.0) | 04:27 |
| chandanc_ | sure | 04:28 |
| *** yushiro_lunch is now known as yushiro | 04:28 | |
| SridarK_ | we may want to look to see if it is easy to call into the driver to clear all iptable rules for a particular port | 04:28 |
| chandanc_ | ok, I am not have the context | 04:29 |
| chandanc_ | but sure i can look at it | 04:29 |
| SridarK_ | https://review.openstack.org/#/c/355576/16/neutron_fwaas/services/firewall/agents/l3reference/firewall_l3_agent_v2.py L#239 | 04:30 |
| chandanc_ | distutils is the standard one in python 2.7 | 04:31 |
| chandanc_ | ok | 04:31 |
| SridarK_ | chandanc_: also if u have a few mins we can do a quick call | 04:31 |
| SridarK_ | ok my setuptools is 26.1.0 | 04:32 |
| chandanc_ | sure | 04:32 |
| chandanc_ | SridarK_, which PS should this change be done on | 04:35 |
| chandanc_ | the original l3 iptables driver is merged | 04:35 |
| *** vishwanathj_zzz has quit IRC | 04:38 | |
| *** vishwanathj_zzz has joined #openstack-fwaas | 04:38 | |
| *** yamamoto has joined #openstack-fwaas | 04:40 | |
| *** vishwanathj_zzz has quit IRC | 04:47 | |
| *** vishwanathj_zzz has joined #openstack-fwaas | 04:47 | |
| *** vishwanathj_zzz has quit IRC | 04:50 | |
| *** vishwanathj_zzz has joined #openstack-fwaas | 04:50 | |
| *** yamamoto has quit IRC | 05:00 | |
| *** vishwanathj_zzz has quit IRC | 05:06 | |
| *** vishwanathj_zzz has joined #openstack-fwaas | 05:07 | |
| *** vishwanathj_zzz has quit IRC | 05:11 | |
| *** vishwanathj_zzz has joined #openstack-fwaas | 05:11 | |
| *** vishwanathj_zzz has quit IRC | 05:15 | |
| *** vishwanathj_zzz has joined #openstack-fwaas | 05:15 | |
| *** vishwanathj_zzz has quit IRC | 05:17 | |
| *** vishwanathj_zzz has joined #openstack-fwaas | 05:17 | |
| *** vishwanathj_zzz has quit IRC | 05:26 | |
| *** vishwanathj_zzz has joined #openstack-fwaas | 05:26 | |
| *** vishwanathj_zzz has quit IRC | 05:38 | |
| *** vishwanathj_zzz has joined #openstack-fwaas | 05:38 | |
| *** vishwanathj_zzz has quit IRC | 05:40 | |
| *** vishwanathj_zzz has joined #openstack-fwaas | 05:40 | |
| *** yamamoto has joined #openstack-fwaas | 05:41 | |
| *** _SarathMekala_ has quit IRC | 05:44 | |
| *** _SarathMekala_ has joined #openstack-fwaas | 05:44 | |
| *** SarathMekala_ has joined #openstack-fwaas | 05:44 | |
| *** SarathMekala has quit IRC | 05:47 | |
| *** yushiro has quit IRC | 06:40 | |
| *** chandanc_ has quit IRC | 06:51 | |
| *** chandanc_ has joined #openstack-fwaas | 06:57 | |
| *** yushiro has joined #openstack-fwaas | 07:07 | |
| *** mickeys has joined #openstack-fwaas | 07:59 | |
| *** chandanc_ has quit IRC | 08:02 | |
| *** mickeys has quit IRC | 08:04 | |
| *** chandanc_ has joined #openstack-fwaas | 08:31 | |
| *** yamamoto has quit IRC | 09:48 | |
| *** yamamoto has joined #openstack-fwaas | 10:42 | |
| *** yamamoto_ has joined #openstack-fwaas | 10:59 | |
| *** chandanc_ has quit IRC | 10:59 | |
| *** yamamoto has quit IRC | 11:02 | |
| *** _SarathMekala_ has quit IRC | 11:04 | |
| *** SarathMekala_ has quit IRC | 11:04 | |
| *** yamamoto_ has quit IRC | 11:08 | |
| *** yamamoto has joined #openstack-fwaas | 11:16 | |
| *** yamamoto has quit IRC | 11:17 | |
| *** yamamoto has joined #openstack-fwaas | 11:29 | |
| *** yamamoto has quit IRC | 11:33 | |
| *** yamamoto has joined #openstack-fwaas | 12:22 | |
| *** xdcc has joined #openstack-fwaas | 13:21 | |
| *** chandanc_ has joined #openstack-fwaas | 13:22 | |
| njohnston | good morning all | 13:40 |
| chandanc_ | Hello njohnston | 13:41 |
| njohnston | How is it going, chandanc_? | 13:42 |
| chandanc_ | njohnston, i could not get too much done today. Had some stuff at office | 13:43 |
| njohnston | I understand. :-) | 13:43 |
| chandanc_ | SridarK_, told me about the driver API for handling router port delete | 13:43 |
| chandanc_ | was working a on that | 13:44 |
| SridarK_ | hi all | 13:47 |
| chandanc_ | One issue is that, the delete port is called in context of router delete, so it will be easy to get the router_info but difficult to get the firewall_group context | 13:47 |
| chandanc_ | *router port delete | 13:47 |
| SridarK_ | ok we will be in the context of the agent ext notification here | 13:48 |
