| eandersson | So if a member is down in for example mitaka it's supposed to show as offlien? | 00:00 |
|---|---|---|
| rm_work | it should yes | 00:00 |
| rm_work | johnsom: hmm, this l7policy POST controller isn't the cleanest code <_< looks like i left some commented code there | 00:00 |
| rm_work | but regardless | 00:01 |
| rm_work | https://github.com/openstack/octavia/blob/master/octavia/api/v2/controllers/l7policy.py#L142-L152 | 00:01 |
| rm_work | the last line there (152) should always set that | 00:01 |
| *** cpuga_ has quit IRC | 00:01 | |
| eandersson | rm_work, is there anything for 3rd party drivers btw? | 00:04 |
| eandersson | or is this only available for octavia (haproxy)? | 00:04 |
| eandersson | We are just looking to see if there is a driver we could implement for a10 | 00:05 |
| johnsom | eandersson So, if a member is down it should show in ERROR, the pool should go to DEGRADED. This should work for drivers other than octavia, but that is up to the vendor to implement and notify lbaas | 00:05 |
| rm_work | johnsom: shouldn't the member show as OFFLINE? | 00:06 |
| rm_work | johnsom: or does n-lbaas not have that status | 00:06 |
| johnsom | No, OFFLINE is admin-state-up=False | 00:06 |
| rm_work | err | 00:06 |
| rm_work | ok | 00:06 |
| rm_work | i think maybe that answers part of a question i've been dealing with recently | 00:06 |
| johnsom | failed is ERROR | 00:06 |
| rm_work | which is to say, why we get ERROR status on members | 00:06 |
| eandersson | Do you have an example driver that implements this? | 00:07 |
| johnsom | Yes, octavia does | 00:07 |
| eandersson | I assume this is it for Octavia | 00:07 |
| eandersson | https://github.com/openstack/octavia/blob/master/octavia/controller/healthmanager/health_manager.py | 00:07 |
| johnsom | But vendor driver, hmmm. Did you say you are trying to use the A10 driver? | 00:07 |
| eandersson | Yes | 00:07 |
| johnsom | eandersson Here is the logic octavia uses to "calculate" a load balancer's health: https://github.com/openstack/octavia/blob/master/octavia/controller/healthmanager/update_db.py#L114 | 00:09 |
| eandersson | Yep - that make sense | 00:10 |
| eandersson | Is that what would have been used in neutron-lbaas as well? | 00:10 |
| johnsom | As for A10, they have a stub in-tree, but the bulk of it is in their own package. | 00:10 |
| eandersson | I would have assumed that the status code would have been implemented on the neutron-lbaas side | 00:10 |
| johnsom | Here are docs BTW: https://developer.openstack.org/api-ref/load-balancer/v2/index.html#status-codes | 00:11 |
| eandersson | So for Octavia if we wanted to add status updates for the a10 we would extend the Health Manager? | 00:12 |
| eandersson | Or would it be completely re-implemented using the vendor driver? | 00:12 |
| johnsom | The vendor driver should provide that status back. For drivers that plug into the new octavia API we have not yet finished deciding how that is going to work. | 00:13 |
| *** JudeC has quit IRC | 00:13 | |
| eandersson | I see | 00:13 |
| johnsom | For a neutron-lbaas vendor driver, it should already be providing that back to lbaas. | 00:14 |
| johnsom | Some vendors calculate "degraded" differently, for example more than 30% of the members failed or such. This way the vendor can make their own status decisions. | 00:14 |
| eandersson | So would the vendor driver by listening to events, and then modifing the database directly? | 00:16 |
| eandersson | I can't see any sort of endpoint to override status | 00:17 |
| johnsom | Well, I really hope they are not modifying the database directly. That is bad coding as we can change the schema. There should be call backs for the drivers. | 00:17 |
| rm_work | well | 00:17 |
| eandersson | Yea exactly | 00:17 |
| eandersson | That is what I am worried about | 00:18 |
| rm_work | member updates would need callbacks... but there isn't an endpoint for that | 00:18 |
| rm_work | sooo | 00:18 |
| rm_work | i think we were trying to get them to use the same event-queue system | 00:18 |
| rm_work | but i doubt anyone implemented it | 00:18 |
| eandersson | I just don't see how the vendors could possible implement something at the moment | 00:18 |
| johnsom | I thought the drivers were using the neutron agent stuff for that | 00:18 |
| rm_work | it wouldn't matter if it was octavia or a vendor dumping statuses into rabbit | 00:18 |
| rm_work | right? | 00:19 |
| rm_work | or is that queue really deeply integrated into the octavia driver | 00:19 |
| rm_work | we may not have made it generic | 00:19 |
| johnsom | Yeah, I don't think that is the right answer. | 00:20 |
