| mfranc213 | because this all worked without the L3 agent ext framework, right? (just to verify; apologies for very low-level questions, since i was out.) | 00:00 |
|---|---|---|
| mfranc213 | which is to confirm with you that the smoking gun would have to be the change to the l3 agent ext framework...? | 00:01 |
| SridarK_ | mfranc213: yes we tested that with the first agent patch (that merged) | 00:02 |
| SridarK_ | no worries | 00:02 |
| mfranc213 | thank you sridar | 00:03 |
| mfranc213 | thinking/looking... | 00:03 |
| mfranc213 | i wonder if it is related to the change at https://review.openstack.org/#/c/355576/22/neutron_fwaas/services/firewall/agents/l3reference/firewall_l3_agent_v2.py@119 | 00:08 |
| SridarK_ | ok i am in the rpc layer, let me pass on this info to u | 00:11 |
| SridarK_ | http://paste.openstack.org/show/564930/ | 00:12 |
| SridarK_ | so on the rpc | 00:13 |
| SridarK_ | In "/usr/local/lib/python2.7/dist-packages/oslo_messaging/rpc/dispatcher.py", line 155, in dispatch | 00:13 |
| SridarK_ | so it does land up at the L3Agent as expected, but look at the attributes of that endpoint - the method of interest is missing | 00:14 |
| SridarK_ | which is probab right in that as without the inheritance - create_firewall_group is not going to be there | 00:15 |
| SridarK_ | so we may either need to sprinkle some other magic dust so we are also part of that rpc endpoint | 00:16 |
| mfranc213 | thinking about what you said. | 00:18 |
| mfranc213 | was there another alternative you were going to present? | 00:18 |
| SridarK_ | no i am not sure on this either | 00:18 |
| mfranc213 | and we know that the fwaas agent extension has been loaded properly? | 00:19 |
| SridarK_ | but looking at the change u pointed to - let me try to undo that | 00:19 |
| SridarK_ | that is something i would like to understand more too | 00:20 |
| SridarK_ | ok undoing that change back to host=host did not help | 00:23 |
| mfranc213 | yeah. what happened then? | 00:24 |
| njohnston | I think I have a theory. What we had before, with the inheritance model, had the FWaaS agent code taking over entirely - it inherited from L3NATAgent and then became the primary identity of the agent. So it had access to all L3_AGENT RPC messages. | 00:25 |
| njohnston | But now with the new code, we are separated, and there are only two ways for messages to get sent to the FWaaS code | 00:26 |
| mfranc213 | the inheritance was the other way around, but i think the principal is the same | 00:26 |
| njohnston | First, the L3NATAgent forwards the message (i.e. L# agent extensions API object) | 00:27 |
| njohnston | Second, the FWaaS agent extension configures itself as a direct RPC listener | 00:27 |
| njohnston | (which we haven't coded yet) | 00:27 |
| mfranc213 | yes | 00:27 |
| mfranc213 | good. | 00:27 |
| mfranc213 | well, bad, but good. | 00:27 |
| SridarK_ | njohnston: yes that makes sense | 00:28 |
| mfranc213 | we were thinking of that inheritance model primarily/solely in terms of getting access to router info | 00:28 |
| njohnston | so the message is getting dropped because we haven't set anything up to route it | 00:28 |
| mfranc213 | but that model also enabled the rpd itself. | 00:28 |
| mfranc213 | rpc | 00:28 |
| njohnston | yes | 00:28 |
| mfranc213 | so on the qos side... | 00:28 |
| njohnston | on the QoS side all that mattered were messages on the PORT RPC topic | 00:28 |
| SridarK_ | hmm but "to forward the RPC msg" - we will need something else in neutron to do this ? | 00:29 |
| mfranc213 | they had both traditional and new-style rpc messaging on qos. | 00:29 |
| mfranc213 | s/had/have | 00:29 |
| mfranc213 | port deletes are done using the traditional rpc mechanism afaict | 00:31 |
| njohnston | The easiest solution at this point, if I am right, would be to go ahead and change the RPC topic for FWG, FWR, etc. objects from L3_AGENT to L3_FWAAS, and set up the FWaaS agent extension to subscribe to that topic directly | 00:31 |
| SridarK_ | njohnston: that is my thought too | 00:31 |
| njohnston | the problem is not new style versus old style messaging, I believe | 00:31 |
| mfranc213 | i'm catching up on this so you two go ahead while i try to wrap my brain around it | 00:33 |
| njohnston | we ave two different kinds of messaging - router updates, and firewall object updates. We handled the first set, but the second set is a totally different kind of messaging. Damn, I wish I had figured this out before. | 00:34 |
| SridarK_ | njohnston: the second set is basically the CRUD | 00:35 |
| SridarK_ | plugin - agent | 00:35 |
| njohnston | right | 00:37 |
| mfranc213 | njohnston are you talking, at least in part, about changing lines 74-76, and 84 on https://review.openstack.org/#/c/355576/22/neutron_fwaas/services/firewall/agents/l3reference/firewall_l3_agent_v2.py? | 00:38 |
| SridarK_ | i am trying out a quick validation - by adding a new topic | 00:40 |
| njohnston | it means fleshing out _register_rpc_customers on line 85 | 00:41 |
| mfranc213 | good. | 00:42 |
| mfranc213 | i'm caught up now. nate, you still have the baton though. | 00:42 |
| mfranc213 | i would defn be way slower than you. | 00:42 |
| njohnston | it'd be something like | 00:43 |
| SridarK_ | i am adding this | 00:43 |
| SridarK_ | def start_rpc_listeners(self): self.endpoints = [FWaaSL3AgentExtension(self)] self.conn = n_rpc.create_connection() self.conn.create_consumer( 'firewall_agent', self.endpoints, fanout=False) return self.conn.consume_in_threads() | 00:43 |
| njohnston | registry.subscribe(fwaas_callback, resources.FWAAS, events.UPDATE) | 00:43 |
| SridarK_ | ouch that is not great | 00:43 |
| SridarK_ | njohnston: ok this is a new style | 00:43 |
| SridarK_ | i just added this - let me quickly change the topic in teh plugin for a quick test | 00:44 |
| njohnston | SridarK_: you're right, I am wrong | 00:44 |
| *** yamamoto_ has joined #openstack-fwaas | 00:44 | |
| njohnston | so will we end up bundling this all together into this change? I kind of think we have to, but I am not sure | 00:48 |
| mfranc213 | i think we have to | 00:49 |
| SridarK_ | njohnston: yes i think so too - lets add this change | 00:56 |