| SridarK_ | https://review.openstack.org/#/c/355576/16/neutron_fwaas/services/firewall/agents/l3reference/firewall_l3_agent_v2.py L#231 | 13:51 |
| SridarK_ | njohnston: did u get a chance to look at my email i sent last night | 13:52 |
| SridarK_ | i wanted to understand if we are getting a BEFORE_ & AFTER_ notification - so we can glean the port_id | 13:53 |
| chandanc_ | SridarK_, yes what i mean is the data that you get will be for a router update, but will have to figure out the firewall group for the port | 13:54 |
| SridarK_ | chandanc_: since this is the port going away, if we find a redirect to the fw chain on this port - can we just clean that up | 13:56 |
| SridarK_ | so wondering if this can be done without worrying abt fwg association | 13:56 |
| chandanc_ | SridarK_, here is a sample of the test output of iptables I captured, http://paste.openstack.org/show/554078/ | 13:58 |
| chandanc_ | L60 | 13:58 |
| njohnston | SridarK_: Do you mean the one titled "Re: Testing the fwaas L3 extension"? | 13:58 |
| SridarK_ | njohnston: yes | 13:59 |
| SridarK_ | chandanc_: looking | 14:00 |
| chandanc_ | the target name is based on fwid | 14:01 |
| SridarK_ | chandanc_: qr-e2d7c3b0-49 - this is the port | 14:04 |
| chandanc_ | ya | 14:04 |
| chandanc_ | neutron-l3-agent-iv4adff2b87 | 14:04 |
| SridarK_ | and if this port is going away can we use that to knock off that rule | 14:04 |
| SridarK_ | that will send the pkt into the fwg chain | 14:05 |
| SridarK_ | that will be the only clean up needed | 14:05 |
| mfranc213 | morning everyone | 14:05 |
| chandanc_ | actually the API for rule delete seems to require the full definition | 14:06 |
| chandanc_ | still looking into it | 14:06 |
| SridarK_ | chandanc_: yes i think we may not be able to piggy back on the rule delete | 14:07 |
| SridarK_ | we will need an api where given the port-id going away - the driver can go thru the rules and knock off any entry corresponding to the port | 14:08 |
| SridarK_ | irrespective of the fwg | 14:08 |
| chandanc_ | let me look into it | 14:10 |
| SridarK_ | chandanc_: thx | 14:11 |
| SridarK_ | mfranc213: GM | 14:14 |
| mfranc213 | good morning SridarK_ | 14:14 |
| mfranc213 | you are up early; you were up late. | 14:15 |
| SridarK_ | fears of FFE :-) | 14:15 |
| mfranc213 | understandable! | 14:15 |
| SridarK_ | looks like we may have until Sep1 | 14:16 |
| SridarK_ | thankfully Aug has 31 days :-) | 14:16 |
| njohnston | indeed; I am on the neutron team meeting as well (#openstack-meeting for those who are interested) | 14:16 |
| mfranc213 | SridarK_ what if it were February!!? | 14:18 |
| SridarK_ | :-) | 14:18 |
| njohnston | then we'd have tons of days before Sept 1st. | 14:27 |
| njohnston | :-P | 14:27 |
| SridarK_ | oh yes, where is that time machine when we need it | 14:29 |
| *** chandanc_ is now known as chandanc_afk | 14:32 | |
| mfranc213 | time machine as a service | 14:34 |
| SridarK_ | yushiro: hi - pls let us know once u are comfortable on the CLI PS - can review that and hopefully have it ready for core reviewers by tomorrow | 14:36 |
| SridarK_ | or late evening US time | 14:36 |
| yushiro | SridarK_: Thanks for your information. I update soon.. | 14:39 |
| yushiro | SridarK_: However, UT hasn't finished yet. | 14:39 |
| mfranc213 | thank you yushiro. i'll review that too once it's ready. | 14:41 |
| yushiro | mfranc213: thank you too :) | 14:41 |
| *** SridarK_ has quit IRC | 14:41 | |
| *** diogogmt has joined #openstack-fwaas | 14:45 | |
| njohnston | xgerman: Do you have the ability to edit the 'implementation' field in https://blueprints.launchpad.net/neutron/+spec/fwaas-api-2.0 perhaps? | 15:01 |
| xgerman | let me check | 15:02 |
| xgerman | yes, I can | 15:02 |
| njohnston | Great! Perhaps 'good progress' is a better fit than 'deferred' at this point | 15:02 |
| xgerman | changed it | 15:02 |
| njohnston | thanks! | 15:03 |
| xgerman | anytime | 15:03 |
| *** diogogmt has quit IRC | 15:06 | |
| *** diogogmt has joined #openstack-fwaas | 15:08 | |
| *** chandanc_afk has quit IRC | 15:15 | |
| *** mickeys has joined #openstack-fwaas | 15:32 | |
