| *** atoth has quit IRC | 00:03 | |
| *** yamamoto has joined #openstack-lbaas | 00:18 | |
| *** yamamoto has quit IRC | 00:23 | |
| *** numans has quit IRC | 00:30 | |
| *** numans has joined #openstack-lbaas | 00:35 | |
| *** numans has quit IRC | 00:40 | |
| *** numans has joined #openstack-lbaas | 00:46 | |
| rm_work | back | 00:56 |
|---|---|---|
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Adds automatic ERD generation for the Octavia DB https://review.openstack.org/509319 | 01:01 |
| johnsom | I have wanted that for a while... I expect requirements to puke, but it's a start. | 01:02 |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Adds automatic ERD generation for the Octavia DB https://review.openstack.org/509319 | 01:05 |
| johnsom | I will put in the g-r tomorrow | 01:07 |
| rm_work | k | 01:09 |
| *** tonygunk has quit IRC | 01:11 | |
| johnsom | Hmm, there is a bindep too... | 01:12 |
| johnsom | http://logs.openstack.org/19/509319/2/check/gate-octavia-python27-ubuntu-xenial/b0aba2e/console.html#_2017-10-04_01_10_12_513299 | 01:12 |
| rm_work | lol such graphviz | 01:16 |
| *** numans has quit IRC | 01:25 | |
| *** numans has joined #openstack-lbaas | 01:25 | |
| rm_work | kong: so all of the code around creating a test backend server is gone :( | 01:26 |
| rm_work | at least, I don't see it | 01:26 |
| kong | rm_work: do you mean L125 in https://review.openstack.org/#/c/486775/20/octavia_tempest_plugin/tests/v2/scenario/test_basic_ops.py? | 01:28 |
| rm_work | ah, maybe | 01:28 |
| kong | yeah, i realized i should copy some code from tempest to octavia-tempest repo | 01:28 |
| kong | from tempest.common import compute is not considered a stable tempest interface | 01:29 |
| rm_work | ok so I can use this, I see | 01:29 |
| kong | although i think it should be a stable interface in tempest :-) | 01:29 |
| kong | i will ask qa team for that | 01:29 |
| *** numans has quit IRC | 01:30 | |
| kong | i see there may be several projects using it | 01:30 |
| kong | octavia, manila, etc | 01:30 |
| *** fnaval has joined #openstack-lbaas | 01:32 | |
| rm_work | kong: the only other thing is that we really don't need two servers -- we only need one, and put a webserver on two ports | 01:34 |
| rm_work | kong: this reduces the overhead a lot and will increase the speed in cases where we are missing nested-virt | 01:35 |
| kong | rm_work: you are right, it could be improved. | 01:36 |
| *** numans has joined #openstack-lbaas | 01:36 | |
| rm_work | wait, does this work? | 01:36 |
| kong | because i was using cirros, the server creation is very quick | 01:36 |
| rm_work | the _copy_file is AFTER the exec to chmod? | 01:36 |
| rm_work | you make the file, change the perms, and then copy the real file over it? | 01:37 |
| rm_work | or... hold on, trying to grok this | 01:37 |
| kong | hah, the order should be reversed, but in the current way, it works :-) | 01:39 |
| * rm_work is slightly confused | 01:39 | |
| rm_work | can we move the stuff around server creation up to base.py? | 01:40 |
| rm_work | because I will need it elsewhere | 01:40 |
| kong | rm_work: i remember | 01:40 |
| kong | if we don't create file and chmod first, the copy file will fail | 01:40 |
| kong | becuase of no permission | 01:41 |
| rm_work | lolwut | 01:41 |
| rm_work | ok | 01:41 |
| rm_work | that's ... super weird | 01:41 |
| rm_work | does tempest not have an scp client already? like, paramiko based or something? | 01:41 |
| rm_work | doing this via an scp command in a shell is weird | 01:42 |