| SridarK_ | but let me see if it works | 00:56 |
| njohnston | sounds good. Do you want to send the necessary changes to me and I add them so you can preserve +2, or do you want to go ahead and add them directly? I am fine either way. | 00:57 |
| mfranc213 | oh boy nate. | 00:57 |
| njohnston | mfranc213: ? | 00:57 |
| mfranc213 | good thinking on that one | 00:57 |
| njohnston | brb, I need to go put my daughter to bed | 00:57 |
| mfranc213 | i would have forgotten about the +2 piece of it. | 00:57 |
| SridarK_ | njohnston: no i will send it across to u | 01:05 |
| mfranc213 | did it work SridarK_ ? | 01:05 |
| SridarK_ | mfranc213: sigh - ran into some stupid issues | 01:08 |
| SridarK_ | mfranc213: do u have a few mins for a quick call ? | 01:08 |
| mfranc213 | yes | 01:09 |
| SridarK_ | ok let me get to a conf room | 01:09 |
| mfranc213 | webex? | 01:09 |
| SridarK_ | y | 01:09 |
| SridarK_ | give me few mins | 01:09 |
| mfranc213 | np | 01:10 |
| *** yamamoto_ has quit IRC | 01:31 | |
| *** yamamoto_ has joined #openstack-fwaas | 01:38 | |
| *** vishwanathj has quit IRC | 01:59 | |
| *** chandanc_ has joined #openstack-fwaas | 02:04 | |
| *** chandanc_ has quit IRC | 02:15 | |
| *** diogogmt has quit IRC | 02:19 | |
| *** SridarK_ has quit IRC | 02:31 | |
| *** yamamoto_ has quit IRC | 02:48 | |
| *** yamamoto_ has joined #openstack-fwaas | 03:02 | |
| mfranc213 | i am fading and need to go to sleep. i tend to wake up very early and i will get back online when i wake up. | 03:13 |
| *** mickeys has quit IRC | 03:13 | |
| njohnston | sleep well mfranc213! I am back on now | 03:14 |
| *** yamamoto_ has quit IRC | 03:22 | |
| *** yamamoto_ has joined #openstack-fwaas | 03:22 | |
| *** yamamoto_ has quit IRC | 03:22 | |
| *** yamamoto_ has joined #openstack-fwaas | 03:25 | |
| *** yamamoto_ has quit IRC | 03:29 | |
| *** diogogmt has joined #openstack-fwaas | 03:44 | |
| *** chandanc_ has joined #openstack-fwaas | 03:47 | |
| chandanc_ | Hello All | 03:47 |
| yushiro | hi | 03:47 |
| chandanc_ | hello yushiro | 03:48 |
| *** SridarK has joined #openstack-fwaas | 03:53 | |
| chandanc_ | Are we going to use only RPC calls ? | 03:54 |
| chandanc_ | Hello SridarK | 03:54 |
| SridarK | chandanc_: Hi | 03:54 |
| chandanc_ | Just going through the diffs | 03:55 |
| SridarK | ok - i think something minor is missing | 03:55 |
| chandanc_ | which one should i take | 03:55 |
| SridarK | hold off | 03:56 |
| chandanc_ | ok | 03:56 |
| SridarK | i dont think either work properly | 03:56 |
| chandanc_ | ok | 03:56 |
| SridarK | i think we can try to get this to quick resolution | 03:56 |
| SridarK | may be we can try to wrap up the mtg quickly and debug this | 03:56 |
| chandanc_ | ok | 03:57 |
| *** padkrish has joined #openstack-fwaas | 04:00 | |
| *** SarathMekala has joined #openstack-fwaas | 04:02 | |
| *** njohnsto_ has joined #openstack-fwaas | 04:16 | |
| *** xdcc has joined #openstack-fwaas | 04:21 | |
| *** yamamoto_ has joined #openstack-fwaas | 04:29 | |
| yushiro | hi | 04:33 |
| SridarK | hi all | 04:33 |
| chandanc_ | Hello All | 04:33 |
| yushiro | folks, I'll continue to fix UTs for CLI patch. | 04:33 |
| chandanc_ | sure yushiro | 04:33 |
| chandanc_ | give me a ping when ready | 04:33 |
| SridarK | yushiro: yes makes sense - we will not disturb u | 04:33 |
| chandanc_ | I will test | 04:33 |
| SridarK | njohnsto_: we will continue the debugging and update thru email | 04:34 |
| yushiro | I'm sorry for slow speed.. If you need some help, please ping me at any time! | 04:34 |
| SridarK | yushiro: no worries - many thx | 04:34 |
| SridarK | chandanc_: are u ok for a call - easier to share screen | 04:34 |
| chandanc_ | yes, was about to ask for a call | 04:35 |
| chandanc_ | ya | 04:35 |
| njohnston | SridarK: Sounds good. Is there anything in particular I can do to help now, or should I focus on being ready in the morning to take the torch from wherever you and chandanc_ have left things to keep it moving? | 04:35 |
| chandanc_ | sounds good njohnston | 04:35 |
| SridarK | njohnston: it seems that i am missing some piece - but we will continue - hopefully if we get this figured out - u can push the changes | 04:36 |
| SridarK | we can sync over email or early am | 04:36 |
| njohnston | SridarK: sounds good. I will leave myself signed in and will read the logs as soon as I wake up. | 04:38 |
| njohnston | chandanc_ SridarK: Thank you very much, ttyl. | 04:39 |
| SridarK | njohnston: GN | 04:39 |
| chandanc_ | GN njohnston | 04:39 |
| *** padkrish has quit IRC | 04:40 | |
| *** padkrish has joined #openstack-fwaas | 04:41 | |
| *** padkrish has quit IRC | 04:41 | |
| *** njohnsto_ has quit IRC | 04:42 | |
| *** diogogmt has quit IRC | 05:17 | |
| *** xdcc has quit IRC | 05:23 | |
| *** chandanc_ has quit IRC | 08:15 | |
| *** mickeys has joined #openstack-fwaas | 08:20 | |
| *** mickeys has quit IRC | 08:25 | |
| *** chandanc_ has joined #openstack-fwaas | 08:45 | |
| *** yamamoto_ has quit IRC | 10:01 | |
| mfranc213 | hello. | 10:08 |
| mfranc213 | looking at chandan's ps now. | 10:08 |
| mfranc213 | i see what he did. very nice. | 10:10 |
| mfranc213 | i'm starting to look at the jenkins 3.4 failures on the plugin now. | 10:25 |
| njohnston | the pep8 failure is easy to fix | 10:32 |
| njohnston | but the failure for test neutron_fwaas.tests.unit.services.firewall.test_fwaas_plugin_v2.TestFirewallPluginBasev2.test_update_firewall_group_with_new_ports_no_polcy | 10:33 |
| njohnston | that one I don't see how it is happening | 10:33 |
| mfranc213 | me neither :( | 10:33 |
| mfranc213 | still looking | 10:33 |
| njohnston | is that the one you're talking about, that has this midway through the traceback? | 10:33 |
| mfranc213 | good morning nate | 10:33 |
| njohnston | File "neutron_fwaas/services/firewall/fwaas_plugin_v2.py", line 230, in create_firewall_group | 10:33 |
| njohnston | fwg_ports) | 10:33 |
| njohnston | good morning | 10:33 |
| mfranc213 | new ports no policy... | 10:39 |
| mfranc213 | (just thinking out loud) | 10:39 |
| SarathMekala | Hi all.. I was looking at Yushiro's patch https://review.openstack.org/#/c/355755/2/etc/policy.json,unified | 10:40 |