| yushiro | Hi, I updated CLI PS but please wait 1 minutes! | 15:34 |
| yushiro | I forgot to fix pep8. | 15:34 |
| mfranc213 | :) | 15:35 |
| mfranc213 | looking now. thank you yushiro! | 15:36 |
| yushiro | mfranc213: you're welcome. I'll go home now :) | 15:37 |
| yushiro | I'm sorry for late. See you tomorrow's IRC meeting :) | 15:37 |
| yushiro | bye | 15:37 |
| *** yushiro has quit IRC | 15:39 | |
| *** vishwanathj_zzz is now known as vishwanathj | 15:56 | |
| *** chandanc_afk has joined #openstack-fwaas | 16:06 | |
| *** chandanc_afk is now known as chandanc_ | 16:09 | |
| *** SridarK_ has joined #openstack-fwaas | 16:11 | |
| *** yamamoto has quit IRC | 16:15 | |
| njohnston | https://github.com/openstack/neutron/blob/master/neutron/agent/l3/router_info.py#L490-L502 | 16:27 |
| njohnston | lunch; biab | 16:29 |
| *** vishwanathj_ has joined #openstack-fwaas | 16:44 | |
| *** diogogmt has quit IRC | 16:52 | |
| *** diogogmt has joined #openstack-fwaas | 17:07 | |
| *** SridarK_ has quit IRC | 17:11 | |
| *** yamamoto has joined #openstack-fwaas | 17:16 | |
| *** yamamoto has quit IRC | 17:22 | |
| chandanc_ | njohnston, i am getting the loading error | 17:32 |
| chandanc_ | 2016-08-30 16:20:30.082 ERROR stevedore.extension [req-efb6c939-46e0-48de-90c5-fdb3f4eb911e None None] Could not load 'fwaas': object.__init__() takes no parameters | 17:32 |
| njohnston | shouldn't it be trying to load fwaasv2? | 17:33 |
| chandanc_ | ya | 17:33 |
| njohnston | actually, not even that, it should be trying to load 'firewall_v2' looking at setup.cfg | 17:35 |
| njohnston | I am looking for where that 'fwaas' string is coming from | 17:35 |
| chandanc_ | oh ok, sorry i dint catch you first | 17:36 |
| mfranc213 | chandanc_ could you post the entire traceback? | 17:36 |
| chandanc_ | i dont get a traceback, just that in the log file | 17:36 |
| njohnston | hmm | 17:37 |
| mfranc213 | what precedes that in the log file? | 17:38 |
| *** chandanc__ has joined #openstack-fwaas | 17:40 | |
| *** chandanc_ has quit IRC | 17:40 | |
| *** xdcc has quit IRC | 17:40 | |
| mfranc213 | chandanc_ and chandanc__ : ) also, which log file is that? | 17:42 |
| chandanc__ | tail -f /opt/stack/logs/q-l3.log |grep -i extension | 17:43 |
| chandanc__ | 2016-08-30 17:25:36.965 ERROR stevedore.extension [req-b99143be-d621-4fed-9f04-09d28a3dbd36 None None] Could not load 'firewall_v2': object.__init__() takes no parameters | 17:43 |
| chandanc__ | 2016-08-30 17:25:36.965 INFO neutron.agent.agent_extension | 17:43 |
| chandanc__ | sorry njohnston , the fwaas was due to my env | 17:43 |
| njohnston | ok | 17:44 |
| chandanc__ | i think it is due to the host=host parameter we discussed yday | 17:44 |
| njohnston | the one that we removed? | 17:45 |
| mfranc213 | i think you're right chandanc__ | 17:46 |
| mfranc213 | that one went back and forth, and i had to add it in to get my UTs to work. | 17:46 |
| chandanc__ | yes looks like its still there | 17:46 |
| chandanc__ | oh ok | 17:46 |
| mfranc213 | something about my env was different | 17:46 |
| mfranc213 | can you remove it locally just to verify that that in fact is the source of the trace you're getting? | 17:47 |
| chandanc__ | ok | 17:47 |
| mfranc213 | thank you. | 17:47 |
| mfranc213 | (and sorry.) | 17:47 |
| chandanc__ | no probs | 17:48 |
| chandanc__ | 2016-08-30 17:30:51.783 CRITICAL neutron [req-f981493a-2118-4edd-a9a3-aa30c2b2f673 None None] AttributeError: 'L3WithFWaaS' object has no attribute '_register_rpc_consumers' | 17:49 |
| mfranc213 | thinking | 17:49 |
| njohnston | OK, that may be a stevedore thing since I don't see _register_rpc_customers in the fwaas repo at all. | 17:50 |
| njohnston | nor in the neutron repo | 17:51 |
| mfranc213 | hmmm | 17:51 |
| njohnston | let's add this to FWaaSL3AgentExtension: | 17:52 |
| njohnston | def _register_rpc_customers(self): | 17:52 |
| njohnston | pass | 17:52 |
| njohnston | and see what happens | 17:52 |
| *** SridarK_ has joined #openstack-fwaas | 17:52 | |
| mfranc213 | https://github.com/openstack/neutron/blob/master/neutron/agent/l2/extensions/qos.py#L202 | 17:54 |