| rm_work | that might have been what was already in the patch before you started working on it, but | 01:42 |
| rm_work | it's still weird :P | 01:42 |
| rm_work | did you steal _copy_file() directly from tempest? | 01:42 |
| kong | checked, tempest doesn't support yet | 01:43 |
| rm_work | hmm | 01:43 |
| rm_work | maybe we can just ... do it | 01:43 |
| rm_work | ? | 01:43 |
| rm_work | paramiko is easy | 01:43 |
| kong | another stuff could be improved :-) | 01:44 |
| kong | seem i have a lot of work to do | 01:45 |
| kong | stay tuned ;-) | 01:45 |
| rm_work | heh k | 01:48 |
| rm_work | i think Jude and I will be spending a lot of time reviewing tomorrow | 01:48 |
| rm_work | maybe we can get some of the first bits merged | 01:49 |
| rm_work | i can drop some of those comments on the first patch maybe | 01:49 |
| kong | rm_work: sure | 01:49 |
| *** longstaff has joined #openstack-lbaas | 01:52 | |
| rm_work | done | 01:52 |
| kong | rm_work: thanks! | 01:53 |
| rm_work | if you can get those comments addressed, we can try to get that merged by EOD tomorrow | 01:53 |
| rm_work | I will make it priority #1 | 01:53 |
| kong | rm_work: ok, will work on it today | 01:53 |
| rm_work | what times do you work? | 01:53 |
| kong | UTC+13 | 01:53 |
| rm_work | OK that ... actually works well for me | 01:54 |
| rm_work | so it's ... like 3pm now? | 01:56 |
| rm_work | we're not far off, just a whole day :P | 01:56 |
| kong | correct | 01:56 |
| rm_work | i'll see you during my afternoon tomorrow -- i have stuff in the morning until after lunchtime or so, which means i'll be getting on a little AFTER you in your morning ;) | 01:57 |
| kong | yeah, New Zealand is the first country to see the sun in the world :-) | 01:57 |
| *** catintheroof has joined #openstack-lbaas | 02:03 | |
| *** tongl has joined #openstack-lbaas | 02:13 | |
| *** catintheroof has quit IRC | 02:19 | |
| *** SumitNaiksatam has joined #openstack-lbaas | 02:21 | |
| *** yamamoto has joined #openstack-lbaas | 02:28 | |
| rm_work | though i'll also be on late today so if you get another patch up i can review sooner ;) | 02:28 |
| *** catintheroof has joined #openstack-lbaas | 02:29 | |
| kong | rm_work: not now, maybe later today. I am working on failover stuff | 02:30 |
| rm_work | ahhh | 02:33 |
| rm_work | what kind | 02:34 |
| rm_work | like, that's what i'm still working on now | 02:34 |
| rm_work | kong: what are you doing with failover | 02:37 |
| kong | write a script to do amphora upgrade (using falover capability provided by health_manager) | 02:38 |
| kong | it's amphora upgrade, not failover, actually | 02:38 |
| rm_work | ah ok so | 02:50 |
| rm_work | i am looking at doing that *in the API* | 02:50 |
| rm_work | right now | 02:50 |
| rm_work | my task this sprint includes: making a spec for an admin API call that can automate failing over large sets of amphorae | 02:50 |
| rm_work | for the purposes of HV maintenance (to get them off of a specific HV or AZ), or for upgrades (to rotate the fleet) | 02:51 |
| rm_work | kong: so ... maybe it would be better to collaborate on exactly what we both are looking for | 02:52 |
| rm_work | but this testing refactor stuff has been taking my time so i keep putting off my spec :P | 02:52 |
| kong | our task contains: create new amphora image, upload to glance set proper tag, do failover process for each slave amphora, make sure the new slave works fine, do the active ones, do some test again, etc. | 02:53 |
| rm_work | hmm k | 02:53 |
| kong | and we are going to use pike, which means we can't use failover api unless we backport internally | 02:54 |