| SarathMekala | I have one question in this regards | 10:40 |
| njohnston | mfranc213: I wonder if the issue is that there was no argument to FirewallGroupPortInvalid https://review.openstack.org/#/c/355576/23/neutron_fwaas/extensions/firewall_v2.py | 10:40 |
| mfranc213 | njohnston: the key error must be in fwg_ports | 10:40 |
| mfranc213 | njohnston: looking at what you just said | 10:41 |
| SarathMekala | In V2 we have renamed shared --> public, so in the policy.json all the existing references to shared_** should be renamed to public_** right? | 10:41 |
| mfranc213 | njohnston: i agree that that's a problem, but i believe it's a separate problem. | 10:42 |
| mfranc213 | the error is a key error | 10:42 |
| mfranc213 | with no policy there's no port_id in the fwg dict | 10:43 |
| mfranc213 | no 'port_id' dict | 10:43 |
| mfranc213 | in fwg_ports dict i should say | 10:43 |
| yushiro | SarathMekala: no, if renames from 'shared' to 'public', fwaas v1 doesn't work at policy check. | 10:43 |
| mfranc213 | but i'm not sure where that is being checked actually. | 10:44 |
| SarathMekala | ok.. so do we have corresponding public_** methods for all the existing commands? | 10:44 |
| mfranc213 | njohnston: i'm going to talk to you privately to not muddle SarathMekala's posts | 10:45 |
| *** mestery has quit IRC | 10:46 | |
| * njohnston thinks SarathMekala brings up an interesting point, but is not sure of the answer himself | 10:46 | |
| SarathMekala | yushiro: I think "create_firewall_policy:shared": "rule:admin_or_owner", should have a corresponding "create_firewall_policy:public": "rule:admin_or_owner", | 10:47 |
| yushiro | SarathMekala: ah, Yes. You're right. | 10:49 |
| yushiro | SarathMekala: I was missing your point. | 10:50 |
| SarathMekala | yushiro : should I leave a comment on the review or do you think its a bit risky at this time | 10:50 |
| yushiro | SarathMekala: hmm, yes it's difficult timing ... let me see.. | 10:51 |
| SarathMekala | yushiro, : one more thing that I have noticed.. there is no public definition for firewall_rule even in V1 | 10:52 |
| SarathMekala | but I remember that a rule can be tagged as public | 10:52 |
| yushiro | SarathMekala: Yes. v1 is also missing 'shared' policy definition. | 10:53 |
| SarathMekala | yeah.. do you want to add it to V1 and V2 now? | 10:54 |
| yushiro | SarathMekala: No I don't. We shuould focus on V2 policy. Regarding v1, we should post as an another patch. | 10:55 |
| SarathMekala | Ok.. Can I go ahead and post the patch for V1. I need to get my hands wet here :) | 10:55 |
| yushiro | SarathMekala: OK. Thanks. BTW, your previous question "ok.. so do we have corresponding public_** methods for all the existing commands?" | 10:56 |
| yushiro | SarathMekala: You mean that validation method on fwaas v2? | 10:57 |
| yushiro | SarathMekala: quick question. | 10:58 |
| SarathMekala | yushiro : your previous answer was good as I see that V1 and V2 both use the same policy.json file | 10:59 |
| SarathMekala | yushiro : so V1 will have shared_** aliases and V2 will public_** aliases | 11:00 |
| SarathMekala | yushiro : but there seems to be one target (policy) that mentioned that needs a corresponding target in V2 | 11:01 |
| yushiro | SarathMekala: OK, I understand. Thank you. | 11:01 |
| SarathMekala | yushiro : need to catch my cab.. will login later tonight.. | 11:01 |
| *** chandanc_ has quit IRC | 11:02 | |
| *** SarathMekala has quit IRC | 11:06 | |
| *** chandanc_ has joined #openstack-fwaas | 11:12 | |
| *** SarathMekala has joined #openstack-fwaas | 11:15 | |
| mfranc213 | ihar just gave a million comments on the l3 agent extension api object. i'm looking these over now. https://review.openstack.org/#/c/357503/ | 11:36 |
| mfranc213 | nate and i are still looking into the plugin UT failures. it's producing a {"NeutronError": {"type": "FirewallGroupPortInvalid", "detail": "", "message": "Firewall Group Port 2aba5c36-38cc-42dc-a68b-996f374794d3 is invalid"}} on the test for new ports no policy | 11:37 |
| *** vishwanathj has joined #openstack-fwaas | 11:44 | |
| mfranc213 | ihar's suggestions will require changes to the agent extension. we'll need to get coordinated on who's doing what. right now, i'm drafting responses to ihar's comments in gerrit. | 11:47 |
| mfranc213 | which means that i'm not looking at the moment at the plugin UT failures... | 11:56 |
| *** yamamoto has joined #openstack-fwaas | 11:59 | |
| *** yamamoto_ has joined #openstack-fwaas | 11:59 | |
| *** yamamoto has quit IRC | 12:03 | |
| *** chandanc_ has quit IRC | 12:16 | |
| *** chandanc_ has joined #openstack-fwaas | 12:32 | |
| chandanc_ | Hello all | 12:36 |
| chandanc_ | i am currently fixing the UT | 12:40 |
| chandanc_ | can someone help me with running the UT as it needs to have the neutron patch in venv | 12:40 |
| *** chandanc_ has quit IRC | 12:41 | |
| *** chandanc_ has joined #openstack-fwaas | 12:41 | |
| chandanc_ | Hello mfranc213 , njohnston | 12:42 |
| njohnston | chandanc_: The easiest thing to do is to do: cd .tox/py27/src/neutron && git review -d 357503 && cd ../../../.. && tox -e py27 | 12:42 |
| chandanc_ | thanks a lot, let me try that | 12:43 |
| mfranc213 | chandanc_ i had to do a git review -s first in that .tox subfolder... | 12:45 |
| chandanc_ | yes i realized that | 12:45 |
| njohnston | mfranc213: Really? Interesting, it worked without that for me | 12:45 |
| mfranc213 | okay, good, that's good to know. | 12:45 |
| *** mestery has joined #openstack-fwaas | 12:52 | |
| *** yamamoto_ has quit IRC | 13:20 | |
| njohnston | chandanc_: How are you doing with the UT fixes for https://review.openstack.org/#/c/355576/ ? | 13:41 |
| chandanc_ | mostly done, | 13:42 |
| njohnston | Excellent! | 13:42 |
| chandanc_ | just doing a final check and i have 6 more :( | 13:42 |
| chandanc_ | njohnston, after discussing with SridarK , i introduced a validation for port owner in the plugin, it is causing some test failure. Need help to fix those | 13:43 |
| chandanc_ | can you help ? | 13:43 |
| chandanc_ | the last 6 UT are failing due to this | 13:44 |
