| *** diogogmt has quit IRC | 00:16 | |
| *** chandanc has joined #openstack-fwaas | 01:00 | |
| *** diogogmt has joined #openstack-fwaas | 01:03 | |
| *** mickeys has quit IRC | 01:13 | |
| *** chandanc has quit IRC | 01:28 | |
| *** chandanc has joined #openstack-fwaas | 02:40 | |
| *** chandanc_ has joined #openstack-fwaas | 02:42 | |
| *** chandanc has quit IRC | 02:45 | |
| *** vishwanathj_ has joined #openstack-fwaas | 02:54 | |
| *** sc68cal has quit IRC | 03:52 | |
| *** sc68cal has joined #openstack-fwaas | 03:52 | |
| *** vishwanathj has quit IRC | 04:20 | |
| *** vishwanathj has joined #openstack-fwaas | 04:21 | |
| *** vishwanathj_ has quit IRC | 05:36 | |
| *** xgerman_ has joined #openstack-fwaas | 05:45 | |
| *** njohnston has quit IRC | 05:48 | |
| *** afranc has quit IRC | 05:48 | |
| *** xgerman has quit IRC | 05:48 | |
| *** mestery has quit IRC | 05:48 | |
| *** njohnston_ has joined #openstack-fwaas | 05:48 | |
| *** afranc has joined #openstack-fwaas | 05:49 | |
| *** xgerman_ is now known as xgerman | 05:49 | |
| *** mestery has joined #openstack-fwaas | 05:51 | |
| *** vishwanathj has quit IRC | 06:15 | |
| *** vishwanathj has joined #openstack-fwaas | 06:15 | |
| *** mickeys has joined #openstack-fwaas | 06:19 | |
| *** mickeys has quit IRC | 06:26 | |
| *** chandanc_ has quit IRC | 07:05 | |
| *** chandanc_ has joined #openstack-fwaas | 08:14 | |
| *** vishwanathj_ has joined #openstack-fwaas | 08:26 | |
| *** chandanc_ has quit IRC | 11:02 | |
| *** vishwanathj_ has quit IRC | 11:06 | |
| *** yamamoto has joined #openstack-fwaas | 12:02 | |
| *** yamamoto has quit IRC | 12:13 | |
| *** yamamoto has joined #openstack-fwaas | 12:15 | |
| *** Trident has joined #openstack-fwaas | 12:16 | |
| *** yamamoto has quit IRC | 12:19 | |
| *** yamamoto has joined #openstack-fwaas | 12:20 | |
| *** yamamoto has quit IRC | 12:29 | |
| *** yamamoto has joined #openstack-fwaas | 13:09 | |
| *** njohnston_ is now known as njohnston | 13:09 | |
| *** chandanc has joined #openstack-fwaas | 13:10 | |
| *** diogogmt has quit IRC | 13:33 | |
| *** diogogmt has joined #openstack-fwaas | 13:34 | |
| *** yamamoto has quit IRC | 13:41 | |
| *** yamamoto has joined #openstack-fwaas | 13:43 | |
| *** yamamoto has quit IRC | 13:43 | |
| *** yamamoto has joined #openstack-fwaas | 13:44 | |
| *** diogogmt has quit IRC | 13:50 | |
| chandanc | Hello Guys | 13:56 |
|---|---|---|
| njohnston | Hello chandanc | 13:56 |
| mfranc213 | hello chandanc and njohnston | 13:56 |
| chandanc | Hey | 13:56 |
| chandanc | just trying to understand, how to start testing margaret's patch | 13:57 |
| mfranc213 | i did unit tests only for the fwaas v2 l3 agent extension. for integration testing i believe that we need to put the following changesets | 13:57 |
| mfranc213 | in place: | 13:57 |
| mfranc213 | the l3 agent extension PS | 13:57 |
| chandanc | ok, | 13:58 |
| mfranc213 | the v2 fwaas plugin | 13:58 |
| mfranc213 | the v2 fwaas driver | 13:58 |
| chandanc | thats my changes | 13:58 |
| chandanc | right ? | 13:58 |
| mfranc213 | yours is the v2 fwaas driver, right? | 13:58 |
| chandanc | yes | 13:58 |
| mfranc213 | then the db piece | 13:59 |
| chandanc | I think we might also need the db | 13:59 |
| chandanc | ya | 13:59 |
| mfranc213 | and it goes on from there | 13:59 |
| chandanc | wow, looks like the complete set :) | 13:59 |
| mfranc213 | because do we need versioned objects? i think not, actually, because at the moment we're not doing the new rpc callback mechanism | 13:59 |
| njohnston | I think these are the ones we were targeting for merge by Friday | 13:59 |
| chandanc | ok | 14:00 |
| mfranc213 | and it would be great to have the cli stuff in order to avoid having to write rest calls | 14:00 |
| mfranc213 | i'm really thinking aloud here: maybe i'm getting stuff wrong. | 14:00 |
| chandanc | hmm, not sure how far the cli stuff is | 14:00 |
| mfranc213 | i'm in the middle of a review. | 14:00 |
| chandanc | no i have the same feeling | 14:00 |
| mfranc213 | it doesn't look all that far off, as far as i can tell. | 14:00 |
| chandanc | ok | 14:01 |
| mfranc213 | neutron's agent extension generalizaation has already been merged. | 14:01 |