| rm_work | well, i also noticed that we did not provide a way for these updates to happen otherwise | 00:20 |
| rm_work | there would need to be something listening | 00:20 |
| johnsom | I mean to some degree neutron doesn't need to keep this status info for vendor driver, it can just query straight through to their driver and query the device right? | 00:20 |
| rm_work | and given that none of the drivers seem to have implemented such a listener in their own driver code | 00:20 |
| rm_work | i don't think it's being done for any vendors | 00:20 |
| rm_work | ah so you mean it's actually not caching statuses | 00:21 |
| rm_work | their driver might do checks syncronously | 00:21 |
| rm_work | when the user does a GET | 00:21 |
| johnsom | I'm not sure vendor drivers really need to... | 00:21 |
| rm_work | to be fair, I kinda wish we'd just done that for octavia | 00:21 |
| rm_work | then we'd have less stupidity with DB redundancy | 00:21 |
| rm_work | it's not like calling out to octavia for member statuses would have been slow | 00:22 |
| rm_work | or really, ALL statuses | 00:22 |
| johnsom | Yeah, I don't know why that was done really. Why does the neutron DB need to care? | 00:22 |
| rm_work | grr where did the backslash-line-continuation check come from | 00:24 |
| rm_work | i can't get this to fail | 00:24 |
| *** aojea has joined #openstack-lbaas | 00:24 | |
| johnsom | Did they just drop it up stream? | 00:24 |
| rm_work | erm i hope not | 00:24 |
| rm_work | because i also hate it | 00:24 |
| rm_work | and if those start infecting us | 00:25 |
| rm_work | I'll cry | 00:25 |
| *** aojea has quit IRC | 00:28 | |
| *** blogan has quit IRC | 00:29 | |
| johnsom | rm_work yeah, testing with octavia, it seems to pass | 00:31 |
| rm_work | bleeegh | 00:33 |
| rm_work | how do we add it back? as an "O" rule? | 00:34 |
| rm_work | I need to find the old check now... | 00:34 |
| rm_work | hmm | 00:34 |
| rm_work | "It is preferred to wrap long lines in parentheses and not a backslash for line continuation." is written in the hacking guide | 00:35 |
| rm_work | but just "preferred" | 00:35 |
| johnsom | You can add one here: https://github.com/openstack/octavia/blob/master/octavia/hacking/checks.py | 00:35 |
| rm_work | yeah | 00:36 |
| rm_work | i need to find how they used to implement it | 00:36 |
| rm_work | so i don't have to figure it out from scratch | 00:36 |
| rm_work | digging through the hacking repo's old tags | 00:36 |
| eandersson | So as a hack for now could I override the status call and just make a call to the vendor driver? | 00:42 |
| johnsom | I have been trying to look at the neutron-lbaas code (before my time) and figure out what the heck is going on for vendor status | 00:43 |
| rm_work | we REALLY need to get vendor drivers in octavia finished and working | 00:44 |
| johnsom | Sigh, yes | 00:45 |
| johnsom | eandersson It looks like the netscaler driver is updating status, maybe have a look at that | 00:45 |
| johnsom | It looks like the way the providers are setup in neutron-lbaas it doesn't pass the status requests through to the providers, so it looks like the DB needs to be updated. | 00:47 |
| johnsom | As short term solutions (Octavia API will have a better answer here), I say either try to use the rabbit queue octavia is using or just setup the driver to write to the DB like netscaler. | 00:49 |
| openstackgerrit | Merged openstack/octavia master: Updated from global requirements https://review.openstack.org/470938 | 00:49 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Minor code cleanup in l7policy controller https://review.openstack.org/471176 | 00:51 |
| johnsom | Ok, I have no idea how I ended up with the null project_id. Hmmm | 00:53 |
| rm_work | <_< | 00:53 |
| rm_work | yeah it's rather odd | 00:53 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: WIP: Floating IP Network Driver (spans L3s) https://review.openstack.org/435612 | 00:53 |
| rm_work | ^^ just rebasing again | 00:54 |
| rm_work | literally just have to refresh that every few weeks | 00:54 |
| rm_work | it's in production here though :) | 00:54 |
| eandersson | RabbitMQ messages would have been perfect, but will try to use the same db approach for now then | 00:54 |
| johnsom | eandersson well, you can try to use the rabbit method the octavia driver uses. I wouldn't stop you | 00:55 |
| johnsom | Just note, this is all for neutron-lbaas which is going to go away | 00:56 |
| eandersson | I think that will be our goal long term, but pre-octavia I think we will try to mimic the code you linked | 00:56 |