| njohnston | Sure thing - why don't you push your changes to gerrit, and I can pull it to run in mye nvironment? | 13:46 |
| *** _SarathMekala_ has joined #openstack-fwaas | 13:47 | |
| *** SarathMekala_ has joined #openstack-fwaas | 13:47 | |
| *** SarathMekala_ has quit IRC | 13:48 | |
| *** _SarathMekala_ has quit IRC | 13:48 | |
| chandanc_ | sure | 13:50 |
| njohnston | chandanc_: Thanks! | 13:50 |
| chandanc_ | let me push it | 13:50 |
| chandanc_ | ok i just pushed it | 13:56 |
| njohnston | Thanks, already testing | 13:56 |
| njohnston | :-) | 13:56 |
| chandanc_ | thanks :) | 13:57 |
| njohnston | In my environment, `tox -e py27` passed all tests | 13:57 |
| chandanc_ | ya it passes, as i commented ou the validation | 13:57 |
| njohnston | Oh, what needs to be uncommented? | 13:57 |
| njohnston | to exercise the issue we need to fix, I mean | 13:58 |
| chandanc_ | https://review.openstack.org/#/c/355576/24/neutron_fwaas/services/firewall/fwaas_plugin_v2.py L208 | 13:58 |
| chandanc_ | basically check to validate for router port | 13:58 |
| njohnston | OK, uncommented and testing | 13:59 |
| chandanc_ | also both njohnston and mfranc213 please review the test related changes | 13:59 |
| njohnston | Will do | 13:59 |
| mfranc213 | i will. thank you chandanc_! | 13:59 |
| *** _SarathMekala_ has joined #openstack-fwaas | 14:03 | |
| njohnston | OK, uncommented and testing; I see that the web call returns this error, which makes sense for a validation fail: {"NeutronError": {"type": "FirewallGroupPortInvalid", "message": "Firewall Group Port 13a1a781-276f-4d17-8df3-275526424219 is invalid", "detail": ""}} | 14:05 |
| chandanc_ | ya | 14:05 |
| chandanc_ | actually the validation tries to access the port['device_owner'] | 14:05 |
| chandanc_ | this may not be populated in the test data | 14:06 |
| njohnston | right, so right before that I checked the value of port_db['device_owner'], and it is indeed set to "network:router_interface" | 14:06 |
| chandanc_ | yes | 14:06 |
| njohnston | so the logic is performing as instructed | 14:06 |
| chandanc_ | ya | 14:06 |
| *** yamamoto has joined #openstack-fwaas | 14:07 | |
| njohnston | So the next question is, why is a router interface in the firewall group? | 14:07 |
| njohnston | Looking at the test_update_firewall_group_with_new_ports_status_pending test, for example, I see router interfaces getting created - but I see no interfaces that are not routers getting created. | 14:08 |
| chandanc_ | this is to make sure that only l3 ports are added to the FWG, as the current implementation is supporting only l3 | 14:08 |
| chandanc_ | but is the device owner set for those ports ? | 14:09 |
| chandanc_ | can you give me the code location | 14:10 |
| chandanc_ | I can also look in parallel | 14:10 |
| njohnston | well, in this case routers == l3, so when we do "self._router_interface_action('add', ..." we're always going to have the port owned by 'network:router_interface' | 14:10 |
| njohnston | neutron_fwaas/tests/unit/services/firewall/test_fwaas_plugin_v2.py line 384 | 14:10 |
| chandanc_ | ok | 14:11 |
| njohnston | so if the l3 agent is managing a port, it is almost certaintly going to be a router interface, right? otherwise it'd be a port being managed by the L2 side of the code | 14:12 |
| chandanc_ | in that case, i am not sure why the tests are failing :(, i could make sure that the error is coming due to this validation | 14:12 |
| njohnston | It seems like the validation logic is opposite, like it should say if port_db['device_owner'] != "network:router_interface": raise fw_ext.FirewallGroupPortInvalid(port_id=port_id) | 14:12 |
| njohnston | change of '==' to '!=' | 14:12 |
| chandanc_ | :) thanks a lot | 14:13 |
| chandanc_ | sorry, i missed it | 14:14 |
| njohnston | changing that causes all tests to pass | 14:14 |
| njohnston | :-) | 14:14 |
| chandanc_ | cool | 14:14 |
| njohnston | OK, feel free to push that fix, and then once mfranc213 is done with 357503 I can change 355576 to adjust to her changes | 14:15 |
| SridarK | Hi All | 14:15 |
| njohnston | Hi SridarK! | 14:15 |
| chandanc_ | I need to be away for 1/2 hr , then will pickup the functional/integ testing | 14:15 |
| chandanc_ | Hello SridarK | 14:15 |
| SridarK | sorry just catchin up | 14:15 |
| chandanc_ | SridarK, i could not proceed with the integ test as there were UT failure on the last PS | 14:17 |
| chandanc_ | i will get to it in some time | 14:17 |
| njohnston | SridarK: So ihrachys had some feedback for 357503, mfranc213 is working it; it will require some adjustment to 355576 because a method name changed etc. And chandanc_ just pushed the last UT fix for 355576. | 14:18 |
| *** xdcc has joined #openstack-fwaas | 14:18 | |
| *** mickeys has joined #openstack-fwaas | 14:18 | |
| *** chandanc_ is now known as chandanc_afk_30m | 14:20 | |
| mfranc213 | chandanc_, njohnston, SridarK 355576 will need a couple changes to coordinate with 357503. | 14:20 |
| *** mickeys has quit IRC | 14:20 | |
| SridarK | ok just caught up, sorry was a long night and got up later than usual | 14:20 |
| SridarK | njohnston: i think u have the issue on the UT failures nailed down | 14:21 |
| njohnston | SridarK: Thanks! | 14:21 |
| SridarK | and chandanc has fixed the PS | 14:21 |
| *** SarathMekala_ has joined #openstack-fwaas | 14:24 | |
| *** SridarK has quit IRC | 14:26 | |
| *** _SarathMekala_ has quit IRC | 14:27 | |
| *** SridarK has joined #openstack-fwaas | 14:27 | |
| *** chandanc_afk_30m has quit IRC | 14:28 | |
| *** yamamoto has quit IRC | 14:29 | |
| *** yamamoto has joined #openstack-fwaas | 14:29 | |
| njohnston | I just pushed a new PS for 355576 that changes the method names to match Margaret's changes in 357503. | 14:34 |
| mfranc213 | i'm finishing up final ihar changes for 357503... | 14:34 |
| njohnston | mfranc213: router_in_ns -> router_in_namespace and get_router_from_port_id -> get_router_hosting_port; are there any other changes that 355576 needs to account for? | 14:35 |
| mfranc213 | njohnston: thinking about this | 14:52 |
| *** chandanc_afk_30m has joined #openstack-fwaas | 14:52 | |
| *** chandanc_afk_30m is now known as chandanc_ | 14:53 | |