| chandanc | ok | 14:01 |
| mfranc213 | and i wrote the l3 agent extension code as totally analogous to what was in place for the l2 side | 14:01 |
| chandanc | let me start pulling the patches | 14:01 |
| mfranc213 | so i'm fairly confident that that one is ready to go | 14:01 |
| mfranc213 | chandanc, does it make sense to make sure we're talking about the same changesets (just to be sure)? | 14:02 |
| chandanc | any chance you have the links for the patches in one place | 14:02 |
| mfranc213 | i mostly do. 1 sec. let me fetch them. | 14:02 |
| mfranc213 | fetch the URLs i should say | 14:02 |
| chandanc | yes, it does | 14:02 |
| chandanc | great, thanks a lot | 14:03 |
| mfranc213 | 1 sec. i haven't fallen off a bridge... | 14:04 |
| chandanc | :) its ok, thats a lot of work saved for me | 14:05 |
| mfranc213 | CLI: https://review.openstack.org/#/c/351582 | 14:05 |
| mfranc213 | REST commands: https://review.openstack.org/#/c/264489 | 14:05 |
| mfranc213 | Plugin: https://review.openstack.org/#/c/267046 | 14:05 |
| mfranc213 | DB: https://review.openstack.org/#/c/311159 | 14:06 |
| chandanc | Thanks a lot | 14:06 |
| mfranc213 | a few more... | 14:06 |
| mfranc213 | L3 agent extension manager: https://review.openstack.org/#/c/339246 | 14:07 |
| mfranc213 | FWaaS agent extension: https://review.openstack.org/#/c/337699 | 14:07 |
| mfranc213 | And yours is the driver ... blagh can't find that one but i'm sure you know it! | 14:08 |
| chandanc | sure, that i will take care of | 14:09 |
| mfranc213 | and the devstack fwaas plugin and the agent extension generalization are already merged. | 14:09 |
| chandanc | ok, cool. Let me get started with | 14:09 |
| mfranc213 | i'm not sure all of these are needed for an integration test | 14:10 |
| mfranc213 | and i'm not sure which of these are ready for an integration test | 14:11 |
| mfranc213 | so those are the two things that you and nate can help with i think | 14:11 |
| chandanc | sure, Let me start pulling in the changes and I will let you know the results | 14:12 |
| mfranc213 | that is excellent | 14:12 |
| mfranc213 | once again you are working very late :( | 14:13 |
| mfranc213 | i'm sorry | 14:13 |
| *** SridarK has joined #openstack-fwaas | 14:14 | |
| chandanc | no worries :) | 14:14 |
| mfranc213 | can you send me the link to the driver changeset? | 14:18 |
| SridarK | mfranc213: chandanc: Hello, so thinks are looking good | 14:18 |
| mfranc213 | hello sridar! | 14:19 |
| chandanc | The driver changeset will be part of your patch | 14:19 |
| SridarK | on the driver piece ? | 14:19 |
| mfranc213 | the driver isn't its own changeset? | 14:19 |
| SridarK | chandanc: from ur pastebin: | 14:19 |
| SridarK | 0 0 neutron-l3-agent-fwaas-defau all qr-c1a76216-8f * ::/0 ::/0 0 0 neutron-l3-agent-fwaas-defau all qr-8a37c725-62 * ::/0 ::/0 0 0 neutron-l3-agent-fwaas-defau all qr-e2d7c3b0-49 * ::/0 ::/0 | 14:19 |
| SridarK | ouch that did not come right | 14:19 |
| SridarK | 0 neutron-l3-agent-fwaas-defau all qr-c1a76216-8f * ::/0 ::/0 | 14:20 |
| SridarK | and 2 other rows below that | 14:20 |
| chandanc | no, I will be merging it with your patch set | 14:20 |
| mfranc213 | wait, does that make sense? the fwaas plugin is its own changeset. | 14:20 |
| chandanc | SridarK, maybe the line number will be better | 14:20 |
| SridarK | chandanc: L# 132-134 | 14:21 |
| njohnston | SridarK: I was wondering if you could comment on the questions I asked here, when you get a chance: https://review.openstack.org/#/c/311159/22/neutron_fwaas/db/migration/alembic_migrations/versions/d6a12e637e28_neutron_fwaas_v2_0.py@60 | 14:22 |
| SridarK | njohnston: yes will do, i can address Yushiro's comment also on the next patch i push up | 14:22 |
| njohnston | Thanks. I am still working on that test failure. | 14:23 |
| chandanc | SridarK: yes i got it, you have a question ? | 14:23 |
| SridarK | njohnston: sorry - i could not look at the UT failure, i ran into an issue testing the policy_update method | 14:23 |
| njohnston | SridarK: no problem :-) | 14:23 |
| SridarK | so have been debugging that, last night i think i made some headway, i need to clean that up and get that out. | 14:24 |