| mfranc213 | this one: https://github.com/openstack/neutron/blob/master/neutron/agent/l2/extensions/qos.py#L210 | 17:54 |
| njohnston | I didn't think we would be loading anything in L2 at this point, would we? | 17:55 |
| mfranc213 | no, sorry, my point is that we need to add a _register_rpc_consumers method in the fwaas agent extension. | 17:55 |
| njohnston | So it seems, though I don't see why | 17:56 |
| chandanc__ | diff --git a/neutron_fwaas/services/firewall/agents/l3reference/firewall_l3_agent_v2.py b/neutron_fwaas/services/firewall/agents/l3reference/firewall_l3_a | 17:56 |
| chandanc__ | index 4556e6f..640a3f2 100644 | 17:56 |
| chandanc__ | --- a/neutron_fwaas/services/firewall/agents/l3reference/firewall_l3_agent_v2.py | 17:56 |
| chandanc__ | +++ b/neutron_fwaas/services/firewall/agents/l3reference/firewall_l3_agent_v2.py | 17:56 |
| chandanc__ | @@ -82,6 +82,9 @@ class FWaaSL3AgentExtension(l3_agent_extension.L3AgentCoreResourceExtension): | 17:56 |
| chandanc__ | self.agent_api = agent_api | 17:56 |
| chandanc__ | self.fwaas_driver.consume_api(self.agent_api) | 17:56 |
| chandanc__ | 17:56 | |
| chandanc__ | + def _register_rpc_consumers(self, connection): | 17:56 |
| chandanc__ | + pass | 17:56 |
| chandanc__ | + | 17:56 |
| chandanc__ | def __init__(self, host, conf): | 17:56 |
| chandanc__ | LOG.debug("Initializing firewall group agent") | 17:56 |
| chandanc__ | self.agent_api = None | 17:56 |
| chandanc__ | @@ -111,7 +114,8 @@ class FWaaSL3AgentExtension(l3_agent_extension.L3AgentCoreResourceExtension): | 17:56 |
| chandanc__ | self.services_sync_needed = False | 17:56 |
| chandanc__ | self.fwplugin_rpc = FWaaSL3PluginApi(f_const.FIREWALL_PLUGIN, | 17:56 |
| chandanc__ | host) | 17:57 |
| chandanc__ | - super(FWaaSL3AgentExtension, self).__init__(host=host) | 17:57 |
| chandanc__ | + #super(FWaaSL3AgentExtension, self).__init__(host=host) | 17:57 |
| chandanc__ | + super(FWaaSL3AgentExtension, self).__init__() | 17:57 |
| chandanc__ | 17:57 | |
| chandanc__ | with this in i am getting the following error | 17:57 |
| chandanc__ | error | 17:57 |
| chandanc__ | 2016-08-30 17:38:06.042 TRACE neutron.agent.l3.agent Traceback (most recent call last): | 17:57 |
| chandanc__ | 2016-08-30 17:38:06.042 TRACE neutron.agent.l3.agent File "/opt/stack/neutron/neutron/agent/l3/agent.py", line 498, in _process_router_update | 17:57 |
| chandanc__ | 2016-08-30 17:38:06.042 TRACE neutron.agent.l3.agent self._process_router_if_compatible(router) | 17:57 |
| chandanc__ | 2016-08-30 17:38:06.042 TRACE neutron.agent.l3.agent File "/opt/stack/neutron/neutron/agent/l3/agent.py", line 435, in _process_router_if_compatible | 17:57 |
| chandanc__ | 2016-08-30 17:38:06.042 TRACE neutron.agent.l3.agent self._process_updated_router(router) | 17:57 |
| chandanc__ | 2016-08-30 17:38:06.042 TRACE neutron.agent.l3.agent File "/opt/stack/neutron/neutron/agent/l3/agent.py", line 452, in _process_updated_router | 17:57 |
| chandanc__ | 2016-08-30 17:38:06.042 TRACE neutron.agent.l3.agent self.l3_ext_manager.update_router(self.context, router) | 17:57 |
| chandanc__ | 2016-08-30 17:38:06.042 TRACE neutron.agent.l3.agent File "/opt/stack/neutron/neutron/agent/l3/l3_agent_extensions_manager.py", line 54, in update_router | 17:57 |
| chandanc__ | 2016-08-30 17:38:06.042 TRACE neutron.agent.l3.agent extension.obj.update_router(context, data) | 17:57 |
| chandanc__ | 2016-08-30 17:38:06.042 TRACE neutron.agent.l3.agent File "/opt/stack/neutron-fwaas/neutron_fwaas/services/firewall/agents/l3reference/firewall_l3_agent_v2.py", line 275, in update_router | 17:57 |
| chandanc__ | 2016-08-30 17:38:06.042 TRACE neutron.agent.l3.agent updated_router.router['id']) | 17:57 |
| chandanc__ | 2016-08-30 17:38:06.042 TRACE neutron.agent.l3.agent AttributeError: 'dict' object has no attribute 'router' | 17:57 |
| mfranc213 | the _register_rpc_consumers method was removed from that changeset somewhere along the way: https://review.openstack.org/#/c/355576/4..18/neutron_fwaas/services/firewall/agents/l3reference/firewall_l3_agent_v2.py | 17:59 |
| mfranc213 | @114 | 17:59 |