| johnsom | If you have input on how vendor drivers should work under Octavia API, feel free to give input. We will be writing up a spec | 00:57 |
| johnsom | Ok | 00:57 |
| eandersson | Yea - that isn't easy | 00:57 |
| *** fnaval has quit IRC | 00:57 | |
| eandersson | Maybe we can help out with some blueprints | 00:58 |
| rm_work | would be appreciated | 00:58 |
| rm_work | we're trying to envision how it needs to work | 00:58 |
| rm_work | but | 00:58 |
| rm_work | we're not a vendor I guess :P so it's hard | 00:58 |
| eandersson | Yea was just gonna say the same | 00:59 |
| rm_work | I wish sam/evgeny/kobi would pop by and give feedback | 00:59 |
| johnsom | Yeah, and we are still focused on getting the API and openstack client finished | 00:59 |
| eandersson | For me ideally vendors would push events | 00:59 |
| eandersson | but guessing that in most cases you would have to do periodical checks | 00:59 |
| *** fnaval has joined #openstack-lbaas | 00:59 | |
| rm_work | yeah, probably similar to event-streamer | 00:59 |
| rm_work | would be ideal | 00:59 |
| rm_work | but they'd need a daemon on their side | 00:59 |
| rm_work | i think a lot of "box solutions" don't do that? | 01:00 |
| rm_work | though maybe a lot actually do now that i think about it | 01:00 |
| rm_work | IIRC stingray had hooks for that, as did f5? | 01:00 |
| *** leitan has joined #openstack-lbaas | 01:01 | |
| johnsom | We just need to have the interface, what the driver does internally (poll or other) is up to the vendor | 01:01 |
| rm_work | yeah | 01:01 |
| rm_work | truth | 01:01 |
| rm_work | vendors are welcome to have a polling daemon | 01:01 |
| rm_work | though -- wouldn't that mean we'd be encouraging them to update our DB? | 01:01 |
| rm_work | and is that... OK? | 01:01 |
| rm_work | I guess it is? | 01:01 |
| johnsom | Like the netscaler, it's forking off a periodic job | 01:02 |
| johnsom | Not native, but through an interface or queue it should be fine | 01:02 |
| rm_work | right, so yeah | 01:02 |
| rm_work | event-streamer would be the ideal interface I think | 01:02 |
| rm_work | which means we'd need to make that available in octavia now as well | 01:03 |
| rm_work | which is fine | 01:03 |
| johnsom | I had talked about a endpoint like we have for octavia health too. | 01:03 |
| johnsom | Ah well, not up to doing that design work tonight. | 01:04 |
| rm_work | lol yeah... | 01:04 |
| rm_work | i'm happy the docs merged... | 01:05 |
| johnsom | I think the OSC l7policy review is good. Just need to do rule tomorrow and those client patches have a review pass. | 01:05 |
| johnsom | Me too! | 01:05 |
| johnsom | Did all three? | 01:05 |
| johnsom | I need to get back to that too | 01:05 |
| rm_work | yes | 01:05 |
| rm_work | i need to run another devstack, my stack exploded with nova errors so i was only able to test create/list/show for LB | 01:06 |
| rm_work | but they seemed fine so I'm +2 | 01:06 |
| johnsom | We still need stats, status, and quota for OSC I think. | 01:07 |
| rm_work | ah yeah | 01:07 |
| johnsom | I still need to write the docs for l7 and quota, then I think that is good. | 01:07 |
| rm_work | well we need stats/status for octavia first :P | 01:07 |
| johnsom | I'm really itching to write up a step-by-step install guide too | 01:07 |
| johnsom | Well, yeah, details.... | 01:08 |
| johnsom | This could use a review too: https://review.openstack.org/#/c/470081/ | 01:09 |
| johnsom | I need to follow up with german tomorrow as I'm not sure I follow his comment | 01:09 |
| *** catintheroof has quit IRC | 01:10 | |
| rm_work | yeah i don't have a good test setup with active/standby yet so i didn't get to testing it | 01:11 |
| rm_work | and our gates don't have one | 01:11 |
| johnsom | Ok, time to make dinner. I will be on later tomorrow morning as I have a dentist appointment. | 01:11 |
| rm_work | k | 01:12 |
| *** gongysh has joined #openstack-lbaas | 01:16 | |
| *** eandersson has quit IRC | 01:25 | |
| *** eandersson has joined #openstack-lbaas | 01:26 | |
| *** armax has quit IRC | 01:44 | |
| leitan | Hi guys | 01:48 |
| leitan | trying to rebuild the aphorae image to get the keepalive race condition patch that has been merged into master | 01:50 |
| leitan | im getting "ERROR: Element 'sysctl' not found in '/tmp/octavia/elements" | 01:50 |
| johnsom | Hmm, sysctl come from the diskimage-builder package | 01:52 |
| johnsom | Is it installed? | 01:52 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron-lbaas-dashboard master: Updated from global requirements https://review.openstack.org/471179 | 01:53 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/octavia-tempest-plugin master: Updated from global requirements https://review.openstack.org/471180 | 01:53 |