| mfranc213 | chandanc, let me know when you come up for air. i'd like to talk further about the idea of having both driver and agent extension code in a single changeset | 14:24 |
| mfranc213 | i would prefer to have these be separate changesets but maybe i am missing something | 14:24 |
| mfranc213 | maybe njohnston and sridark can chime in on this question also | 14:25 |
| njohnston | Well, we would want the driver to be in use in an L2 context as well as an L3 context, right? | 14:25 |
| chandanc | margaret, sure will ping you | 14:26 |
| SridarK | mfranc213: for the L3, perhaps chandanc can push an update to ur patch | 14:26 |
| chandanc | njohnston: the l3 driver is different from l2 driver | 14:26 |
| SridarK | so we can avoid having another patch in progress | 14:26 |
| *** vishwanathj_ has joined #openstack-fwaas | 14:26 | |
| njohnston | chandanc: Ah, I see. | 14:27 |
| chandanc | l2 driver will be a separate patch | 14:27 |
| mfranc213 | i think it will be more difficult to review a single changeset with two pieces of disparate functionality. i know this topic was discussed earlier; i may be in the minority here. | 14:28 |
| SridarK | mfranc213: since the change is confined to the driver - perhaps it will be easier to avoid another level of dependency for the merge | 14:29 |
| chandanc | I would prefer a separate patch, but looks like we are running out of time for reviews and the changes on the driver looked small enough | 14:30 |
| SridarK | with that in ur patch | 14:30 |
| SridarK | mfranc213: my concern along chandanc: 's comment above | 14:30 |
| mfranc213 | chandanc, sridark: okay. sounds good to me. | 14:31 |
| SridarK | mfranc213: thx for ur understanding | 14:31 |
| chandanc | by the way, in our previous discussion (me and margaret) we came to the conclusion that to test the driver we need the following set of patches | 14:32 |
| chandanc | CLI: https://review.openstack.org/#/c/351582 | 14:32 |
| chandanc | REST commands: https://review.openstack.org/#/c/264489 | 14:32 |
| chandanc | Plugin: https://review.openstack.org/#/c/267046 | 14:32 |
| chandanc | DB: https://review.openstack.org/#/c/311159 | 14:32 |
| chandanc | L3 agent extension manager: https://review.openstack.org/#/c/339246 | 14:32 |
| chandanc | FWaaS agent extension: https://review.openstack.org/#/c/337699 | 14:32 |
| chandanc | njohnston, SridarK any thought | 14:33 |
| SridarK | chandanc: u are correct | 14:33 |
| SridarK | we can avoid the CLI | 14:33 |
| SridarK | let me publish the curl cmds | 14:33 |
| chandanc | i was about to ask that :) | 14:33 |
| SridarK | so u can use that to drive the operations | 14:34 |
| mfranc213 | chandanc https://review.openstack.org/#/c/337699 has two unit tests missing | 14:34 |
| mfranc213 | njohnston has uncovered a scenario in which two developers can overwrite one another's changes if they don't push the PSs up in a particular | 14:34 |
| mfranc213 | way. | 14:34 |
| mfranc213 | njohnston: could describe this for us? | 14:34 |
| SridarK | chandanc: i dont think u need 339246 & 337699 righ now | 14:34 |
| mfranc213 | chandanc: i will work on fixing those 2 unit tests. i won't push anything until you and i know how to avoid stomping on each other's feet. | 14:34 |
| chandanc | SridarK: thanks that will help | 14:35 |
| SridarK | mfranc213: yes we can do some offline coordination to avoid this | 14:35 |
| mfranc213 | maybe nate has published this somewhere | 14:36 |
| chandanc | SridarK: do you think i should upload my changes and update it later ? | 14:37 |
| SridarK | chandanc: if mfranc213: is at a logical point and not abt to push a patch up, u can make sure that u put ur changes over her latest version | 14:38 |
| SridarK | and push it up | 14:38 |
| chandanc | sure | 14:38 |
| mfranc213 | i'm at a logical stopping point other than the 2 unit tests--but maybe chandanc can fix those! :) | 14:39 |
| njohnston | sorry, I stepped away to get myself some breakfast | 14:39 |
| njohnston | catching up... | 14:39 |
| *** yamamoto has quit IRC | 14:40 | |
| njohnston | mfranc213: The way to avoid having two developers overwrite patchsets is just to always make sure you're up to date with `git review -d` before developing, and scrupulously check the change before uploading a patchset to make sure someone else hasn't uploaded one in the mean time. | 14:40 |