| mfranc213 | njohnston: it's router_in_ns -> is_router_in_namespace. other than those 2 method name changes i can't think of anything else | 15:00 |
| chandanc_ | Hello mfranc213 , njohnston | 15:13 |
| mfranc213 | hello chandanc_ | 15:13 |
| chandanc_ | i think we may need to add some entry points to the setup.cfg | 15:13 |
| *** diogogmt has joined #openstack-fwaas | 15:14 | |
| chandanc_ | should i send you a paste url ? | 15:14 |
| mfranc213 | yes please | 15:14 |
| chandanc_ | http://paste.openstack.org/show/565212/ | 15:14 |
| chandanc_ | please have a look | 15:15 |
| chandanc_ | you can always adjust the alias names | 15:15 |
| njohnston | added setup.cfg entries to https://review.openstack.org/355576 | 15:18 |
| SridarK | ok perfect on the setup.cfg | 15:18 |
| mfranc213 | pushed hopefully last PS for https://review.openstack.org/357503 | 15:19 |
| SridarK | mfranc213: :-) | 15:19 |
| mfranc213 | fingers crossed | 15:19 |
| mfranc213 | i keep making stupid mistakes today | 15:20 |
| SridarK | sigh no worries - i am on that boat too | 15:20 |
| SridarK | mfranc213: thx for getting all those comments addressed quickly | 15:21 |
| njohnston | We're all a little brain-fried %-) | 15:21 |
| SridarK | +1 | 15:21 |
| SridarK | now with the setup.cfg entries does that do all the magic for us to stack and start up with everything setup ? | 15:22 |
| njohnston | I *think* so | 15:22 |
| njohnston | as long as we do the python setup.py develop | 15:22 |
| chandanc_ | ya, that should be all we need, but we will need the ini file configs | 15:22 |
| SridarK | ok - something chandanc_ and i were also discussing | 15:22 |
| SridarK | l3_agent.ini ? | 15:23 |
| chandanc_ | ya | 15:23 |
| SridarK | adding the ext fwaas ? | 15:23 |
| chandanc_ | let me share mine | 15:23 |
| njohnston | I wish there was a way to put something in etc/oslo-config-generator that would mod l3_agent.ini | 15:23 |
| mfranc213 | biab | 15:24 |
| SridarK | should we try to get this in to 355576 or handle it as a bug | 15:24 |
| SridarK | also do we want the devstack script to set the plugin to v2 ? | 15:24 |
| njohnston | I think these changes are orthogonal to 355576 | 15:25 |
| SridarK | should we just keep v1 as default and have a doc entry to enable v2 | 15:25 |
| SridarK | njohnston: ok agreed | 15:26 |
| chandanc_ | http://paste.openstack.org/show/565216/ | 15:26 |
| *** yamamoto has quit IRC | 15:26 | |
| SridarK | we can probab add this as part of the q-fwaas tag - the processing for that | 15:27 |
| SridarK | but i think there was some issue in us being able to define another tag due to the L3 Agent interaction | 15:28 |
| njohnston | SridarK: you mean in the devstack plugin? | 15:29 |
| SridarK | njohnston: yes | 15:29 |
| SridarK | but if we want to keep v1 as default | 15:29 |
| SridarK | and turn on v2 that may be good too | 15:29 |
| SridarK | *turn on v2 if the user desires | 15:30 |
| SridarK | i have to drive in to work for an important meeting - so will be offline for a bit, but will log back in as soon as i am at work | 15:30 |
| SridarK | chandanc_: will u be able to start a sanity test on ur env ? | 15:30 |
| njohnston | SridarK: Drive safe! | 15:30 |
| SridarK | :-) | 15:31 |
| chandanc_ | sure will start | 15:31 |
| chandanc_ | I was planning to pull in the CLI | 15:31 |
| SridarK | ok will sync up in abt an hour | 15:31 |
| chandanc_ | ok sure | 15:31 |
| SridarK | i think yushiro is probab nearing the final stage on this | 15:32 |
| chandanc_ | I hope so | 15:32 |
| yushiro | SridarK: yes.. | 15:32 |
| SridarK | ok great | 15:32 |
| SridarK | be back soon | 15:32 |
| yushiro | I finished firewall_group and firewall_policy. the last is firewall_rule. | 15:33 |
| *** _SarathMekala_ has joined #openstack-fwaas | 15:35 | |
| mfranc213 | great yushiro! | 15:36 |
| *** SridarK has quit IRC | 15:37 | |
| *** SarathMekala_ has quit IRC | 15:39 | |
| chandanc_ | yushiro, just sent you a mail, some issue with CLI | 15:43 |
| yushiro | chandanc_: Thanks! but I'll push it at once and reflect your feedback! | 15:43 |
| chandanc_ | ok | 15:44 |
| chandanc_ | sure, more like a FYI | 15:44 |
| yushiro | chandanc_: If you'd like to through fwaas v2 API more easily, I'll send you a short term solution. | 15:47 |
| yushiro | maybe 3 month ago, I made ruby script to though REST API. | 15:48 |
| yushiro | Can I send to you? | 15:48 |
| chandanc_ | I have some shell scripts :) | 15:48 |
| yushiro | OK. | 15:49 |
| chandanc_ | using curl | 15:49 |
| *** SarathMekala has quit IRC | 15:53 | |
| *** _SarathMekala_ has quit IRC | 15:55 | |
| njohnston | bbl, meeting my father for lunch | 15:57 |
| mfranc213 | bye nate | 15:57 |
| chandanc_ | mfranc213, which is the neutron patch that we need to apply for the last adjustment you did on 355576 | 16:01 |
| mfranc213 | 1 sec chandanc_ | 16:04 |
| mfranc213 | 27 is the last one, and it is the one nate pushed in order to work with PS 18 of https://review.openstack.org/#/c/357503/ | 16:05 |
| chandanc_ | ok thanks mfranc213 | 16:06 |
| mfranc213 | please let me know if there's anything else you need. | 16:06 |
| chandanc_ | mfranc213, although i could do one end to end fwg create and populate iptables | 16:10 |
| chandanc_ | there is an issue | 16:10 |
| chandanc_ | 2016-08-31 15:51:31.245 TRACE neutron_fwaas.services.firewall.agents.l3reference.firewall_l3_agent_v2 AttributeError: 'FirewallPluginV2' object has no attribute '_get_ports_for_firewall_group' | 16:10 |
| chandanc_ | this happens when i restarted the l3 agent | 16:10 |
| chandanc_ | mfranc213, here is the full trace on plugin side http://paste.openstack.org/show/565234/ | 16:12 |
| chandanc_ | afk for dinner | 16:12 |
| *** chandanc_ is now known as chandanc_afk | 16:13 | |
| mfranc213 | thank you. looking now. | 16:15 |
| *** yamamoto has joined #openstack-fwaas | 16:17 | |
| *** mickeys has joined #openstack-fwaas | 16:29 | |
| mfranc213 | chandanc_ the db layer has changed: https://github.com/openstack/neutron-fwaas/blob/master/neutron_fwaas/db/firewall/v2/firewall_db_v2.py#L694 | 16:30 |
| *** yamamoto has quit IRC | 16:30 | |
| mfranc213 | the method name is now _get_ports_in_firewall_group | 16:31 |