| leitan | johnsom: yes, i needed to install it cause if its not present the binary disk-image-create is not availabe | 01:55 |
| leitan | and you dont get to that point without it | 01:55 |
| johnsom | Ok. Hmm, it should be getting that element from there | 01:56 |
| leitan | ii python-diskimage-builder 1.0.0-1 all image building tools for Openstack | 01:56 |
| johnsom | You are just running diskimage-create -s 2 ? | 01:56 |
| johnsom | That is how our gates run it | 01:57 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-octaviaclient master: Updated from global requirements https://review.openstack.org/471182 | 01:57 |
| leitan | johnsom: just the diskimage-create.sh script inside octavia/diskimage-create folder | 01:59 |
| johnsom | Try adding -s 2 | 01:59 |
| leitan | Building elements: base base vm ubuntu haproxy-octavia-ubuntu rebind-sshd no-resolvconf amphora-agent-ubuntu keepalived-octavia-ubuntu pip-cache certs-ramfs | 02:00 |
| leitan | ERROR: Element 'sysctl' not found in '/tmp/octavia/elements:/usr/bin/../share/diskimage-builder/elements' | 02:00 |
| leitan | same | 02:00 |
| leitan | fresh clone, from today, both repos | 02:01 |
| johnsom | Hmm. That last path, usr/bin... if you ls that directory, is sysctl there? | 02:02 |
| johnsom | I know it was working today as we merged patches. Each patch builds an image that way | 02:03 |
| leitan | johnsom: ill try pip installing it | 02:04 |
| leitan | nothing there | 02:05 |
| johnsom | Also, make sure the script is using the same version of python and installed it | 02:05 |
| johnsom | I.e. Py3 vs py2 | 02:05 |
| leitan | hmm, the pip install didnt added anything, it has to be something related with my host, ill dig in | 02:07 |
| leitan | the only difference is that im trying to run it on xenial | 02:08 |
| leitan | now, and when it worked on other server, was trusty | 02:09 |
| johnsom | Yeah, on xenial both python 3 and python 2 install by default. I have had problems where the pip install goes into py3 directory but the script was running py2. We support both 2 & 3 just not mixed installs | 02:12 |
| leitan | johnsom: launching a 14.04 to try it our | 02:14 |
| leitan | out | 02:14 |
| johnsom | My transition to xenial was a bit rough too | 02:16 |
| leitan | johnsom: running like a charm on 14.04 | 02:22 |
| *** links has joined #openstack-lbaas | 02:27 | |
| *** links has quit IRC | 02:30 | |
| *** links has joined #openstack-lbaas | 02:31 | |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add pool commands to client https://review.openstack.org/461992 | 02:31 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron-lbaas-dashboard master: Updated from global requirements https://review.openstack.org/471185 | 02:40 |
| *** cpuga has joined #openstack-lbaas | 02:41 | |
| *** sanfern has joined #openstack-lbaas | 02:44 | |
| leitan | Anyone knows if on glance v2 is there a method to update the image file, keeping the original id, etc ? | 02:55 |
| leitan | want to update the amphorae image without changhing the api config | 02:55 |
| johnsom | leitan We use a glance "tag" now instead of an ID, so you can set the tag in the config and just move the tag to the new image in glance | 03:02 |
| leitan | johnsom: yes i configured the tag, totally forgot that, thats the best way to promote it thanks | 03:03 |
| leitan | thanks !! | 03:03 |
| *** yamamoto_ has joined #openstack-lbaas | 03:25 | |
| *** cpuga_ has joined #openstack-lbaas | 03:28 | |
| *** cpuga has quit IRC | 03:28 | |
| *** gans has joined #openstack-lbaas | 03:35 | |
| *** cpuga_ is now known as cpuga | 03:36 | |
| cpuga | johnsom: I've updated flavor spec (https://review.openstack.org/#/c/392485/) patch so that index.rst has link to spec. | 03:39 |
| cpuga | johnsom: also responded to previous comments from previous patchset. | 03:39 |
| cpuga | johnsom: let me know if you have any questions or suggestions, we can discuss. | 03:39 |
| *** sanfern has quit IRC | 04:04 | |
| *** sanfern has joined #openstack-lbaas | 04:14 | |
| *** armax has joined #openstack-lbaas | 04:44 | |
| *** leitan has quit IRC | 04:44 | |
| *** leitan has joined #openstack-lbaas | 04:45 | |
| *** leitan has quit IRC | 04:49 | |
| *** gongysh has quit IRC | 04:50 | |
| *** yamamoto_ has quit IRC | 05:07 | |
| *** cpuga has quit IRC | 05:15 | |
| *** gcheresh has joined #openstack-lbaas | 05:23 | |
| *** gongysh has joined #openstack-lbaas | 05:30 | |
| *** yamamoto has joined #openstack-lbaas | 05:51 | |