| njohnston | So yes, I think chandanc should push the driver updates as a new PS for 337699 | 14:42 |
| njohnston | brb | 14:43 |
| mfranc213 | thank you nate | 14:44 |
| SridarK | chandanc: as part of ur testing, will u be setting up a devstack env with the REST, db, plugin, 337699 (+ ur changes) | 14:44 |
| chandanc | yes | 14:44 |
| SridarK | i have been testing on devstack with the first 3 patches | 14:45 |
| SridarK | i will add 337699 once i wrap up the issue i am debugging now | 14:45 |
| *** yamamoto has joined #openstack-fwaas | 14:47 | |
| *** diogogmt has joined #openstack-fwaas | 14:51 | |
| chandanc | I just posted the patch, will update it once the tests are done | 14:57 |
| SridarK | chandanc: thanks | 14:58 |
| SridarK | chandanc: i have updated #link https://etherpad.openstack.org/p/fwaas-v2-l2-agent with curl cmds for create | 14:58 |
| chandanc | sure, plz wait for my update, might take a little time | 14:59 |
| SridarK | chandanc: no worries - i am dealing with db issues now - it will be a while b4 i get to it, but u can use the curl cmds for ur testing | 14:59 |
| SridarK | i have put in some samples, u will need to change the uuid's of the resources accordingly | 15:00 |
| SridarK | stepping away for abt 30 mins or so | 15:04 |
| chandanc | any idea how to trigger db migration ? | 15:18 |
| njohnston | chandanc: I believe this document describes how to do it with `neutron-db-manage`: http://docs.openstack.org/developer/neutron/devref/alembic_migrations.html | 15:21 |
| *** yamamoto has quit IRC | 15:28 | |
| *** mickeys has joined #openstack-fwaas | 15:28 | |
| *** mickeys has quit IRC | 15:32 | |
| SridarK | chandanc: one other alternative is to build out the tables manually | 15:37 |
| SridarK | i have been doing this | 15:37 |
| SridarK | if the migration script runs correctly will be the easiest way to do it | 15:40 |
| SridarK | if the basic set of tables get built out by the migration script - then if u uncover some missing column - u can always add that manually | 15:42 |
| njohnston | but if you find a missing column, comment about it on 311159 :-) | 15:44 |
| *** mickeys has joined #openstack-fwaas | 15:49 | |
| chandanc | made it to work with "neutron-db-manage --config-file /etc/neutron/neutron.conf upgrade heads" | 15:51 |
| chandanc | but looks like some tables creation order needs to change | 15:51 |
| SridarK | chandanc: ok tht is good | 15:52 |
| SridarK | if u do a describe <tablename> - for our tables - does it look decent | 15:53 |
| chandanc | I will send the diff to nate | 15:53 |
| SridarK | chandanc: i assume this is to do with a foreign key or backref ? | 15:54 |
| chandanc | ya | 15:54 |
| chandanc | thats the issue | 15:54 |
| chandanc | mysql> show tables like "%firewall%"; | 15:54 |
| chandanc | +---------------------------------------------+ | 15:54 |
| chandanc | | Tables_in_neutron (%firewall%) | | 15:54 |
| chandanc | +---------------------------------------------+ | 15:54 |
| chandanc | | cisco_firewall_associations | | 15:54 |
| chandanc | | firewall_policies | | 15:54 |
| chandanc | | firewall_router_associations | | 15:54 |
| chandanc | | firewall_rules | | 15:54 |
| chandanc | | firewall_v2_address_group_cidr_associations | | 15:54 |
| chandanc | | firewall_v2_address_groups | | 15:54 |
| chandanc | | firewall_v2_group_port_associations | | 15:54 |
| chandanc | | firewall_v2_groups | | 15:54 |
| chandanc | | firewall_v2_policies | | 15:54 |
| chandanc | | firewall_v2_policy_rule_associations | | 15:54 |
| chandanc | | firewall_v2_rules | | 15:54 |
| chandanc | | firewalls | | 15:54 |
| chandanc | | nsxv_firewall_rule_bindings | | 15:54 |
| chandanc | +---------------------------------------------+ | 15:54 |
| SridarK | that looks good | 15:55 |
| chandanc | mysql> desc firewall_v2_groups; | 15:56 |
| chandanc | +----------------------------+---------------+------+-----+---------+-------+ | 15:56 |
| chandanc | | Field | Type | Null | Key | Default | Extra | | 15:56 |
| chandanc | +----------------------------+---------------+------+-----+---------+-------+ | 15:56 |