| mfranc213 | chandanc__ would you be able to past the git diff and the traceback in a pastebin? | 17:59 |
| chandanc__ | http://paste.openstack.org/show/564876/ | 18:01 |
| njohnston | Thanks chandanc__, re-pull 355576 and give it another try; I incorporated your edits. | 18:01 |
| chandanc__ | njohnston, holdon i have some more | 18:02 |
| njohnston | ok | 18:02 |
| mfranc213 | thank you njohnston. | 18:03 |
| mfranc213 | njohnston: do your UTs still work? | 18:03 |
| njohnston | mfranc213: Running them now, they take their sweet time | 18:03 |
| mfranc213 | oh boy yeah | 18:04 |
| chandanc__ | http://paste.openstack.org/show/564877/ | 18:05 |
| chandanc__ | with this i get the following error | 18:05 |
| chandanc__ | error | 18:05 |
| chandanc__ | 2016-08-30 17:45:50.851 TRACE neutron_fwaas.services.firewall.agents.l3reference.firewall_l3_agent_v2 AttributeError: 'L3AgentExtensionAPI' object has no attribute 'router_in_ns' | 18:05 |
| njohnston | does your neutron have the most recent patchset of https://review.openstack.org/357503 | 18:06 |
| chandanc__ | I am not sure if all these changes are compatible with the RPC and ext APIs | 18:06 |
| chandanc__ | let me check | 18:07 |
| chandanc__ | hmm looks like i have a older version of that | 18:11 |
| mfranc213 | so many moving parts :( | 18:12 |
| chandanc__ | sorry for that | 18:12 |
| chandanc__ | trying to pull it | 18:12 |
| mfranc213 | no apologies necessary!! you are doing the heavy lifting for us now. | 18:13 |
| SridarK_ | +1 | 18:13 |
| chandanc__ | :) thanks | 18:13 |
| mfranc213 | fyi we know we'll have at least one more PS for the v2 l3 agent extension fixing an issue with an edge case that i just saw and talked to nate about. this won't affect current testing though. | 18:18 |
| njohnston | chandanc__: I also just updated https://review.openstack.org/355576 with a minor fix | 18:20 |
| SridarK_ | njohnston:, mfranc213: do we also need to refactor to using a different rpc topic - we talked some abt this | 18:21 |
| njohnston | yes. question is, should that be a part of this patch? | 18:21 |
| mfranc213 | njohnston: i still believe that self.fwaas_driver.consume_api(self.agent_api) must be there, just only if fwaas is enabled. | 18:22 |
| njohnston | but the driver doesn't use the agent_api at all | 18:22 |
| mfranc213 | yes, but that's THIS driver. | 18:22 |
| mfranc213 | i think this is an abc abstract method that the driver must implement even if just to pass? can't remember actually. | 18:23 |
| njohnston | if in the future we have a driver that needs it then we'll solve that issue at that time | 18:23 |
| SridarK_ | not sure on that, if it is just a change on the topic we can pick this up | 18:23 |
| mfranc213 | njohnston: this is technical debt we can easily avoid though. | 18:23 |
| njohnston | SridarK_: Let's make it a separate change - we don't want to confuse resolution of this loading issue | 18:24 |
| SridarK_ | njohnston: that is fine | 18:24 |
| *** chandanc_ has joined #openstack-fwaas | 18:24 | |
| SridarK_ | i am for getting this basic integration done - so agreed | 18:25 |
| mfranc213 | well i see that consume_api i see was removed from the driver also somewhere along the way. i do think that this should be an abc abstract class | 18:25 |
| mfranc213 | but it is the 11th hour. | 18:25 |
| njohnston | mfranc213: I think it is creating technical debt if we leave it in. We don't know, really, what the requirements of a future driver will be. We might imagine we do, but we don't. So we would be leaving in code that is currently useless and may always be so, which to me is definitely technical debt. Let's cross that bridge when we get to it. | 18:25 |
| SridarK_ | so now the plugin - agent will use the current rpc topic | 18:25 |
| mfranc213 | i disagree but let's move along :) | 18:26 |
| *** chandanc__ has quit IRC | 18:26 | |
| njohnston | SridarK_: Also, if we can get in a topic change before the cutoff and feel sure that it works, so much the better. But if not, then it's not the end of the world. :-) | 18:26 |
| SridarK_ | njohnston: yes - i want to be able to test that well - so lets get this done first | 18:27 |
