| *** bobmel has joined #openstack-net-bgpvpn | 00:01 | |
| *** bobmel has quit IRC | 00:06 | |
| *** bobmel has joined #openstack-net-bgpvpn | 02:02 | |
| *** bobmel has quit IRC | 02:07 | |
| *** bobmel has joined #openstack-net-bgpvpn | 04:03 | |
| *** bobmel has quit IRC | 04:07 | |
| *** bobmel has joined #openstack-net-bgpvpn | 06:04 | |
| *** bobmel has quit IRC | 06:09 | |
| *** bobmel has joined #openstack-net-bgpvpn | 06:22 | |
| *** bobmel has quit IRC | 06:27 | |
| *** bobmel has joined #openstack-net-bgpvpn | 06:43 | |
| *** bobmel_ has joined #openstack-net-bgpvpn | 06:46 | |
| *** bobmel_ has quit IRC | 06:48 | |
| *** bobmel has quit IRC | 06:49 | |
| *** bobmel has joined #openstack-net-bgpvpn | 06:50 | |
| *** bobmel_ has joined #openstack-net-bgpvpn | 07:00 | |
| *** bobmel has quit IRC | 07:02 | |
| *** matrohon has joined #openstack-net-bgpvpn | 07:51 | |
| *** tmorin has joined #openstack-net-bgpvpn | 08:01 | |
| *** tmorin has quit IRC | 08:13 | |
| *** openstackgerrit has quit IRC | 09:03 | |
| *** mickeys has quit IRC | 09:14 | |
| *** tmorin has joined #openstack-net-bgpvpn | 10:00 | |
| *** openstackgerrit has joined #openstack-net-bgpvpn | 10:01 | |
| openstackgerrit | Thomas Morin proposed openstack/networking-bgpvpn master: bagpipe driver: cleanup, remove OVSInterceptBridge https://review.openstack.org/425933 | 10:01 |
|---|---|---|
| *** mickeys has joined #openstack-net-bgpvpn | 10:15 | |
| *** mickeys has quit IRC | 10:19 | |
| *** mickeys has joined #openstack-net-bgpvpn | 11:16 | |
| *** mickeys has quit IRC | 11:20 | |
| *** matrohon has quit IRC | 11:25 | |
| *** bobmel_ has quit IRC | 11:27 | |
| *** mickeys has joined #openstack-net-bgpvpn | 12:16 | |
| *** mickeys has quit IRC | 12:21 | |
| *** matrohon has joined #openstack-net-bgpvpn | 13:33 | |
| pcarver | vks1: hi. are you still around? | 13:52 |
| *** mickeys has joined #openstack-net-bgpvpn | 14:00 | |
| *** mickeys has quit IRC | 14:04 | |
| pcarver | tmorin: got a few minutes? | 14:09 |
| tmorin | hi Paul | 14:09 |
| tmorin | yes I do | 14:09 |
| pcarver | I've finally got some time free of meetings and I'm tackling this change to use the API definition from neutron-lib | 14:10 |
| pcarver | I'm looking for suggestions on the best way to start troubleshooting. | 14:10 |
| pcarver | specifically whether I should start in with a debugger or if there are easier things to try first | 14:10 |
| pcarver | The py27 tox results are not especially helpful | 14:11 |
| pcarver | I have Devstack running with my modified bgpvpn code and commands like neutron net-list work but neutron bgpvpn-list fails with a not very useful output | 14:12 |
| tmorin | pcarver: I need to know your starting point | 14:12 |
| tmorin | ok | 14:12 |
| tmorin | what I would first do is run "tox -epy27" | 14:12 |
| tmorin | then dig into each failing test | 14:12 |
| tmorin | to dig, one thing I have found useful is the following: | 14:12 |
| tmorin | a) go into the py27 venv: source .tox/py27/bin/activate | 14:13 |
| pcarver | All the tests are failing with what looks like a fairly generic server error | 14:13 |
| tmorin | b) run: python -m unittest -v full.path.to.test_which_is_failing | 14:14 |
| pcarver | I tried activating the py27 env and running ipython and importing my modified .py file but I wasn't sure if I was heading down a useful path | 14:14 |
| tmorin | what is the issue you get ? | 14:14 |
| tmorin | pcarver: it could help detecting some basic errors, but other errors may not be | 14:14 |
| tmorin | .. may not be detected by just importing the file | 14:14 |
| pcarver | so first off, running "neutron bgpvpn-list" with a running devstack gives "The resource could not be found.<br /><br />" | 14:14 |
| pcarver | I'm re-running py27 now so I can copy and paste the error message to you. I've forgotten the exact wording I was seeing yesterday. | 14:15 |
| pcarver | BTW, how long does py27 take for you? They just migrated me to a new laptop and running in VirtualBox with four vCPUs still seems to me to take a long time. This new machine should be pretty decent. | 14:17 |
| pcarver | It's a Core i7 at 2.7GHz with 32GB RAM and an SSD and I've given the Devstack VM 8GB. I haven't timed py27 but it sure seems to take a long time. | 14:19 |
| * tmorin running tox -ep27 to give pcarver a timing... | 14:19 | |