| chandanc | | id | varchar(36) | NO | PRI | NULL | | | 15:56 |
| chandanc | | name | varchar(255) | YES | | NULL | | | 15:56 |
| chandanc | | description | varchar(1024) | YES | | NULL | | | 15:56 |
| chandanc | | project_id | varchar(36) | YES | | NULL | | | 15:56 |
| chandanc | | egress_firewall_policy_id | varchar(36) | NO | UNI | NULL | | | 15:56 |
| chandanc | | ingress_firewall_policy_id | varchar(36) | NO | UNI | NULL | | | 15:56 |
| chandanc | +----------------------------+---------------+------+-----+---------+-------+ | 15:56 |
| chandanc | mysql> desc firewall_v2_policies; | 15:56 |
| chandanc | +-------------+---------------+------+-----+---------+-------+ | 15:56 |
| chandanc | | Field | Type | Null | Key | Default | Extra | | 15:56 |
| chandanc | +-------------+---------------+------+-----+---------+-------+ | 15:56 |
| chandanc | | id | varchar(36) | NO | PRI | NULL | | | 15:56 |
| chandanc | | name | varchar(255) | YES | | NULL | | | 15:56 |
| chandanc | | description | varchar(1024) | YES | | NULL | | | 15:56 |
| chandanc | | project_id | varchar(36) | YES | | NULL | | | 15:56 |
| chandanc | | audited | tinyint(1) | YES | | NULL | | | 15:56 |
| chandanc | | public | tinyint(1) | YES | | NULL | | | 15:56 |
| chandanc | +-------------+---------------+------+-----+---------+-------+ | 15:56 |
| chandanc | 6 rows in set (0.01 sec) | 15:56 |
| SridarK | Hmm the v2 seems to be at the end in the sqlalchemy stuff | 15:56 |
| SridarK | https://review.openstack.org/#/c/311159/22/neutron_fwaas/db/firewall/v2/firewall_db_v2.py L#91 | 15:57 |
| njohnston | we should probably change the sqlalchemy to match the migration | 15:58 |
| SridarK | njohnston: yes either way - we can track this as a cleanup | 15:58 |
| SridarK | chandanc: a -1 with with this comment pls | 15:59 |
| SridarK | njohnston: since with our file naming, we seem to be putting v2 at the end | 16:00 |
| SridarK | shall we adopt a similar thing | 16:01 |
| njohnston | sure, I have no strong opinions on it | 16:01 |
| SridarK | njohnston: i am defn not religious on this naming thing | 16:01 |
| SridarK | njohnston: :-) | 16:01 |
| SridarK | chandanc: would it help if quickly push rev up to keep this matching | 16:02 |
| chandanc | done | 16:02 |
| SridarK | ie the migration table name and the sqlalc | 16:02 |
| chandanc | sorry didnot get your last comment | 16:02 |
| SridarK | chandanc: the naming does not match | 16:03 |
| SridarK | look at : https://review.openstack.org/#/c/311159/22/neutron_fwaas/db/firewall/v2/firewall_db_v2.py L#91 | 16:03 |
| SridarK | the table name above is firewall_v2_policies | 16:03 |
| SridarK | and in L#91 | 16:04 |
| SridarK | it firewall_policies_v2 | 16:04 |
| SridarK | i am wondering if that is the issue u hit | 16:04 |
| chandanc | oh | 16:04 |
| SridarK | i created the tables manually | 16:04 |
| SridarK | so i did not run the script - so i matched up with sqlal | 16:05 |
| chandanc | no this means I pulled in wron ps | 16:05 |
| SridarK | hmm not sure | 16:06 |
| SridarK | what issue did u see actually | 16:06 |
| SridarK | it is likely that since the sqlal did not match the table name, it would have complained | 16:07 |
| SridarK | njohnston: are u in the midst of pushing up a PS, if not let me quickly change this | 16:07 |
| njohnston | I'm working on changing it | 16:08 |
| SridarK | njohnston: ok perfect | 16:08 |
| chandanc | got it, the migration script needs update | 16:08 |
| chandanc | the model is correct | 16:08 |
| SridarK | ok u can pick up the next PS | 16:08 |
| chandanc | ya | 16:09 |
| chandanc | nate, will you also take care of the ordering issue ? | 16:09 |
| SridarK | chandanc: what ordering issue did u hit ? | 16:09 |
| chandanc | the backref and foreign key refs | 16:10 |
| njohnston | I just updated 311159 for the table names | 16:11 |
| njohnston | Can you be more specific about the backref/foreign key issue? | 16:12 |
| njohnston | chandanc: ^^ | 16:12 |
| SridarK | njohnston: ok perfect | 16:12 |
| SridarK | njohnston: i think in the script, u want the create table ordering as rules, policies, firewall_groups, policy_rule_association | 16:14 |
| SridarK | chandanc: can u pls confirm | 16:14 |