| mfranc213 | njohnston: with the latest PS, my UTs no longer trace. thank you! | 18:29 |
| njohnston | Sure thing! I have a TODO left in there to figure out why host is a required argument, but that is a Topic For Further Student Research, as my secondary school math teacher would say. | 18:30 |
| *** mickeys has quit IRC | 18:32 | |
| njohnston | How is it going chandanc_? | 18:33 |
| *** chandanc__ has joined #openstack-fwaas | 18:33 | |
| *** chandanc_ has quit IRC | 18:36 | |
| *** chandanc_ has joined #openstack-fwaas | 18:37 | |
| *** chandanc__ has quit IRC | 18:38 | |
| chandanc_ | njohnston, mfranc213 : i am still getting similar errors | 18:39 |
| chandanc_ | 2016-08-30 18:20:56.835 TRACE neutron_fwaas.services.firewall.agents.l3reference.firewall_l3_agent_v2 File "/opt/stack/neutron-fwaas/neutron_fwaas/services/firewall/agents/l3reference/firewall_l3_agent_v2.py", line 258, in add_router | 18:39 |
| chandanc_ | 2016-08-30 18:20:56.835 TRACE neutron_fwaas.services.firewall.agents.l3reference.firewall_l3_agent_v2 self._process_router_update(new_router) | 18:39 |
| chandanc_ | 2016-08-30 18:20:56.835 TRACE neutron_fwaas.services.firewall.agents.l3reference.firewall_l3_agent_v2 File "/opt/stack/neutron-fwaas/neutron_fwaas/services/firewall/agents/l3reference/firewall_l3_agent_v2.py", line 236, in _process_router_update | 18:39 |
| chandanc_ | 2016-08-30 18:20:56.835 TRACE neutron_fwaas.services.firewall.agents.l3reference.firewall_l3_agent_v2 ctx = context.Context('', updated_router.router['tenant_id']) | 18:39 |
| chandanc_ | 2016-08-30 18:20:56.835 TRACE neutron_fwaas.services.firewall.agents.l3reference.firewall_l3_agent_v2 AttributeError: 'dict' object has no attribute 'router' | 18:39 |
| chandanc_ | 2016-08-30 18:20:56.835 TRACE neutron_fwaas.services.firewall.agents.l3reference.firewall_l3_agent_v2 | 18:39 |
| njohnston | ooh, can you put that in a paste? | 18:40 |
| njohnston | hard to read in my IRC client | 18:40 |
| chandanc_ | sure | 18:40 |
| chandanc_ | http://paste.openstack.org/show/564880/ | 18:40 |
| chandanc_ | basically dict/object issue | 18:41 |
| mfranc213 | looking now. thank you chandanc_ | 18:41 |
| mfranc213 | is that the entirety of the trace? | 18:41 |
| mfranc213 | usually you get more | 18:41 |
| chandanc_ | here is one more change that was different | 18:41 |
| chandanc_ | CC229 #if router_id not in self.agent_api.router_in_ns(router_id): | 18:41 |
| chandanc_ | 230 if not self.agent_api.router_in_ns(router_id): | 18:41 |
| mfranc213 | yes, the uncommented line is correct. njohnston: ^^ (that is my fault and nate will be the one fixing it!) | 18:43 |
| mfranc213 | looking further at the trace back now. | 18:43 |
| mfranc213 | nvm don't need more of the traceback, chandanc_ | 18:44 |
| chandanc_ | i think you are getting a dict for the router arg | 18:44 |
| chandanc_ | sure | 18:44 |
| njohnston | chandanc_: Here's the thing - your error occurs at line 236, where `updated_router.router['tenant_id']` is referenced. | 18:45 |
| mfranc213 | njohnston: i find this traceback strange. why wouldn't line 229 have traced? | 18:45 |
| njohnston | that is the question I was just about to ask! :-D | 18:46 |
| mfranc213 | something is goofy | 18:46 |
| njohnston | lines 227-229 should have traced | 18:46 |
| njohnston | 229 is: router_id = updated_router.router['id'] | 18:46 |
| mfranc213 | yep | 18:46 |
| mfranc213 | and there's no conditional. it's funky! | 18:46 |
| chandanc_ | 227 updated_router['id']) | 18:47 |
| chandanc_ | 228 router_id = updated_router['id'] | 18:47 |
| chandanc_ | CC229 #if router_id not in self.agent_api.router_in_ns(router_id): | 18:47 |
| chandanc_ | 230 if not self.agent_api.router_in_ns(router_id): | 18:47 |
| mfranc213 | hmmm | 18:48 |
| mfranc213 | your is different from ours | 18:48 |
| njohnston | chandanc_: OK, so your code is different from outs | 18:48 |
| chandanc_ | i would have corrected during the local changes | 18:48 |
| njohnston | so let's shift to just doing a straight dict here, we can adjust the UTs later | 18:48 |
| mfranc213 | yeah, but njohnston, do you see how to fix now? | 18:48 |
| mfranc213 | yes njohnston | 18:49 |