| mfranc213 | dealing with the fix now | 16:31 |
| mfranc213 | not quite sure how the db patch made it in. SridarK am i right that gate allowed this problem in? see link above compared with https://github.com/openstack/neutron-fwaas/blob/master/neutron_fwaas/services/firewall/fwaas_plugin_v2.py#L114 | 16:35 |
| mfranc213 | (don't trust myself) | 16:36 |
| mfranc213 | okay, SridarK isn't here at the moment. looking further as to which direction to change things | 16:38 |
| *** SridarK has joined #openstack-fwaas | 16:41 | |
| *** yamamoto has joined #openstack-fwaas | 16:47 | |
| *** chandanc_afk has quit IRC | 16:49 | |
| mfranc213 | chandanc_ https://review.openstack.org/#/c/355576 has the correct method name in the plugin but i'm trying to figure out why your version of the plugin has the wrong method name, since i know you're using 355576 | 16:52 |
| SridarK | mfranc213: chandanc_: which method is this ? sorry just got back online | 16:53 |
| mfranc213 | _get_ports_for_firewall_group (in chandanc_ 's plugin) verses _get_ports_in_firewall_group in the version of the plugin in 355576 | 16:54 |
| mfranc213 | the db has "def _get_ports_in_firewall_group" | 16:55 |
| mfranc213 | but i don't see where https://review.openstack.org/#/c/355576 changed that part of the plugin | 16:55 |
| SridarK | hmm | 16:55 |
| mfranc213 | so i'm trying to figure out how chandanc_ has the wrong version of the plugin, and how he got it! | 16:56 |
| SridarK | i am wondering if this was an issue that slipped thru | 16:59 |
| mfranc213 | it did, yes. | 17:00 |
| mfranc213 | but some patch fixed it, but it's not https://review.openstack.org/#/c/355576, which *contains* the fixed code but did not fix it itself | 17:01 |
| SridarK | i think it is a typo in _get_ports_for_firewall_group | 17:01 |
| mfranc213 | yes, i agree. | 17:02 |
| mfranc213 | i think _get_ports_in_firewall_group | 17:02 |
| mfranc213 | makes more sense | 17:02 |
| SridarK | agreed | 17:02 |
| mfranc213 | but where/how is chandanc_ getting the old code? | 17:02 |
| SridarK | https://github.com/openstack/neutron-fwaas/blob/master/neutron_fwaas/services/firewall/fwaas_plugin_v2.py#L100 | 17:02 |
| SridarK | i think we need to fix the typo here | 17:03 |
| mfranc213 | no, no, wait SridarK | 17:03 |
| mfranc213 | i don't think i'm making myself clear. | 17:03 |
| mfranc213 | let me link something to you. one sec | 17:03 |
| SridarK | oh sorry go ahead | 17:03 |
| SridarK | sorry i am in a mtg so i am a bit distracted | 17:04 |
| mfranc213 | https://review.openstack.org/#/c/355576/27/neutron_fwaas/services/firewall/fwaas_plugin_v2.py@169 | 17:04 |
| SridarK | and that looks to be the right thing | 17:05 |
| mfranc213 | yes, but chandanc_ 's environment has the wrong thing. how? | 17:05 |
| SridarK | oh i wonder if he might have fixed something | 17:06 |
| SridarK | lets wait for him to get back online | 17:06 |
| mfranc213 | could there be another patchset that he has overlaid on the top of 355576, where the other patchset reverts that line of code? | 17:07 |
| SridarK | not sure, but this is an opportunity to fix it | 17:07 |
| SridarK | as we get the final PS pushed for 35576 | 17:08 |
| mfranc213 | right | 17:08 |
| SridarK | *355576 | 17:08 |
| mfranc213 | yep :) | 17:08 |
| mfranc213 | but SridarK, 35576 did not fix that--if you compare latest PS to its base, that line was already fixed. | 17:09 |
| mfranc213 | so what fixed that line in the plugin, do you know? | 17:09 |
| njohnston | back | 17:11 |
| mfranc213 | hi nate | 17:12 |
| SridarK | mfranc213: i am wondering if L#169 was not a problem | 17:12 |
| SridarK | it was as is | 17:12 |
| SridarK | i think the method around L#100 had the snafu | 17:13 |
| mfranc213 | oh, ok. | 17:13 |
| mfranc213 | no, wait, i retract that "oh, ok" | 17:14 |
| mfranc213 | :) | 17:14 |
| mfranc213 | i agree with you that that line 100 has the problem. | 17:14 |
| SridarK | ok good | 17:14 |
| mfranc213 | i guess my questions are: 1) where and how did it get fixed? 2) how come chandanc_ 's environment doesn't have the fix, given that he's using 355576 ? | 17:15 |
| SridarK | that i am also wondering | 17:18 |
| *** chandanc_afk has joined #openstack-fwaas | 17:23 | |
| SridarK | chandanc_afk: hi | 17:31 |
| SridarK | mfranc213: i sent an email out on related to what u found | 17:32 |
| chandanc_afk | Hello All | 17:32 |
| SridarK | pls check if that diff makes sense and we will pick that up on the next rev | 17:32 |
| *** chandanc_afk is now known as chandanc_ | 17:32 | |
| mfranc213 | SridarK reading it now! | 17:33 |
| *** yamamoto has quit IRC | 17:34 | |
| chandanc_ | Hello SridarK | 17:35 |
| chandanc_ | just catching up | 17:35 |
| *** SridarK has quit IRC | 17:37 | |
| *** SridarK has joined #openstack-fwaas | 17:39 | |
| SridarK | mfranc213: my version of 355576 did not have the change in the method around L#100 | 17:40 |
| SridarK | i pulled it late last night | 17:40 |
| SridarK | chandanc_: did u do a sanity with the latest PS | 17:41 |
| njohnston | https://github.com/openstack/neutron-fwaas/blame/master/neutron_fwaas/services/firewall/fwaas_plugin_v2.py#L107 has the 'for' not the 'in' syntax and says the v2 driver plugin had it | 17:41 |
| SridarK | njohnston: yes that needs to be fixed | 17:42 |
| njohnston | pushing a new PS for 355576 that has the fix now | 17:42 |
| mfranc213 | it already has the fix!!! | 17:42 |
| mfranc213 | is already has the fix | 17:43 |
| mfranc213 | oh my goodnes! | 17:43 |
| njohnston | mfranc213: I just did 'git review -d 355576' and it didn't have the fix | 17:43 |
| SridarK | mfranc213: that is what i think too | 17:43 |
| mfranc213 | https://review.openstack.org/#/c/355576/27/neutron_fwaas/services/firewall/fwaas_plugin_v2.py@169 | 17:43 |
| mfranc213 | is it wrong in another place? | 17:43 |
| chandanc_ | L110 | 17:44 |
| mfranc213 | ah | 17:44 |
| mfranc213 | 110 | 17:44 |
| chandanc_ | L114 | 17:44 |
| njohnston | and L114 | 17:44 |
| mfranc213 | ok, i'm sorry | 17:44 |
| mfranc213 | i thought i was going crazy | 17:44 |
| mfranc213 | and now i know that i am | 17:44 |
| SridarK | mfranc213: relax - u are fine | 17:44 |
| chandanc_ | same here :) | 17:44 |