| kong | i will evaluate | 02:54 |
| rm_work | so, my plan for the spec was just "select a set of loadbalancers; do failover process for backup; trigger failover to backup; do failover process for new backup" | 02:54 |
| rm_work | testing it internally was not part of what i was planning | 02:54 |
| rm_work | that gets a little ... finnicky | 02:55 |
| kong | rm_work: but before we replace both of the master and slave , how do you garantee the lb functionality is well? | 02:55 |
| rm_work | it's not easy to do | 02:55 |
| rm_work | since you don't have any guarantees about what is behind the LB | 02:55 |
| rm_work | i've worked this problem before -- there's no real good way to do it that I found, so basically you just have to move forward | 02:55 |
| rm_work | the best i came up with was "see what happens beforehand; make sure it's at least that good after" | 02:56 |
| kong | ok, good to know you have done some investigation | 02:56 |
| rm_work | so like, if you get timeout or 5xx beforehand, and it's a 5xx or timeout after... then, no change really | 02:56 |
| rm_work | but it's a LOT of work, and not necessarily something people will want | 02:56 |
| rm_work | because it means we are hitting the LB | 02:56 |
| rm_work | which ... | 02:57 |
| kong | hmm... i have not think about it in detail, just some plan | 02:57 |
| kong | something in my head | 02:57 |
| rm_work | if you think of a good way to do it... | 02:57 |
| rm_work | i mean, i have considered that we have some known-good node | 02:57 |
| kong | i will keep it as a secret :-) | 02:57 |
| rm_work | and we temporarily add a listener to some high port (that we verify isn't used) on the LB | 02:58 |
| rm_work | and then we use that for testing | 02:58 |
| rm_work | that could work | 02:58 |
| rm_work | but... O_o | 02:58 |
| rm_work | it'd have to be a node that's public-routable because we don't want to plug more networks to the amp (especially because there is no unplug) | 02:58 |
| rm_work | and i'm a little uncomfortable messing with the actual LB configs (ESPECIALLY en-masse) | 02:59 |
| rm_work | but it's the best option i've been able to think of | 02:59 |
| kong | no, we don't want to touch so deeply into the customers amphora | 02:59 |
| rm_work | lulz i suppose I should specify that if you think of someting ... you should tell me ;P | 02:59 |
| kong | rm_work: sure | 02:59 |
| rm_work | yeah, that is my feeling too | 02:59 |
| rm_work | in which case, we're back to being a little stuck | 03:00 |
| kong | ok, maybe just keep it simple as the first step | 03:00 |
| rm_work | i honestly was going to go with "as an admin, you should very that LBs work on the new image, before you trigger a mass failover" | 03:00 |
| rm_work | and then trust that the system works as it was designed | 03:00 |
| rm_work | johnsom: ^^ thoughts | 03:00 |
| kong | rm_work: thanks for the informative thoughts :-) | 03:01 |
| kong | especially before i actually do something | 03:01 |
| kong | i will bypass the verification part for now | 03:02 |
| kong | ;-) | 03:02 |
| * kong goes to change the patch | 03:02 | |
| rm_work | i mean, if we're *inside the API code*, we can at least reach out to the amphora-agent info endpoint, to confirm it's reachable | 03:03 |
| rm_work | that's one minimal thing i think is realistic | 03:03 |
| rm_work | one think i've been struggling with though is ... | 03:03 |
| rm_work | well, nm | 03:04 |
| rm_work | the answer is pretty easy | 03:04 |
| rm_work | this is an async thing so it needs to go to the CW | 03:04 |