| mfranc213 | you still have the baton, right? | 18:49 |
| njohnston | yes mfranc213 | 18:49 |
| mfranc213 | groovy | 18:49 |
| njohnston | and line 240 becomes this, right? for port in updated_router['internal_ports']: | 18:49 |
| mfranc213 | chandanc_ do you want to paste a git diff? | 18:49 |
| mfranc213 | yes njohnston | 18:50 |
| chandanc_ | sure | 18:50 |
| mfranc213 | njohnston you'll see a few occurrences of update_router.router | 18:51 |
| njohnston | http://paste.openstack.org/show/564882/ | 18:51 |
| mfranc213 | but also njohnston you'll need to fix line 261 | 18:51 |
| *** mestery has quit IRC | 18:52 | |
| chandanc_ | http://paste.openstack.org/show/564883/ | 18:52 |
| njohnston | and line 153 | 18:52 |
| mfranc213 | njohnston, not sure yet about 153... thinking | 18:53 |
| mfranc213 | still not sure, actually. | 18:53 |
| njohnston | If that data structure doesn't exist, I would think it never existed | 18:53 |
| mfranc213 | think your prob right, just not sure. | 18:54 |
| mfranc213 | but i'm good with changing that. we can change it back if needed :) | 18:54 |
| njohnston | openstackgerrit Nate Johnston proposed openstack/neutron-fwaas: FWaaS v2 utilize L3 Agent Extension framework https://review.openstack.org/355576 | 18:55 |
| mfranc213 | this is the ultimate in pair programming! :) | 18:56 |
| *** mestery has joined #openstack-fwaas | 18:57 | |
| mfranc213 | and it's not even pair programming! it's triplet programming! | 18:58 |
| *** chandanc__ has joined #openstack-fwaas | 18:58 | |
| mfranc213 | njohnston: i'm running another UT and then am going to check out that line 153 thing. i'm sure you're right though. | 18:59 |
| *** chandanc_ has quit IRC | 19:00 | |
| mfranc213 | UTs check out. | 19:00 |
| SridarK_ | so now we are getting a router_info, as per the companion change in neutron | 19:00 |
| SridarK_ | ^^^ single router_info corresponding to the router | 19:00 |
| njohnston | excellent! | 19:01 |
| mfranc213 | sweet! | 19:01 |
| mfranc213 | njohnston: i think we need to revert @153. check it out: https://review.openstack.org/#/c/357503/16/neutron/agent/l3/l3_agent_extension_api.py | 19:04 |
| mfranc213 | am i getting that right? | 19:04 |
| mfranc213 | the line in the link is @50 | 19:04 |
| njohnston | hmm, yes, it seems that is ripe for reversion | 19:05 |
| mfranc213 | so, to clarify, i believe that because of https://review.openstack.org/#/c/357503/16/neutron/agent/l3/l3_agent_extension_api.py@50, we need to revert https://review.openstack.org/#/c/355576/21/neutron_fwaas/services/firewall/agents/l3reference/firewall_l3_agent_v2.py@153 | 19:05 |
| njohnston | I have the change staged for the next PS | 19:06 |
| chandanc__ | I will logout now, will catch up in the morning | 19:06 |
| mfranc213 | thank you chandanc__ | 19:06 |
| mfranc213 | SridarK_ are you still able to test? | 19:06 |
| chandanc__ | no probs | 19:06 |
| chandanc__ | awaiting ext patch update, will catch up in the morning | 19:07 |
| mfranc213 | thank you so much chandanc__ i hope you have a good night. | 19:07 |
| SridarK_ | mfranc213: let me pick this up | 19:07 |
| mfranc213 | thank you SridarK_ | 19:08 |
| SridarK_ | chandanc__: thx | 19:08 |
| SridarK_ | so we will take the latest PS | 19:08 |
| mfranc213 | let's let njohnston make the @153 fix first | 19:08 |
| SridarK_ | chandanc__: b4 u sign off | 19:08 |
| mfranc213 | since we know that needs to be done. | 19:08 |
| SridarK_ | ok good | 19:08 |
| njohnston | done | 19:08 |
| SridarK_ | chandanc__: did u get to the point with a l3 agent restart things were good | 19:09 |
| chandanc__ | the patch loads but needs some more code changes, njohnston, mfranc213 are updating the patch | 19:09 |
| SridarK_ | chandanc__: ok | 19:10 |
| SridarK_ | i pulling in the latest PS | 19:11 |
| chandanc__ | ok | 19:11 |
| chandanc__ | drop me a mail, i will take it up in the morning | 19:11 |
| SridarK_ | ok chandanc__ GN | 19:12 |
| chandanc__ | GN SridarK_ | 19:13 |
| njohnston | thanks much chandanc__!! | 19:15 |
| *** vishwanathj_ has quit IRC | 19:16 | |
| *** chandanc__ has quit IRC | 19:17 | |
| SridarK_ | ok now my devstack misbehaves | 19:25 |