| chandanc | basically create the referenced table before the table referencing them, firewall_groups_v2 to be created before firewall_group_port_associations_v2 | 16:15 |
| njohnston | gotcha | 16:15 |
| njohnston | i'll do that now | 16:15 |
| SridarK | njohnston: also i think we are missing the group port table | 16:16 |
| chandanc | cool :), do i get another shot at the migration script with the new patch ? or i have to go manual ? | 16:16 |
| SridarK | let me point to that in gerrit | 16:16 |
| SridarK | njohnston: take that back | 16:17 |
| SridarK | it is there | 16:17 |
| SridarK | chandanc: yes pls give it another shot | 16:17 |
| SridarK | after njohnston 's update | 16:17 |
| chandanc | problem is the db is marked as updated, right ? | 16:17 |
| SridarK | this makes it so much easier over manually creating tables :-( | 16:18 |
| chandanc | I know :) | 16:18 |
| SridarK | i am afraid to mess with my devstack env for this reason | 16:18 |
| chandanc | i am assuming mine is already messed, i am not going to commit from this one any more, | 16:20 |
| chandanc | ubuntu@devstack5:~$ neutron-db-manage current | 16:20 |
| chandanc | Running current for neutron ... | 16:20 |
| chandanc | INFO [alembic.runtime.migration] Context impl MySQLImpl. | 16:20 |
| chandanc | INFO [alembic.runtime.migration] Will assume non-transactional DDL. | 16:20 |
| chandanc | a5648cfeeadf (head) | 16:20 |
| chandanc | 7d9d8eeec6ad (head) | 16:20 |
| chandanc | OK | 16:20 |
| chandanc | Running current for neutron-fwaas ... | 16:20 |
| chandanc | INFO [alembic.runtime.migration] Context impl MySQLImpl. | 16:20 |
| chandanc | INFO [alembic.runtime.migration] Will assume non-transactional DDL. | 16:20 |
| chandanc | d6a12e637e28 (head) | 16:20 |
| chandanc | f83a0b2964d0 (head) | 16:20 |
| chandanc | 4b47ea298795 (head) | 16:20 |
| chandanc | OK | 16:20 |
| chandanc | d6a12e637e28 is ours | 16:20 |
| chandanc | AFK for sometime, its getting late :( | 16:21 |
| SridarK | chandanc: understand - i think the best is to pull njohnston 's latest and restack with OFFLINE=True and RECLONE=no | 16:23 |
| njohnston | The table additions are now in correct order in the migration file in https://review.openstack.org/311159 | 16:23 |
| SridarK | njohnston: perfect | 16:23 |
| SridarK | thx | 16:23 |
| chandanc | will try it and let you know | 16:24 |
| njohnston | thanks! | 16:25 |
| *** vishwanathj_ has quit IRC | 16:36 | |
| chandanc | still need some change | 16:45 |
| chandanc | diff --git a/neutron_fwaas/db/migration/alembic_migrations/versions/d6a12e637e28_neutron_fwaas_v2_0.py b/neutron_fwaas/db/migration/alembic_migrations/versions/d6a12e637e28_neutron_fwaas_v2_0.py | 16:45 |
| chandanc | index 9f46f13..89bab52 100644 | 16:45 |
| chandanc | --- a/neutron_fwaas/db/migration/alembic_migrations/versions/d6a12e637e28_neutron_fwaas_v2_0.py | 16:45 |
| chandanc | +++ b/neutron_fwaas/db/migration/alembic_migrations/versions/d6a12e637e28_neutron_fwaas_v2_0.py | 16:45 |
| chandanc | @@ -32,6 +32,22 @@ import sqlalchemy as sa | 16:45 |
| chandanc | def upgrade(): | 16:45 |
| chandanc | 16:45 | |
| chandanc | op.create_table( | 16:45 |
| chandanc | + 'firewall_policies_v2', | 16:45 |
| chandanc | + sa.Column('id', sa.String(length=36), primary_key=True), | 16:46 |
| chandanc | + sa.Column('name', sa.String(length=255)), | 16:46 |
| chandanc | + sa.Column('description', sa.String(length=1024)), | 16:46 |
| chandanc | + sa.Column('project_id', sa.String(length=36)), | 16:46 |
| chandanc | + sa.Column('audited', sa.Boolean), | 16:46 |
| chandanc | + sa.Column('public', sa.Boolean)) | 16:46 |
| chandanc | + | 16:46 |
| chandanc | + op.create_table( | 16:46 |
| chandanc | + 'firewall_address_groups_v2', | 16:46 |
| chandanc | + sa.Column('id', sa.String(length=36), primary_key=True), | 16:46 |
| chandanc | + sa.Column('name', sa.String(length=255)), | 16:46 |
| chandanc | + sa.Column('description', sa.String(length=1024)), | 16:46 |
| chandanc | + sa.Column('project_id', sa.String(length=36))) | 16:46 |
| chandanc | + | 16:46 |
| chandanc | + op.create_table( | 16:46 |
| njohnston | chandanc: probably better to use paste.openstack.org | 16:46 |
| chandanc | 'firewall_rules_v2', | 16:46 |