| njohnston | new PS piushed for 355576 | 17:44 |
| mfranc213 | thank you njohnston | 17:44 |
| SridarK | njohnston: thx | 17:45 |
| njohnston | my pleasure | 17:45 |
| chandanc_ | I could do, one end-to-end fwg-create | 17:46 |
| chandanc_ | all seemed good except the one issue | 17:46 |
| chandanc_ | with the new PS we should be good | 17:46 |
| njohnston | fantastic, chandanc_ | 17:48 |
| SridarK | chandanc_: the issue that u hit was the change njohnston just pushed ? | 17:48 |
| SridarK | ie, problem with get_firewall_groups_for_project | 17:49 |
| chandanc_ | http://paste.openstack.org/show/565234/ yes | 17:49 |
| chandanc_ | it has the trace | 17:49 |
| chandanc_ | L114 | 17:49 |
| chandanc_ | njohnston, yes looks like we are mostly there | 17:51 |
| SridarK | chandanc_: ok perfect | 17:51 |
| mfranc213 | jeopardy music | 17:54 |
| mfranc213 | who wants to go out for a beer with me once all this merges? | 17:55 |
| chandanc_ | :) too bad for me | 17:55 |
| SridarK | virtual beer :-) | 17:55 |
| chandanc_ | thats one thing not for the cloud | 17:56 |
| mfranc213 | :) | 17:56 |
| SridarK | :-) | 17:57 |
| xgerman | Sangria in Barca? | 18:01 |
| xgerman | too bad I don’t have funding | 18:01 |
| njohnston | xgerman: :-( | 18:02 |
| *** diogogmt has quit IRC | 18:02 | |
| SridarK | xgerman: sigh :-( | 18:02 |
| xgerman | no worries, maybe next time will be better ;-) | 18:03 |
| SridarK | lets hope for that | 18:06 |
| mfranc213 | does this mess us up or help us: Use model_base from neutron-lib https://review.openstack.org/363845 | 18:09 |
| mfranc213 | (just proposed by Henry Gessau) | 18:09 |
| xgerman | good thing | 18:09 |
| xgerman | neutron_lib is supposed to be the future | 18:10 |
| njohnston | yep. I had a change already to do this, but I abandoned it in favor of Henry's | 18:20 |
| njohnston | Just keeping us up to date with the developments in neutron_lib as of v0.4.0 | 18:21 |
| mfranc213 | cool | 18:21 |
| njohnston | I also made a full list of all the imports fwaas has from neutron proper and passed that along to dougwig for them to use as a punchlist | 18:21 |
| mfranc213 | how will they use that? | 18:23 |
| mfranc213 | (you are always at least 2 steps ahead of me, btw) | 18:23 |
| njohnston | So their charter is to take all those places and move them from neutron to neutron_lib so fwaas won't import from neutron anymore. | 18:23 |
| mfranc213 | ohhh, of course. | 18:24 |
| njohnston | I am hoping to help with that next cycle | 18:24 |
| mfranc213 | nice. | 18:24 |
| njohnston | I have to step into a meeting with my new VP now, back in :30 | 18:24 |
| njohnston | Ideally in the meantime we can get a +2+A from carl_baldwin on https://review.openstack.org/#/c/357503/ - I think that will clear our biggest impediment. | 18:25 |
| mfranc213 | bfn | 18:25 |
| carl_baldwin | njohnston: I have it up, was meaning to check out what was suggested since my last review. | 18:26 |
| mfranc213 | thank you carl_baldwin | 18:26 |
| *** chandanc_ has quit IRC | 18:32 | |
| *** yamamoto has joined #openstack-fwaas | 18:35 | |
| *** mickeys has quit IRC | 18:39 | |
| *** xdcc has quit IRC | 18:40 | |
| *** yamamoto has quit IRC | 18:42 | |
| *** diogogmt has joined #openstack-fwaas | 18:48 | |
| yushiro | Hi, all. | 18:54 |
| yushiro | I just updated CLI patch. UT is good bat more UTs are necessary. I'll add more. | 18:55 |
| yushiro | mfranc213: Thanks for your comment but I'll reflect the next PS. | 18:55 |
| yushiro | So, I'll go home and come back to my office. | 18:55 |
| yushiro | See you 5 hours later! | 18:55 |
| *** diogogmt has quit IRC | 18:57 | |
| mfranc213 | thank you yushiro! | 19:01 |
| SridarK | yushiro:thx - i think we can track the need for more UT as a bug | 19:02 |
| mfranc213 | +2+1 from carl on l3 agent extension api | 19:20 |
| SridarK | yes saw that - that is so great - njohnston: , mfranc213: really nice job in pushing that thru | 19:28 |
| mfranc213 | thank you SridarK! | 19:32 |
| njohnston | thanks to everyone | 19:32 |
| mfranc213 | seconded | 19:33 |
| SridarK | +1 | 19:33 |
| SridarK | i am on 355576, some quick thoughts - sort of stating the obvious | 19:33 |
| SridarK | now that we have uncoupled the inheritance relationship, we have broken v1 | 19:34 |
| SridarK | and the tempest failures | 19:35 |
| njohnston | Understandable - what do we need to do to adapt v1 to the new methodology? | 19:35 |
| SridarK | which is expected | 19:35 |
| mfranc213 | can the delta in https://review.openstack.org/#/c/355576 simply be applied to v1, or is it not that simple? | 19:36 |
| SridarK | the changes could be ccery similar | 19:36 |
| SridarK | *very | 19:36 |
| SridarK | shall we do a quick call ? | 19:36 |
| SridarK | i think it is not that complicated to do this | 19:37 |
| mfranc213 | SridarK what would be the desired timelines for issuing such patches? | 19:37 |
| mfranc213 | yes, i can do a call :) | 19:37 |
| njohnston | me too | 19:37 |
| SridarK | njohnston: are u avail as well ? | 19:37 |
| SridarK | ok got it let me go to a conf room | 19:37 |
| njohnston | thanks SridarK | 19:37 |
| mfranc213 | sounds good | 19:38 |
| *** yushiro has quit IRC | 19:38 | |
| *** mickeys has joined #openstack-fwaas | 19:40 | |
| *** mickeys has quit IRC | 19:45 | |
| *** diogogmt has joined #openstack-fwaas | 20:14 | |
| SridarK | CLI - pep8 was simple, looking at UT | 20:19 |
| SridarK | mfranc213: can u pls take a look here: | 20:26 |
| SridarK | http://logs.openstack.org/82/351582/11/check/gate-python-neutronclient-python27-ubuntu-xenial/881db50/testr_results.html.gz | 20:26 |
| SridarK | it seems the failures are unrelated | 20:26 |
| mfranc213 | yes | 20:26 |
| mfranc213 | SridarK i see four failures all of the sort AttributeError: 'module' object has no attribute 'add_project_domain_option_to_parser' | 20:28 |
| SridarK | yes exactly | 20:28 |
| mfranc213 | are there others that i'm missing? | 20:28 |
| SridarK | no all others pass | 20:28 |
| SridarK | now these are not tests that are part of this PS | 20:29 |
| mfranc213 | should a rebase be done, since the failures are all in osc_utils? | 20:29 |