| *** catintheroof has quit IRC | 03:09 | |
| rm_work | aight i am out for a bit but will be back on later | 03:09 |
| rm_work | though not sure how late you work | 03:09 |
| rm_work | i'll look for patches later ;) | 03:09 |
| *** numans has quit IRC | 03:11 | |
| *** numans has joined #openstack-lbaas | 03:11 | |
| *** longstaff has quit IRC | 03:29 | |
| *** longstaff has joined #openstack-lbaas | 03:34 | |
| *** longstaff has quit IRC | 03:44 | |
| *** fnaval has quit IRC | 03:48 | |
| *** longstaff has joined #openstack-lbaas | 03:56 | |
| *** longstaff has quit IRC | 03:58 | |
| *** numans has quit IRC | 03:58 | |
| *** longstaff has joined #openstack-lbaas | 04:00 | |
| *** numans has joined #openstack-lbaas | 04:00 | |
| *** armax has quit IRC | 04:05 | |
| *** numans has quit IRC | 04:05 | |
| *** longstaff has quit IRC | 04:06 | |
| *** numans has joined #openstack-lbaas | 04:12 | |
| *** mnaser has quit IRC | 04:15 | |
| *** sanfern has joined #openstack-lbaas | 04:29 | |
| *** mnaser has joined #openstack-lbaas | 04:35 | |
| *** numans has quit IRC | 04:39 | |
| *** numans has joined #openstack-lbaas | 04:42 | |
| *** eezhova has joined #openstack-lbaas | 04:53 | |
| *** aojea has joined #openstack-lbaas | 05:08 | |
| *** eezhova has quit IRC | 05:23 | |
| *** aojea has quit IRC | 05:35 | |
| *** tongl has quit IRC | 05:37 | |
| *** tongl has joined #openstack-lbaas | 05:42 | |
| *** numans has quit IRC | 05:50 | |
| *** numans has joined #openstack-lbaas | 05:51 | |
| *** aojea has joined #openstack-lbaas | 05:57 | |
| *** tongl has quit IRC | 06:02 | |
| *** aojea has quit IRC | 06:10 | |
| *** rm_work has quit IRC | 06:11 | |
| *** rm_work has joined #openstack-lbaas | 06:14 | |
| *** numans has quit IRC | 06:15 | |
| *** numans has joined #openstack-lbaas | 06:18 | |
| *** gcheresh_ has joined #openstack-lbaas | 06:29 | |
| *** rcernin has joined #openstack-lbaas | 06:30 | |
| *** numans has quit IRC | 06:45 | |
| *** numans has joined #openstack-lbaas | 06:48 | |
| *** pcaruana has joined #openstack-lbaas | 06:48 | |
| *** spectr has quit IRC | 06:54 | |
| *** spectr has joined #openstack-lbaas | 06:55 | |
| *** gcheresh_ has quit IRC | 06:55 | |
| *** nrado has joined #openstack-lbaas | 07:14 | |
| *** tesseract has joined #openstack-lbaas | 07:16 | |
| *** numans has quit IRC | 08:16 | |
| *** numans has joined #openstack-lbaas | 08:17 | |
| *** salmankhan has joined #openstack-lbaas | 08:25 | |
| *** eezhova has joined #openstack-lbaas | 08:25 | |
| *** salmankhan has quit IRC | 08:32 | |
| *** numans has quit IRC | 08:45 | |
| *** numans has joined #openstack-lbaas | 08:48 | |
| *** yamamoto has quit IRC | 09:01 | |
| *** gcheresh_ has joined #openstack-lbaas | 09:17 | |
| *** gcheresh_ has quit IRC | 09:27 | |
| *** yamamoto has joined #openstack-lbaas | 09:27 | |
| openstackgerrit | Dai Dang Van proposed openstack/neutron-lbaas master: Implement policy in code (1) https://review.openstack.org/509389 | 09:47 |
| openstackgerrit | Lingxian Kong proposed openstack/octavia master: Fix KeyError in failover error log https://review.openstack.org/509390 | 09:48 |
| *** aojea has joined #openstack-lbaas | 10:06 | |
| *** aojea has quit IRC | 10:10 | |
| *** yamamoto has quit IRC | 10:30 | |
| *** tesseract has quit IRC | 10:40 | |
| *** numans has quit IRC | 10:40 | |
| *** numans has joined #openstack-lbaas | 10:42 | |
| *** tesseract has joined #openstack-lbaas | 10:44 | |
| *** yamamoto has joined #openstack-lbaas | 10:51 | |
| *** spectr-RH has joined #openstack-lbaas | 10:59 | |
| *** spectr has quit IRC | 11:02 | |