| *** aojea has joined #openstack-lbaas | 05:55 | |
| *** reedip has quit IRC | 06:00 | |
| *** cpuga has joined #openstack-lbaas | 06:00 | |
| *** kobis has joined #openstack-lbaas | 06:06 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron-lbaas-dashboard master: Updated from global requirements https://review.openstack.org/471215 | 06:08 |
| *** rcernin has joined #openstack-lbaas | 06:10 | |
| *** aojea has quit IRC | 06:15 | |
| *** aojea has joined #openstack-lbaas | 06:15 | |
| *** aojea has quit IRC | 06:16 | |
| *** aojea has joined #openstack-lbaas | 06:16 | |
| *** cpuga has quit IRC | 06:22 | |
| *** cpuga has joined #openstack-lbaas | 06:23 | |
| *** reedip has joined #openstack-lbaas | 06:33 | |
| *** pcaruana has joined #openstack-lbaas | 06:39 | |
| *** yamamoto has quit IRC | 06:46 | |
| *** yamamoto has joined #openstack-lbaas | 06:47 | |
| *** cpuga has quit IRC | 06:48 | |
| *** tesseract has joined #openstack-lbaas | 07:13 | |
| *** aojea has quit IRC | 07:24 | |
| *** aojea has joined #openstack-lbaas | 07:25 | |
| *** armax has quit IRC | 07:29 | |
| *** aojea has quit IRC | 07:29 | |
| *** eanderson_ has joined #openstack-lbaas | 07:44 | |
| openstackgerrit | Erik Olof Gunnar Andersson proposed openstack/neutron-lbaas-dashboard master: Replace SortedDict with OrderedDict https://review.openstack.org/471241 | 07:53 |
| eanderson_ | Just fixing the gate with the above patch ^ | 07:59 |
| *** links has quit IRC | 08:02 | |
| *** gcheresh has quit IRC | 08:06 | |
| *** gcheresh_ has joined #openstack-lbaas | 08:06 | |
| openstackgerrit | Erik Olof Gunnar Andersson proposed openstack/neutron-lbaas-dashboard master: Replace SortedDict with OrderedDict https://review.openstack.org/471241 | 08:06 |
| *** links has joined #openstack-lbaas | 08:19 | |
| *** eanderson_ has quit IRC | 08:21 | |
| *** gcheresh_ has quit IRC | 08:34 | |
| *** gcheresh_ has joined #openstack-lbaas | 08:34 | |
| *** aojea has joined #openstack-lbaas | 09:02 | |
| openstackgerrit | Ngo Quoc Cuong proposed openstack/octavia master: Remove deprecated oslo_messaging.get_transport https://review.openstack.org/471272 | 09:11 |
| *** links has quit IRC | 09:25 | |
| *** links has joined #openstack-lbaas | 09:43 | |
| *** gongysh has quit IRC | 10:26 | |
| *** sanfern has quit IRC | 10:51 | |
| *** yamamoto has quit IRC | 10:52 | |
| *** gans has quit IRC | 10:53 | |
| *** rcernin has quit IRC | 11:52 | |
| *** pcaruana has quit IRC | 11:53 | |
| *** yamamoto has joined #openstack-lbaas | 11:56 | |
| *** belharar has joined #openstack-lbaas | 12:00 | |
| *** rcernin has joined #openstack-lbaas | 12:07 | |
| *** atoth has joined #openstack-lbaas | 12:07 | |
| *** pcaruana has joined #openstack-lbaas | 12:09 | |
| *** aojea has quit IRC | 12:14 | |
| *** sanfern has joined #openstack-lbaas | 12:17 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron-lbaas-dashboard master: Updated from global requirements https://review.openstack.org/471329 | 12:21 |
| *** yamamoto has quit IRC | 12:21 | |
| *** leitan has joined #openstack-lbaas | 12:26 | |
| *** leitan_ has joined #openstack-lbaas | 12:34 | |
| *** leitan has quit IRC | 12:34 | |
| *** yamamoto has joined #openstack-lbaas | 12:38 | |
| *** catintheroof has joined #openstack-lbaas | 12:42 | |
| *** links has quit IRC | 12:56 | |
| *** fnaval has quit IRC | 13:07 | |
| *** fnaval has joined #openstack-lbaas | 13:08 | |
| *** yamamoto has quit IRC | 13:10 | |
| *** yamamoto has joined #openstack-lbaas | 13:11 | |
| xgerman_ | o/ | 13:14 |
| *** sanfern has quit IRC | 13:17 | |
| *** kobis has quit IRC | 13:17 | |
| *** sanfern has joined #openstack-lbaas | 13:17 | |
| *** yamamoto has quit IRC | 13:21 | |
| *** yamamoto has joined #openstack-lbaas | 13:23 | |
| *** cpuga has joined #openstack-lbaas | 13:27 | |
| *** gcheresh_ has quit IRC | 13:31 | |
| *** cpuga has quit IRC | 13:33 | |
| *** pcaruana has quit IRC | 13:34 | |
| *** rcernin has quit IRC | 13:36 | |
| *** aojea has joined #openstack-lbaas | 13:47 | |
| *** rcernin has joined #openstack-lbaas | 13:49 | |
| *** pcaruana has joined #openstack-lbaas | 13:51 | |
| openstackgerrit | Hengqing Hu proposed openstack/octavia master: Implement zero downtime reload with multibinder https://review.openstack.org/466260 | 13:58 |
| *** pcaruana has quit IRC | 13:59 | |
| *** pcaruana has joined #openstack-lbaas | 14:00 | |
| *** rcernin has quit IRC | 14:04 | |
| *** rcernin has joined #openstack-lbaas | 14:05 | |
| *** rcernin has quit IRC | 14:07 | |
| *** rcernin has joined #openstack-lbaas | 14:07 | |
| *** links has joined #openstack-lbaas | 14:12 | |