| SridarK_ | restacking | 19:25 |
| *** mickeys has joined #openstack-fwaas | 19:33 | |
| *** mickeys has quit IRC | 19:39 | |
| *** mickeys has joined #openstack-fwaas | 20:36 | |
| *** mickeys has quit IRC | 20:41 | |
| njohnston | SridarK_: How's it going? | 20:45 |
| *** mickeys has joined #openstack-fwaas | 20:46 | |
| njohnston | In my devstack I am back to the 'ImportError: Plugin 'firewall_v2' not found.' error that Chandan worked around. I figured it out before, but it's been a while and I have forgotten what I did. | 20:54 |
| mfranc213 | SridarK_: i am heading out soon but will be checking in periodically. | 20:58 |
| SridarK_ | mfranc213: njohnston: finally got my devstack up with the patches | 20:59 |
| SridarK_ | DEBUG neutron.wsgi [-] agent.extensions = ['fwaas'] from (pid=30466) log_opt_values /usr/local/lib/python2.7/dist-packages/oslo_config/cfg.py:2626 | 20:59 |
| mfranc213 | that is most excellent | 20:59 |
| njohnston | excellent! I am about to relocate home, but I will hop on from there | 21:00 |
| SridarK_ | so i think it is loaded | 21:00 |
| SridarK_ | let me try the curl cmds | 21:00 |
| mfranc213 | i am available too sridar, throughout the night. | 21:00 |
| mfranc213 | well, the evening :) | 21:00 |
| SridarK_ | mfranc213: ok :-) | 21:00 |
| SridarK_ | yes lets hash this out | 21:00 |
| SridarK_ | and have it up by night | 21:00 |
| SridarK_ | thx | 21:01 |
| mfranc213 | i'm leaving to bring my son to the orthodontist in 1/2 hour but will be back i think by 4 pm your time. SridarK_ did you get my telephone number? | 21:02 |
| SridarK_ | mfranc213: ok got it many thx | 21:27 |
| *** diogogmt has quit IRC | 22:20 | |
| *** diogogmt has joined #openstack-fwaas | 22:32 | |
| SridarK_ | njohnston: , mfranc213: i think we will need to look into the rpc init | 23:03 |
| SridarK_ | mfranc213: http://paste.openstack.org/show/564918/ | 23:38 |
| *** yushiro has joined #openstack-fwaas | 23:41 | |
| mfranc213 | ping Sridark_ | 23:48 |
| SridarK_ | mfranc213: hi | 23:51 |
| mfranc213 | hi there. | 23:52 |
| SridarK_ | mfranc213: pls go ahead | 23:52 |
| mfranc213 | so it may seem obvious, but can you tell me exactly what you were doing: a curl call, etc.? | 23:52 |
| SridarK_ | oh yes | 23:53 |
| mfranc213 | can you send me the curl call you were making? i just need to situate myself. | 23:53 |
| SridarK_ | i did a curl call to create a fwg with a port | 23:53 |
| SridarK_ | yes sure | 23:53 |
| mfranc213 | cause that command is everywhere, plugin, etc. | 23:53 |
| SridarK_ | curl -X POST -H "x-auth-token:$TOKEN" -H "Content-type: application/json" -d '{"firewall_group": {"name": "fwaas-test", "ports": ["abac7784-30b0-4726-98a2-d10914815f76"], "ingress_firewall_policy_id": "7eb74342-2628-465a-8552-415ec0359f8e", "egress_firewall_policy_id": "7eb74342-2628-465a-8552-415ec0359f8e"}}' 172.29.87.168:9696/v2.0/fwaas/firewall_groups | 23:53 |
| SridarK_ | yes i verified that the db entries were created correctly | 23:54 |
| mfranc213 | aha that is good to know. | 23:54 |
| mfranc213 | have you been able to do any other actions: update or delete? | 23:54 |
| mfranc213 | do all creates fail? | 23:54 |
| SridarK_ | ah, the rules and policies do not message to the agent | 23:55 |
| SridarK_ | so for our integ test - we really need to use the create fwg | 23:55 |
| SridarK_ | my initial thought was that agent was not loaded properly | 23:56 |
| mfranc213 | yeah, you said in text that it was, and that you thought it was a wiring issue. | 23:56 |
| mfranc213 | can you put the breakpoint lower down in the stack? | 23:57 |
| SridarK_ | but with it is getting loaded - as verified with a breakpoint in __init__ | 23:57 |
| SridarK_ | ok in the rpc - ok let me look at that | 23:57 |
| SridarK_ | For the plugin to Agent we use the L3Agent topic | 23:58 |
| SridarK_ | so with the switch to the L3 Agent ext framework - we will need to see if there is something more needed - may be u can look at that | 23:59 |
| SridarK_ | let me see if i can dig up some info from rpc | 23:59 |
| mfranc213 | yes, looking at that now | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!