| *** Alex_Staf has joined #openstack-lbaas | 11:15 | |
| *** tonygunk has joined #openstack-lbaas | 11:16 | |
| *** tonygunk has quit IRC | 11:20 | |
| *** sanfern has quit IRC | 11:22 | |
| *** numans has quit IRC | 11:30 | |
| *** numans has joined #openstack-lbaas | 11:33 | |
| *** salmankhan has joined #openstack-lbaas | 11:39 | |
| *** numans has quit IRC | 11:40 | |
| *** numans has joined #openstack-lbaas | 11:46 | |
| *** tonygunk has joined #openstack-lbaas | 11:50 | |
| *** atoth has joined #openstack-lbaas | 11:58 | |
| *** sanfern has joined #openstack-lbaas | 12:07 | |
| *** nrado1 has joined #openstack-lbaas | 12:18 | |
| *** nrado has quit IRC | 12:21 | |
| *** ianychoi has quit IRC | 12:25 | |
| *** ianychoi has joined #openstack-lbaas | 12:26 | |
| *** numans has quit IRC | 12:40 | |
| *** numans has joined #openstack-lbaas | 12:42 | |
| *** salmankhan has quit IRC | 12:48 | |
| *** spectr-RH has quit IRC | 13:00 | |
| *** sanfern has quit IRC | 13:02 | |
| *** sanfern has joined #openstack-lbaas | 13:02 | |
| *** pcaruana has quit IRC | 13:02 | |
| *** spectr-RH has joined #openstack-lbaas | 13:12 | |
| *** leitan has joined #openstack-lbaas | 13:14 | |
| *** pcaruana has joined #openstack-lbaas | 13:15 | |
| *** spectr-RH has quit IRC | 13:16 | |
| *** spectr-RH has joined #openstack-lbaas | 13:17 | |
| *** chlong_ has joined #openstack-lbaas | 13:22 | |
| *** catintheroof has joined #openstack-lbaas | 13:24 | |
| *** yamamoto has quit IRC | 13:27 | |
| *** longstaff has joined #openstack-lbaas | 13:31 | |
| *** numans has quit IRC | 13:31 | |
| *** numans has joined #openstack-lbaas | 13:31 | |
| *** longstaff has quit IRC | 13:43 | |
| *** armax has joined #openstack-lbaas | 13:45 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-octaviaclient master: Updated from global requirements https://review.openstack.org/509458 | 13:45 |
| *** tonygunk has quit IRC | 13:46 | |
| *** numans has quit IRC | 13:50 | |
| *** numans has joined #openstack-lbaas | 13:52 | |
| *** avtar-_ has joined #openstack-lbaas | 13:53 | |
| avtar-_ | hello fellas, | 13:55 |
| avtar-_ | I am facing the issue that octavia plugin is not loaded by neutron | 13:55 |
| avtar-_ | WARNING stevedore.named [req-45fd1a31-e2c3-4774-bbce-9fc8b40f1527 - - - - -] Could not load neutron_lbaas.drivers.octavia.driver.OctaviaDriver | 13:56 |
| johnsom | avtar-_ I believe that is just a neutron bug and that it will actually load. Have you tried it? | 13:57 |
| avtar-_ | I am getting this error while creating load-balancer | 14:01 |
| avtar-_ | There was an error in the driver: AttributeError: 'NoneType' object has no attribute 'rstrip' | 14:01 |
| avtar-_ | Full trace is here http://paste.openstack.org/show/622664/ | 14:02 |
| *** yamamoto has joined #openstack-lbaas | 14:03 | |
| *** yamamoto has quit IRC | 14:03 | |
| johnsom | avtar-_ So the Octavia driver loaded and ran ok, it's a keystone error you are seeing: | 14:03 |
| johnsom | File "/usr/lib/python2.7/site-packages/keystoneauth1/identity/v3/base.py", line 72, in token_url | 14:04 |
| johnsom | return '%s/auth/tokens' % self.auth_url.rstrip('/') | 14:04 |
| johnsom | So, this is some kind of configuration issue. | 14:05 |
| johnsom | I think it is the service_auth section in one of the neutron config files. I'm looking to refresh my mind. | 14:06 |
| johnsom | Yes, the "service_auth" section in the neutron_lbaas.conf file is likely not configured correctly. | 14:08 |
| johnsom | There is also a "service_auth" section in the neutron.conf that should be checked | 14:09 |