| *** armax has joined #openstack-lbaas | 14:15 | |
| *** links has quit IRC | 14:17 | |
| *** leitan_ has quit IRC | 14:20 | |
| *** cpuga has joined #openstack-lbaas | 14:21 | |
| *** cpuga has quit IRC | 14:22 | |
| *** cpuga has joined #openstack-lbaas | 14:23 | |
| *** pksingh has joined #openstack-lbaas | 14:23 | |
| pksingh | johnsom: Hello | 14:25 |
| pksingh | johnsom: I was going through octavia code base, I found octavia is not using oslo-versioned-objects, is my understanding correct? | 14:27 |
| pksingh | johnsom: if yes, is there ant specfic reason not to use that? | 14:28 |
| pksingh | sbalukoff_: can you answer my above question? ^^ | 14:30 |
| johnsom | pksingh Two main reasons. 1. They didn't exist when we created our data model. 2. No one has taken the time to convert our datamodel over to OVO. | 14:44 |
| johnsom | I expect it will happen at some point in the future, but we just don't have people working on it now. | 14:44 |
| pksingh | johnsom: I think they will be helpful, would octavia community accept the feature? | 14:45 |
| pksingh | johnsom: if yes then i can submmit a blurprint | 14:45 |
| pksingh | johnsom: ok great | 14:45 |
| johnsom | I don't think there would be push back, so feel free to put in a spec. | 14:46 |
| pksingh | johnsom: and wsme is deprecated, shouldn't we use any alternative like jsonschema? | 14:46 |
| pksingh | johnsom: i registered a BP here, https://blueprints.launchpad.net/octavia/+spec/jsonschema-api-validation | 14:47 |
| johnsom | When was wsme deprecated???? | 14:47 |
| johnsom | Hmm, I will take a look | 14:48 |
| pksingh | johnsom: i am core member of openstack zun(container service), and in past we decided to move to jsonschema | 14:48 |
| pksingh | johnsom: and i work with sanfern in @walmartlabs, he talked a lot about you | 14:49 |
| *** rcernin has quit IRC | 14:54 | |
| *** yamamoto has quit IRC | 14:55 | |
| *** pksingh has quit IRC | 14:58 | |
| *** armax has quit IRC | 14:58 | |
| *** armax has joined #openstack-lbaas | 15:00 | |
| *** kobis has joined #openstack-lbaas | 15:11 | |
| *** kobis has quit IRC | 15:11 | |
| *** tomtomtom has joined #openstack-lbaas | 15:12 | |
| *** blogan has joined #openstack-lbaas | 15:12 | |
| tomtomtom | hello, trying to install octavia in newton using openstack-ansible. I've gotten the playbook/role into newton and am able to deploy it, but I have some questions about octavia setup. Octavia asks for a lbaas-mgmt network, is that network one you create in neutron just like any other openstack network? | 15:16 |
| tomtomtom | It seems to be required is that true? Also, is br-lbaas something you MUST create? | 15:16 |
| *** reedip_ has joined #openstack-lbaas | 15:27 | |
| xgerman_ | tomtomtom the management net is a provider network… I have put in under br-mgmt as an example | 15:28 |
| xgerman_ | you can do it any other way — it is needed to bridge between containers in the control plabe to the vms running the load balancer | 15:28 |
| *** chlong has joined #openstack-lbaas | 15:29 | |
| *** belharar has quit IRC | 15:35 | |
| *** belharar has joined #openstack-lbaas | 15:36 | |
| *** kobis has joined #openstack-lbaas | 15:37 | |
| *** kobis has quit IRC | 15:41 | |
| *** KeithMnemonic1 has quit IRC | 15:53 | |
| *** aojea has quit IRC | 15:58 | |
| tomtomtom | xgerman_ thanks, I hadn't though of just using br-mgmt for this. | 15:58 |
| tomtomtom | that will simplify the whole process | 15:58 |
| *** gans has joined #openstack-lbaas | 16:00 | |
| xgerman_ | yeah, I used br-mgmt before but I have been told most people wanted to keep things separate | 16:06 |
| *** belharar has quit IRC | 16:07 | |
| *** gans has quit IRC | 16:13 | |
| *** gans has joined #openstack-lbaas | 16:17 | |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add Listener commands to client https://review.openstack.org/461231 | 16:18 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add pool commands to client https://review.openstack.org/461992 | 16:18 |
| *** belharar has joined #openstack-lbaas | 16:22 | |
| *** gans has quit IRC | 16:22 | |
| *** yamamoto has joined #openstack-lbaas | 16:26 | |
| *** yamamoto has quit IRC | 16:31 | |
| *** tesseract has quit IRC | 16:32 | |
| *** pcaruana has quit IRC | 16:34 | |
| *** armax has quit IRC | 16:35 | |
| *** armax has joined #openstack-lbaas | 16:38 | |
| *** leitan has joined #openstack-lbaas | 16:41 | |
| *** kobis has joined #openstack-lbaas | 16:42 | |
| *** sanfern has quit IRC | 16:44 | |
| *** sanfern has joined #openstack-lbaas | 16:45 | |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add member commands to client https://review.openstack.org/463035 | 16:48 |