| pcarver | The run right now is probably extra slow because I deleted .tox again just to make sure I wasn't getting tripped up by that | 14:19 |
| pcarver | oh, this is different than what I was seeing yesterday. Yesterday I was seeing a bunch of successful tests and a bunch of failed tests with all the failures looking like pretty much the same error about a server response | 14:21 |
| pcarver | Today after deleting .tox I'm seeing something different. I'll post a paste | 14:21 |
| tmorin | yes, the longest part of the run is cloning the neutron repo.... | 14:22 |
| *** malos has joined #openstack-net-bgpvpn | 14:22 | |
| pcarver | http://paste.openstack.org/show/601943/ | 14:23 |
| pcarver | I'm going to checkout master and re-run to make sure I'm still getting success on master | 14:23 |
| pcarver | ok, now I'm seeing approximately the same error shown in that paste when I run py27 with a checkout of master for networking-bgpvpn. I was getting success yesterday. | 14:25 |
| tmorin | I know the reason for this error: ImportError: cannot import name context | 14:27 |
| tmorin | you need a recent master of net-bgpvpn that has a change required to follow a change in neutron master | 14:28 |
| tmorin | (was merged two days ago : https://review.openstack.org/#/c/422266/ ) | 14:28 |
| pcarver | ok, so I should pull master and then rebase my change on top of it? | 14:29 |
| tmorin | yep | 14:33 |
| pcarver | tmorin: ok, after rebasing py27 works on master again. The not very helpful message on all of the failing bgpvpn test cases on my modified branch is "webob.exc.HTTPClientError: The server could not comply with the request since it is either malformed or otherwise incorrect." | 14:45 |
| pcarver | so I assume I've got some fairly basic mistake that is just preventing the bgpvpn code from running | 14:45 |
| tmorin | then I would run a single test in the venv with "python -m unittest -v full.path.to.test_which_is_failing" | 14:46 |
| pcarver | and so my question is, are there any easy things I should check before I start using a debugger to step through | 14:46 |
| tmorin | the above ^ | 14:46 |
| tmorin | because it may let you see error messages that tox is eating | 14:46 |
| pcarver | ok, I'll try that | 14:47 |
| pcarver | tmorin: I've got to go deal with someone else right now, but there seems to be something related to BaGPipe that I'll need to figure out. | 14:51 |
| pcarver | I get this error from tox on my branch of networking-bgpvpn "Could not load networking_bgpvpn.neutron.services.service_drivers.bagpipe.bagpipe.BaGPipeBGPVPNDriver" | 14:51 |
| tmorin | you should be able to get a more detailed error log | 14:51 |
| pcarver | no, I take it back. I get that same error on master but the test passes | 14:51 |
| tmorin | which test is giving that ?? | 14:52 |
| pcarver | I'll continue after I finish with this other person. The test I selected (at random out of the failing ones) is networking_bgpvpn.tests.unit.services.test_plugin.TestBGPVPNServicePlugin.test_bgpvpn_router_assoc_create | 14:52 |
| tmorin | seems fishy... | 14:52 |
| tmorin | and does python -m unittest -v ... gives you a more detailed log ? | 14:53 |
| tmorin | ok | 14:53 |
| tmorin | go ahead with the other person, I can also do things in parallel :) | 14:53 |
| pcarver | http://paste.openstack.org/show/601950/ | 14:53 |
| *** matrohon has quit IRC | 15:23 | |
| *** matrohon has joined #openstack-net-bgpvpn | 15:42 | |
| tmorin | hmm.. harder to investigate indeed | 15:51 |
| tmorin | I guess the API validation framework in neutron is failing to validate the bgpvpn creation, but the error is not reported to us ... | 15:52 |
| openstackgerrit | Merged openstack/networking-bgpvpn master: Switch to use stable data_utils https://review.openstack.org/442776 | 16:12 |
| *** mickeys has joined #openstack-net-bgpvpn | 17:11 | |
| *** tmorin has quit IRC | 17:14 | |
| pcarver | tmorin: This is very puzzling. I've been poking around in a debugger and everything looks quite reasonable in networking_bgpvpn/neutron/extensions/bgpvpn.py | 17:44 |
| pcarver | I think the next thing I'm going to do is back out my changes and add them a few at a time to see if I can isolate something specific that triggers the failures. | 17:46 |
| *** matrohon has quit IRC | 18:03 | |
| *** mickeys has quit IRC | 23:13 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!