| avtar-_ | is there any simple configuration guide for this. | 14:09 |
| avtar-_ | ? | 14:09 |
| johnsom | https://docs.openstack.org/neutron/latest/configuration/neutron.html | 14:10 |
| avtar-_ | I mean for Octavia | 14:11 |
| johnsom | Oh, yes! | 14:11 |
| johnsom | Octavia docs are here: https://docs.openstack.org/octavia/latest/ | 14:12 |
| johnsom | Configuration option docs are: https://docs.openstack.org/octavia/latest/configuration/configref.html | 14:12 |
| johnsom | However I will note, the error you have is a neutron error as opposed to an Octavia issue. | 14:13 |
| johnsom | Neutron is attempting to get a keystone token to use when connecting to Octavia | 14:14 |
| *** tongl has joined #openstack-lbaas | 14:19 | |
| avtar-_ | which way do you prefer to install octavia . | 14:22 |
| avtar-_ | pip or yum | 14:23 |
| johnsom | The OpenStack team manages the pip package. The other packages are created by third parties and we don't always know what is included, etc. However, there is RedHat representation here in the channel | 14:24 |
| johnsom | Did you find a config example for the service_auth? I can paste one for you if it will help | 14:25 |
| *** tongl has quit IRC | 14:26 | |
| xgerman_ | o/ | 14:26 |
| avtar-_ | please paste | 14:27 |
| johnsom | https://www.irccloud.com/pastebin/Tjb1yFuz/ | 14:28 |
| johnsom | This is a sample from a devstack setup | 14:28 |
| *** pcaruana has quit IRC | 14:33 | |
| *** longstaff has joined #openstack-lbaas | 14:33 | |
| johnsom | FYI: https://review.openstack.org/#/c/509389 | 14:37 |
| johnsom | I am blocking this oslo policy patch in neutron-lbaas. | 14:37 |
| *** fnaval has joined #openstack-lbaas | 14:40 | |
| *** spectr-RH has quit IRC | 14:41 | |
| *** spectr has joined #openstack-lbaas | 14:42 | |
| *** pcaruana has joined #openstack-lbaas | 14:45 | |
| *** tonygunk has joined #openstack-lbaas | 14:55 | |
| *** numans has quit IRC | 15:00 | |
| *** numans has joined #openstack-lbaas | 15:02 | |
| *** eezhova has quit IRC | 15:03 | |
| *** yamamoto has joined #openstack-lbaas | 15:04 | |
| *** yamamoto has quit IRC | 15:13 | |
| *** rcernin has quit IRC | 15:13 | |
| *** pcaruana has quit IRC | 15:28 | |
| *** tesseract has quit IRC | 15:29 | |
| *** eezhova has joined #openstack-lbaas | 15:33 | |
| *** mdavidson has quit IRC | 15:39 | |
| *** jniesz has joined #openstack-lbaas | 15:43 | |
| *** nrado1 has quit IRC | 15:48 | |
| *** salmankhan has joined #openstack-lbaas | 16:00 | |
| *** tongl has joined #openstack-lbaas | 16:01 | |
| *** salmankhan has quit IRC | 16:11 | |
| *** eezhova has quit IRC | 16:26 | |
| rm_work | johnsom: ah, will miss the meeting today T_T | 16:48 |
| johnsom | Ok, bummer | 16:49 |
| johnsom | https://wiki.openstack.org/wiki/Octavia/Weekly_Meeting_Agenda#Meeting_2017-10-04 | 16:49 |
| *** tonygunk has quit IRC | 16:50 | |
| *** eezhova has joined #openstack-lbaas | 16:53 | |
| *** eezhova has quit IRC | 16:57 | |
| johnsom | Octavia meeting starting soon on #openstack-meeting | 16:58 |
| *** numans has quit IRC | 16:59 | |
| *** salmankhan has joined #openstack-lbaas | 17:00 | |
| *** SumitNaiksatam has quit IRC | 17:01 | |
| *** numans has joined #openstack-lbaas | 17:02 | |
| *** salmankhan has quit IRC | 17:17 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron-lbaas-dashboard master: Imported Translations from Zanata https://review.openstack.org/509555 | 17:19 |
| *** sshank has joined #openstack-lbaas | 17:24 | |
| *** sanfern has quit IRC | 17:24 | |
| *** salmankhan has joined #openstack-lbaas | 17:27 | |