| chandanc | sa.Column('id', sa.String(length=36), primary_key=True), | 16:46 |
| chandanc | sa.Column('name', sa.String(length=255)), | 16:46 |
| chandanc | @@ -54,15 +70,6 @@ def upgrade(): | 16:46 |
| chandanc | sa.Column('enabled', sa.Boolean)) | 16:46 |
| chandanc | 16:46 | |
| chandanc | op.create_table( | 16:46 |
| chandanc | - 'firewall_policies_v2', | 16:46 |
| chandanc | - sa.Column('id', sa.String(length=36), primary_key=True), | 16:46 |
| chandanc | - sa.Column('name', sa.String(length=255)), | 16:46 |
| chandanc | - sa.Column('description', sa.String(length=1024)), | 16:46 |
| chandanc | - sa.Column('project_id', sa.String(length=36)), | 16:46 |
| chandanc | - sa.Column('audited', sa.Boolean), | 16:46 |
| chandanc | - sa.Column('public', sa.Boolean)) | 16:47 |
| chandanc | - | 16:47 |
| chandanc | - op.create_table( | 16:47 |
| chandanc | 'firewall_groups_v2', | 16:47 |
| chandanc | sa.Column('id', sa.String(length=36), primary_key=True), | 16:47 |
| njohnston | please stop | 16:47 |
| chandanc | sa.Column('name', sa.String(length=255)), | 16:47 |
| chandanc | @@ -76,13 +83,6 @@ def upgrade(): | 16:47 |
| chandanc | nullable=False, unique=True)) | 16:47 |
| chandanc | 16:47 | |
| chandanc | op.create_table( | 16:47 |
| chandanc | - | 16:47 |
| chandanc | sure | 16:47 |
| chandanc | http://paste.openstack.org/show/554226/ | 16:48 |
| chandanc | oh sorry dint get you | 16:48 |
| njohnston | thanks! | 16:48 |
| njohnston | https://review.openstack.org/311159 updated | 16:50 |
| *** mickeys has quit IRC | 17:11 | |
| *** mickeys has joined #openstack-fwaas | 17:11 | |
| *** mickeys has quit IRC | 17:16 | |
| *** mickeys has joined #openstack-fwaas | 17:48 | |
| *** mickeys_ has joined #openstack-fwaas | 17:50 | |
| *** mickeys has quit IRC | 17:53 | |
| *** SumitNaiksatam has joined #openstack-fwaas | 17:55 | |
| njohnston | SridarK: ping | 18:19 |
| SridarK | njohnston: yes pls tell me | 18:19 |
| njohnston | SridarK: I see the issue with the test in 311159. It is doing "POST /firewall_policies.json HTTP/1.0" but it should be doing "POST /v2.0/fw/firewall_policies.json HTTP/1.0" | 18:20 |
| njohnston | SridarK: Also, Ihar has a question about the l3 agent extensions patch, and I am not sure of the answer: https://review.openstack.org/#/c/339246/12/neutron/agent/l3/agent.py | 18:20 |
| SridarK | njohnston: aha - ok so wrong path | 18:20 |
| njohnston | I'm not sure if the agent extension should be notified of router deletion contemporaneously with it getting enqueued to go to the agent, or if it should wait until the router is deleted | 18:22 |
| SridarK | njohnston: i guess if someone wants to do something b4 the router gets deleted, we may need to know when the intent is expressed | 18:24 |
| SridarK | then the ext entities can do something | 18:24 |
| njohnston | For the FWaaS use case, though, are we particular about it? | 18:24 |
| SridarK | in our case actually - the deletion is some we dont care abt at all | 18:24 |
| SridarK | *something | 18:25 |
| SridarK | as the rules will get cleaned up | 18:25 |
| SridarK | njohnston: pls give me a few mins, chandanc would like to debug an issue with his test | 18:25 |
| njohnston | OK, then I will acquesce to Ihar's suggestion, it seems sane to me | 18:25 |
| SridarK | njohnston: i was abt to get on a call with him | 18:26 |
| SridarK | njohnston: ok, i can piggyback there too if u need - i need to understand the workflow on that a bit more | 18:26 |
| *** mickeys_ has quit IRC | 18:27 | |
| *** mickeys has joined #openstack-fwaas | 18:27 | |
| *** mickeys has quit IRC | 18:37 | |
| *** mickeys has joined #openstack-fwaas | 18:38 | |
| *** chandanc_ has joined #openstack-fwaas | 18:39 | |
| *** mickeys has quit IRC | 18:42 | |
| *** chandanc has quit IRC | 18:42 | |
| SridarK | njohnston: it should be POST /v2.0/fwaas/ | 18:55 |
| njohnston | SridarK: But in the examples at https://etherpad.openstack.org/p/fwaas-v2-l2-agent it is given as /v2.0/fw not /v2.0/fwaas: "<your_devstack>:9696/v2.0/fw/firewall_rules" | 19:03 |
| SridarK | njohnston: yes i have to fix that - it was based on some hacks i had in my workspace | 19:03 |
| SridarK | but in terms of UT it should show up correctly | 19:04 |