| *** sanfern has quit IRC | 16:50 | |
| *** sanfern has joined #openstack-lbaas | 16:51 | |
| *** sshank has joined #openstack-lbaas | 16:55 | |
| *** reedip_ has quit IRC | 16:58 | |
| *** leitan_ has joined #openstack-lbaas | 17:14 | |
| *** leitan has quit IRC | 17:14 | |
| *** aojea has joined #openstack-lbaas | 17:16 | |
| *** belharar has quit IRC | 17:19 | |
| *** aojea has quit IRC | 17:20 | |
| *** cpuga has quit IRC | 17:27 | |
| *** cpuga has joined #openstack-lbaas | 17:28 | |
| *** kobis has quit IRC | 17:31 | |
| *** cpuga has quit IRC | 17:34 | |
| *** cpuga has joined #openstack-lbaas | 17:34 | |
| *** belharar has joined #openstack-lbaas | 17:42 | |
| *** belharar has quit IRC | 17:46 | |
| *** belharar has joined #openstack-lbaas | 17:46 | |
| *** aojea has joined #openstack-lbaas | 17:50 | |
| eandersson | xgerman_, johnsom pushed a quick patch for the old lbaas dashboard to unblock the gate https://review.openstack.org/#/c/471241/ | 17:58 |
| johnsom | Cool, thanks, I saw that this morning, haven't had a chance to take a look yet | 17:59 |
| *** belharar has quit IRC | 17:59 | |
| johnsom | I need to finish getting octavia-dashboard going too | 18:00 |
| *** fnaval has quit IRC | 18:20 | |
| rm_work | johnsom / JudeC: on member show, the order right now is "loadbalancer member show <member> <pool>" | 18:21 |
| rm_work | is that right? | 18:21 |
| rm_work | it FEELS to me like it should be pool then memberid | 18:22 |
| rm_work | but | 18:22 |
| rm_work | maybe that's just some weird feeling i have right now | 18:22 |
| johnsom | I think I commented that it was different than the old client | 18:22 |
| rm_work | OH | 18:22 |
| rm_work | yeah you made him switch it | 18:22 |
| rm_work | >_> | 18:22 |
| rm_work | I kinda want to comment and make him switch it back | 18:23 |
| johnsom | neutron client is: MEMBER POOL | 18:23 |
| rm_work | we can play core ping-pong | 18:23 |
| openstackgerrit | Merged openstack/neutron-lbaas-dashboard master: Replace SortedDict with OrderedDict https://review.openstack.org/471241 | 18:23 |
| rm_work | ugh | 18:23 |
| rm_work | that doesn't ... | 18:23 |
| rm_work | make sense | 18:23 |
| rm_work | right? | 18:23 |
| rm_work | do we have to be exactly client backwards compatible? >_> | 18:23 |
| *** fnaval has joined #openstack-lbaas | 18:23 | |
| johnsom | Ha, well, I was mostly leaning towards "keep it similar", but if you feel strongly.... | 18:23 |
| rm_work | well | 18:23 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add l7policy commands to client https://review.openstack.org/462431 | 18:23 |
| rm_work | is it possible to have an arg be positional, but ALSO allow it to be keyworded? | 18:24 |
| johnsom | It's odd, I have to say either way IMO | 18:24 |
| rm_work | yeah to be fair if a member is an ID, it should be unique so why specify the pool | 18:24 |
| rm_work | >_> | 18:24 |
| johnsom | I mean logically it should be pool then member | 18:24 |
| johnsom | We take names | 18:24 |
| rm_work | yes | 18:24 |
| rm_work | right | 18:25 |
| johnsom | My other gripe is I think L7policy does --listener | 18:25 |
| *** aojea has quit IRC | 18:25 | |
| johnsom | Which is inconsistent too | 18:25 |
| rm_work | orly | 18:25 |
| * rm_work gets out his pitchfork and torch | 18:25 | |
| *** atoth has quit IRC | 18:25 | |
| rm_work | i'd almost rather have --pool | 18:25 |
| rm_work | on this | 18:25 |
| rm_work | because then it's just not an issue | 18:26 |
| rm_work | dunno | 18:26 |
| rm_work | the way it is now is the WORST option IMO | 18:26 |
| johnsom | Yeah, maybe it wasn't l7, maybe health monitor? looking | 18:26 |
| rm_work | i hate all of this "carry forward dumb stuff because that's how it worked before" | 18:26 |
| johnsom | Yeah, on the client I am much more open to let's fix it now | 18:27 |
| rm_work | yep | 18:27 |
| rm_work | (openstack) loadbalancer member show pool1 member1 | 18:28 |
| rm_work | Unable to locate member1 in pools | 18:28 |
| rm_work | specifically that message is bad too | 18:28 |
| rm_work | "pools"? | 18:28 |
| *** yamamoto has joined #openstack-lbaas | 18:29 | |
| rm_work | i guess "unable to locate pool member1" would make more sense | 18:29 |
| rm_work | i dunno if it's a very generic format though that wouldn't work the same for everything? | 18:29 |
| johnsom | Ah, it was pool create that had "--listener" and <load balancer> | 18:30 |
| johnsom | That one was odd | 18:31 |
| johnsom | I haven't gone back and looked at his new versions yet | 18:31 |
| johnsom | I was just about to start my review of l7rules | 18:31 |