| *** gcheresh_ has joined #openstack-lbaas | 17:27 | |
| *** SumitNaiksatam has joined #openstack-lbaas | 17:28 | |
| *** jniesz has quit IRC | 17:34 | |
| *** sshank has quit IRC | 17:39 | |
| *** gcheresh_ has quit IRC | 17:39 | |
| *** dulek has quit IRC | 17:41 | |
| *** sshank has joined #openstack-lbaas | 17:42 | |
| *** sshank has quit IRC | 17:42 | |
| *** sshank has joined #openstack-lbaas | 17:42 | |
| *** gcheresh_ has joined #openstack-lbaas | 17:48 | |
| *** dulek has joined #openstack-lbaas | 17:53 | |
| *** SumitNaiksatam has quit IRC | 17:58 | |
| *** SumitNaiksatam has joined #openstack-lbaas | 18:01 | |
| *** nrado has joined #openstack-lbaas | 18:04 | |
| *** numans has quit IRC | 18:04 | |
| *** numans has joined #openstack-lbaas | 18:06 | |
| *** gcheresh_ has quit IRC | 18:10 | |
| *** jniesz has joined #openstack-lbaas | 18:12 | |
| *** avtar-_ has quit IRC | 18:12 | |
| *** tonygunk has joined #openstack-lbaas | 18:16 | |
| *** salmankhan has quit IRC | 18:22 | |
| *** krypto has joined #openstack-lbaas | 18:42 | |
| *** eezhova has joined #openstack-lbaas | 18:44 | |
| *** sshank has quit IRC | 18:45 | |
| *** eezhova has quit IRC | 18:47 | |
| *** Alex_Staf has quit IRC | 18:51 | |
| *** oomichi is now known as oomichi_afk | 19:05 | |
| *** chlong_ has quit IRC | 19:16 | |
| *** tonygunk has quit IRC | 19:20 | |
| *** tonygunk has joined #openstack-lbaas | 19:23 | |
| *** atoth has quit IRC | 19:33 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Adds automatic ERD generation for the Octavia DB https://review.openstack.org/509319 | 19:53 |
| *** eezhova has joined #openstack-lbaas | 19:58 | |
| *** eezhova has quit IRC | 20:01 | |
| *** sshank has joined #openstack-lbaas | 20:05 | |
| *** gcheresh_ has joined #openstack-lbaas | 20:20 | |
| *** jniesz has quit IRC | 20:26 | |
| rm_work | kong: i'm around if you're working on patches today | 20:27 |
| rm_work | kong: i left a couple more comments | 20:42 |
| kong | rm_work: ok. the test failed last night when i use paramiko for scp implementation (cirros doesn't support sftp package), so i am going to keep the copy file thing | 20:42 |
| rm_work | err | 20:42 |
| rm_work | I am pretty sure it is possible to do it with cirros | 20:42 |
| rm_work | hold on | 20:42 |
| johnsom | Yeah, scp works fine with cirros | 20:42 |
| johnsom | use it all of the time | 20:43 |
| kong | i was using open_sftp() | 20:43 |
| johnsom | Yeah, try not to use sftp | 20:43 |
| rm_work | yeah i'm trying to find where in paramiko you do file transfers without sftp though | 20:43 |
| kong | would be great if you can show me something :-) | 20:44 |
| * kong goes to stand up meeting | 20:44 | |
| rm_work | i guess not in core paramiko? | 20:48 |
| rm_work | I see a couple of other options but not sure any are in g-r | 20:48 |
| rm_work | checked, they are not | 20:48 |
| kong | there is a scp module in pypi https://pypi.python.org/pypi/scp, but its license is LGPL, i believe we don't want to use it | 20:53 |
| rm_work | yeah | 20:54 |
| rm_work | also not in g-r | 20:54 |
| rm_work | hmmm | 20:54 |
| rm_work | so yeah I guess maybe we are stuck :/ | 20:54 |
| rm_work | wish they'd just added that to paramiko directly <_< | 20:55 |
| kong | so, can we just keep copy file thing as is, and maybe improve it in future, so i could test again quickly and make sure the new code works | 20:55 |
| rm_work | yeah, for now i guess that's fine | 20:56 |
| *** longstaff has left #openstack-lbaas | 21:00 | |
| *** nrado has quit IRC | 21:01 | |
| *** leitan has quit IRC | 21:01 | |