| njohnston | ok | 19:04 |
| SridarK | i have updated the ether pad as well | 19:06 |
| njohnston | So I have the extensions patch (264489) patch pulled in to this... yet what I am seeing is: | 19:13 |
| njohnston | Request is: POST /v2.0/fwaas/firewall_policies.json HTTP/1.0 | 19:13 |
| njohnston | Accept: application/json | 19:13 |
| njohnston | Content-Length: 202 | 19:13 |
| njohnston | {"firewall_policy_v2": {"audited": true, "description": "default description", "shared": true, "firewall_rules": [], "tenant_id": "46f70361-ba71-4bd0-9769-3573fd227c4b", "name": "firewall_policy_v2_1"}} | 19:13 |
| njohnston | Result is: 404 Not Found | 19:13 |
| njohnston | Is there any other dependency I need to make a call like that work? | 19:13 |
| *** mickeys has joined #openstack-fwaas | 19:15 | |
| *** chandanc_ has quit IRC | 19:22 | |
| *** SumitNaiksatam has quit IRC | 19:41 | |
| *** SumitNaiksatam has joined #openstack-fwaas | 19:46 | |
| SridarK | njohnston: in a mtg so give me a few mins | 19:53 |
| njohnston | np | 19:53 |
| SridarK | njohnston: this is via the UT ? | 19:54 |
| SridarK | njohnston: one thing is that tenant_id is actually project_id with respect to the ext patch | 19:55 |
| SridarK | i had to hack some of this initially | 19:55 |
| SridarK | in the devstack env | 19:55 |
| SridarK | but with UT it should come out correctly unless there is something missing | 19:56 |
| njohnston | yes, this is from the UT; I put some print statements in | 20:07 |
| njohnston | But tenant_id versus project_id shouldn't give a 404, it's like the /v2.0/fwaas/firewall_policies.json endpoint doesn't exist, or doesn't support the POST method | 20:09 |
| njohnston | I figure it's something obvious I am missing | 20:24 |
| njohnston | these unit tests are more like functional tests if you ask me | 20:31 |
| njohnston | I was basing them off of the fwaas v1 UTs | 20:31 |
| *** mickeys has quit IRC | 20:39 | |
| SridarK | njohnston: the other thing possible is that without the plugin - perhaps a mock or something is needed for the new endpoint ? | 20:45 |
| SridarK | maybe running the UT where the ext / plugin patch is also present - i am a bit puzzled on this as well | 20:46 |
| njohnston | so I would need to pull in https://review.openstack.org/#/c/267046/ to get this to work, you think? | 20:47 |
| njohnston | The alternative would be to separate the unit tests out into a separate patch, that could then be expressed with dependencies on whatever else is needed | 20:48 |
| njohnston | I don't like that alternative, but it may be what we need to get going and get things merged | 20:49 |
| SridarK | njohnston: yes i agree too - the multiple patch thing is defn a pain | 20:55 |
| SridarK | i am trying to push out the policy update, keep getting sucked into multiple things , i am going go grab some lunch and let me put a wrap on that and i can also debug this issue | 20:57 |
| njohnston | thanks! | 20:57 |
| SridarK | if we can avoid another patch that will nice, but i agree - thats our plan B | 20:58 |
| *** mickeys has joined #openstack-fwaas | 20:58 | |
| *** SridarK has quit IRC | 21:02 | |
| njohnston | I just committed my PS for 311159 that has print statements, so you can see what I am seeing. | 21:05 |
| njohnston | I think that if we haven't gotten this sorted by tomorrow morning then we should probably split out the UTs, in order to make the Friday merge deadline. | 21:12 |
| njohnston | I intend to be in this channel a lot tomorrow, trying to get things merged. SridarK, do you know if shwetaap could be online to help us expedite getting her change merged? | 21:14 |
| njohnston | OK, I am going to stop talking to myself and go make dinner. See you all tomorrow! | 21:14 |
| *** njohnston is now known as njohnston_ | 21:59 | |
| *** njohnston has joined #openstack-fwaas | 22:01 | |
| *** njohnsto_ has joined #openstack-fwaas | 22:10 | |
| *** njohnston has quit IRC | 22:14 | |
| *** mickeys has quit IRC | 22:22 | |
| *** mickeys has joined #openstack-fwaas | 22:24 | |
| *** mickeys has quit IRC | 22:29 | |
| *** SridarK has joined #openstack-fwaas | 22:31 | |
| *** njohnsto_ has quit IRC | 22:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!