| njohnston | SridarK: quick question - do we need to change FWaaSL3PluginApi in https://github.com/openstack/neutron-fwaas/blob/master/neutron_fwaas/services/firewall/agents/l3reference/firewall_l3_agent.py#L33 - I think it can stay the same, but want a doublecheck | 20:30 |
| mfranc213 | looking | 20:30 |
| mfranc213 | i do not see a need to change this other than the tenant-->project vernacular. | 20:31 |
| SridarK | njohnston: yes u are correct no change needed there | 20:32 |
| mfranc213 | i see you asked SridarK! | 20:32 |
| SridarK | mfranc213: i wonder too - i am also looking into the code there | 20:33 |
| *** mickeys has joined #openstack-fwaas | 20:41 | |
| njohnston | SridarK: It looks like the v1 file does not have a class like 'L3WithFWaaS' at https://review.openstack.org/#/c/355576/28/neutron_fwaas/services/firewall/agents/l3reference/firewall_l3_agent_v2.py@501 - is this something we need to add for v1? | 20:41 |
| *** mickeys has quit IRC | 20:42 | |
| *** mickeys_ has joined #openstack-fwaas | 20:42 | |
| SridarK | njohnston: yes we will need that, i think it was present in one of the files we are deleting as part of this change | 20:44 |
| njohnston | SridarK: Great, thanks! | 20:45 |
| SridarK | np | 20:45 |
| njohnston | Status update: I just pushed a new PS for https://review.openstack.org/355576 and there are 3 UTs still broken, I haven't even tried to run the tempest tests yet | 21:01 |
| njohnston | It looks like what needs to happen is for the tests to create an agent extension api and pass it to the extension and then it will work. | 21:02 |
| njohnston | The errors for all 3 failures are: | 21:02 |
| njohnston | File "neutron_fwaas/services/firewall/agents/l3reference/firewall_l3_agent.py", line 126, in _get_router_info_list_for_tenant | 21:02 |
| njohnston | self.agent_api.get_routers_in_project(tenant_id) | 21:02 |
| njohnston | AttributeError: 'NoneType' object has no attribute 'get_routers_in_project' | 21:02 |
| njohnston | I think that code can be copied from the v2 tests | 21:03 |
| njohnston | The 3 tests failing are all in neutron_fwaas.tests.unit.services.firewall.agents.l3reference.test_firewall_l3_agent.TestFwaasL3AgentRpcCallback | 21:03 |
| njohnston | SridarK, mfranc213, and *, I am going to head home now, I will be on later | 21:04 |
| mfranc213 | njohnston, thank you. excellent work. i'll look at fixing the final UTs now as you head home. | 21:05 |
| SridarK | njohnston: cool | 21:08 |
| *** mickeys_ has quit IRC | 21:13 | |
| *** mickeys has joined #openstack-fwaas | 21:14 | |
| SridarK | i can get some UT's to pass in changing an import to use a different package (openstackclient) on the CLI patch | 21:30 |
| mfranc213 | SridarK that is most excellent | 21:31 |
| SridarK | i will step away for a bit - be back online in abt 30 or 40 mins | 21:31 |
| mfranc213 | SridarK: bye for now | 21:31 |
| SridarK | i am not so sure on that still - let me clean that up do more verification and share that - we can decide on what next after that | 21:32 |
| SridarK | bye be back soon | 21:32 |
| mfranc213 | SridarK: sounds good. | 21:35 |
| *** SridarK has quit IRC | 21:37 | |
| *** diogogmt has quit IRC | 21:48 | |
| mfranc213 | ping njohnston | 22:16 |
| *** SridarK_ has joined #openstack-fwaas | 22:22 | |
| SridarK_ | njohnston: & mfranc213: on 355576 - we probab want the default to load up v1 so tempest can run as opposed to loading up v2 | 22:26 |
| *** njohnsto_ has joined #openstack-fwaas | 22:57 | |
| *** njohnsto_ has quit IRC | 23:01 | |
| *** njohnsto_ has joined #openstack-fwaas | 23:02 | |
| SridarK_ | njohnsto_: hi so 3 failures on the v1 agent file, | 23:05 |
| SridarK_ | just need to step away for 10 mins or so - will get back | 23:06 |
| njohnsto_ | Ok so those are still an issue then? | 23:09 |
| *** njohnsto_ has quit IRC | 23:20 | |
| SridarK_ | looking | 23:25 |
| *** njohnsto_ has joined #openstack-fwaas | 23:29 | |
| mfranc213 | njohnston: i am working on these | 23:31 |
| mfranc213 | the fixes revealed some other issues. | 23:32 |
| njohnsto_ | Gotcha, thanks! | 23:32 |
| mfranc213 | i can push the PS shortly to have you take a look | 23:32 |
| *** SarathMekala has joined #openstack-fwaas | 23:32 | |
| SridarK_ | mfranc213: ok | 23:32 |
| SridarK_ | mfranc213: neutron_fwaas/services/firewall/agents/l3reference/firewall_l3_agent.py", line 126 | 23:33 |
| mfranc213 | yes, i've changed that | 23:33 |
| SridarK_ | self.agent_api is not getting set for the agent ext to pick up the ... | 23:33 |
| SridarK_ | ah ok cool | 23:34 |
| mfranc213 | yes, i've fixed that part of things. i'm down to one UT failure. | 23:34 |
| *** njohnsto_ has quit IRC | 23:34 | |
| *** njohnsto_ has joined #openstack-fwaas | 23:35 | |
| SridarK_ | one other thing is i wonder if we can actually have it run v1 for tempest | 23:35 |
| mfranc213 | njohnsto_ do you want to take a look? | 23:37 |
| mfranc213 | you'll see what i've done. it has a print in there to show the namespace. the agent_api is being consumed, some other issues that were then revealed have been resolved. | 23:38 |
| mfranc213 | but the one test failure remains and i think fresh eyes would be super. | 23:38 |
| njohnsto_ | Sure, when I get home; on via my iPhone running an errand ATM. | 23:38 |
| SridarK_ | mfranc213: which test ? | 23:38 |
| mfranc213 | i'm going to take a break and will come back a bit later. | 23:39 |
| mfranc213 | SridarK: neutron_fwaas.tests.unit.services.firewall.agents.l3reference.test_firewall_l3_agent.TestFwaasL3AgentRpcCallback.test_get_router_info_list_two_routers_one_without_router_info | 23:39 |
| *** yushiro has joined #openstack-fwaas | 23:39 | |
| SridarK_ | mfranc213: ok let me pull ur PS | 23:39 |
| mfranc213 | thank you sridar | 23:39 |
| yushiro | morning, all I just come to office. | 23:39 |
| yushiro | ^^ I saw IRC log now. Conguras for L3-agent ! | 23:41 |
| njohnsto_ | Thanks, yushiro! | 23:42 |
| *** diogogmt has joined #openstack-fwaas | 23:46 | |
| yushiro | oh... my CLI patch got '-1' from jenkins Thouth UT has succeeded in local. | 23:48 |
| yushiro | I'll fix it now. | 23:48 |
| *** _SarathMekala_ has joined #openstack-fwaas | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!