| *** gcheresh_ has quit IRC | 21:09 | |
| *** krypto has quit IRC | 21:20 | |
| kong | rm_work: did you test failover api before? | 21:39 |
| *** tonygunk has quit IRC | 21:39 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Adds automatic ERD generation for the Octavia DB https://review.openstack.org/509319 | 21:40 |
| kong | the o-cw will fail when it's updating the lb provision_status | 21:41 |
| *** catintheroof has quit IRC | 21:41 | |
| *** sshank has quit IRC | 21:46 | |
| *** sshank has joined #openstack-lbaas | 21:48 | |
| rm_work | i have not tested the existing failover-api | 21:51 |
| rm_work | i was going to test it before i +2'd it but it was merged before I got a chance to review | 21:52 |
| rm_work | i have no idea as to its efficacy | 21:52 |
| rm_work | i got the feeling it was pushed through to get it in the pike release as a "feature" so we could add bugfixes to it later but still have it in pike <_< | 21:53 |
| rm_work | kong: so it turns out, the spec i'm writing first is for maintenance mode, not managed-failover | 21:53 |
| rm_work | that one will be next | 21:53 |
| johnsom | I don't think that was in pike | 21:53 |
| rm_work | really? | 21:53 |
| rm_work | then what WAS the hurry then? | 21:54 |
| johnsom | IDK | 21:54 |
| kong | it's not in pike, i am testing on master | 21:56 |
| rm_work | yeah k | 21:56 |
| kong | but found that obvious bug | 21:56 |
| rm_work | well i haven't had a chance to look at it yet | 21:56 |
| rm_work | it was on my plate but review got merged to i moved on, and i guess i'll look at it when i try to use it | 21:56 |
| rm_work | *so i moved on | 21:57 |
| kong | so, i can not add two members using the same ip address but different ports? | 21:59 |
| rm_work | err, you can | 21:59 |
| rm_work | it is allowed | 21:59 |
| rm_work | the UNIQUE constraint is on IP+Port I believe -- maybe also Subnet? | 22:00 |
| rm_work | kong: the original code did have two members with the same IP, and it worked fine for me | 22:00 |
| kong | hmm, i know the reason | 22:01 |
| kong | my devstrack was setup for pike | 22:01 |
| kong | need run db migration | 22:01 |
| rm_work | ah | 22:05 |
| rm_work | i mean, we should NEVER have had that constraint | 22:05 |
| kong | ignore me | 22:06 |
| rm_work | but, i guess probably an unrelated issue that was just making it seem like it was a uniqueness issue :P | 22:06 |
| kong | for port in [80, 80]: | 22:06 |
| rm_work | heh | 22:06 |
| kong | maybe i need a coffee now | 22:06 |
| * kong goes to grab a cup of coffee | 22:06 | |
| rm_work | i could also use some coffee but that's not going to happen for a while T_T | 22:20 |
| *** fnaval has quit IRC | 22:46 | |
| *** fnaval has joined #openstack-lbaas | 23:00 | |
| *** armax has quit IRC | 23:01 | |
| johnsom | Ugh, the gates are slow.... queued for an hour and a half so far. I just want to see a five minute job run | 23:03 |
| *** sshank has quit IRC | 23:06 | |
| *** sshank has joined #openstack-lbaas | 23:08 | |
| *** SumitNaiksatam has quit IRC | 23:11 | |
| rm_work | T_T | 23:21 |
| rm_work | I wonder if even at 80% it's below the threshold for "draining" and is just always slowly increasing | 23:22 |
| rm_work | I was kinda afraid that we were close to capacity in general | 23:22 |
| *** sshank has quit IRC | 23:26 | |
| *** sshank has joined #openstack-lbaas | 23:33 | |
| *** tongl has quit IRC | 23:44 | |
| *** sshank has quit IRC | 23:48 | |
| *** sshank has joined #openstack-lbaas | 23:55 | |
| *** armax has joined #openstack-lbaas | 23:58 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!