| *** leitan_ has quit IRC | 18:31 | |
| johnsom | Then circle back. | 18:31 |
| eandersson | I like how there are 4 duplicate global req pull requests for the lbaas-dashboard | 18:31 |
| *** leitan has joined #openstack-lbaas | 18:32 | |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add l7policy commands to client https://review.openstack.org/462431 | 18:32 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add Listener commands to client https://review.openstack.org/461231 | 18:32 |
| johnsom | Joy, yeah, that is odd | 18:32 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add member commands to client https://review.openstack.org/463035 | 18:32 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add pool commands to client https://review.openstack.org/461992 | 18:32 |
| johnsom | rm_work Propose what you would like to see. I just commented that it was different than the old client, but that can be ok. | 18:33 |
| rm_work | well would you prefer them to be real args | 18:33 |
| rm_work | err | 18:33 |
| rm_work | positional args | 18:33 |
| *** JudeC has joined #openstack-lbaas | 18:33 | |
| rm_work | or just make --pool | 18:33 |
| rm_work | i could be ok with either | 18:33 |
| rm_work | if we did --pool | 18:33 |
| rm_work | it could try to get member by ID first? without the pool if it's an ID | 18:34 |
| rm_work | oh wait no | 18:34 |
| rm_work | our API doesn't even allow that <_< nm | 18:34 |
| rm_work | (really it should) | 18:34 |
| johnsom | Yeah, let's not get fancy on the lookup logic | 18:34 |
| johnsom | Yet at least | 18:34 |
| rm_work | ok so | 18:34 |
| *** yamamoto has quit IRC | 18:34 | |
| JudeC | Sorry I am here. | 18:34 |
| rm_work | member show <pool> <member> | 18:34 |
| rm_work | or | 18:34 |
| rm_work | member show --pool <pool> member | 18:35 |
| johnsom | So positional or mandatory --pool | 18:35 |
| rm_work | yep | 18:35 |
| johnsom | xgerman_ Have an opinion on this? | 18:35 |
| rm_work | otherwise i'm ready to +2 everything again | 18:35 |
| rm_work | though i didn't try every single possible flag/option | 18:35 |
| johnsom | I'm still going to circle back and re-review (though should be faster) | 18:36 |
| rm_work | i assume we can take bug reports for that stuff if individual random things are broken | 18:36 |
| rm_work | as long as the main happy path works | 18:36 |
| xgerman_ | I think —pool is what we had before? positional can get confusing fast | 18:36 |
| johnsom | Yeah, the code is pretty simple really | 18:36 |
| johnsom | https://www.irccloud.com/pastebin/8TtGjjTb/ | 18:36 |
| johnsom | That is what we had before | 18:36 |
| rm_work | xgerman_: i'm leaning towards --pool | 18:37 |
| rm_work | just to make it not even a question | 18:37 |
| JudeC | To do the show for the member from the API dont you need to know what the pool is? | 18:37 |
| johnsom | I lean towards --pool <pool> myself, now that I look at it more | 18:37 |
| rm_work | yep | 18:37 |
| rm_work | JudeC: you do, unfortunately | 18:37 |
| johnsom | Yeah, it is mandatory | 18:37 |
| rm_work | yes | 18:37 |
| rm_work | mandatory --pool | 18:37 |
| xgerman_ | +1 | 18:37 |
| rm_work | instead of positional | 18:37 |
| JudeC | Oh | 18:37 |
| JudeC | Easy | 18:37 |
| rm_work | yep | 18:37 |
| JudeC | I will make the change right now | 18:38 |
| rm_work | i'm gonna go +2 the chain | 18:38 |
| rm_work | up to member | 18:38 |
| johnsom | Yeah, so, that is quorum in my book. We should switch it to mandatory --pool | 18:38 |
| JudeC | I think I fixed that issue rm_work where the latest listener patch wasnt in the whole chain. | 18:38 |
| johnsom | I will circle back and +2 today if it all looks good. Just want to do my first pass on l7rules | 18:38 |
| rm_work | JudeC: no worries they're about to all merge anyway I hope | 18:38 |
| *** atoth has joined #openstack-lbaas | 18:39 | |
| johnsom | JudeC Thank you for your patience on this, it's awesome to have! | 18:39 |
| *** atoth has quit IRC | 18:39 | |
| *** atoth has joined #openstack-lbaas | 18:40 | |
| openstackgerrit | Michael Johnson proposed openstack/neutron-lbaas-dashboard master: Updated from global requirements https://review.openstack.org/471329 | 18:40 |
| johnsom | Rebase that onto the fix we just merged | 18:40 |
| johnsom | rm_work did you comment on the member patch? Did we capture that decision? | 18:41 |
| JudeC | johnsom: :) I am just happy I can see the light at the end of the tunnel on this. :) | 18:41 |
| rm_work | i didn't comment, he's just doing it i think lol | 18:41 |
| johnsom | Ok, well hopefully all of my comments made sense. | 18:42 |
| rm_work | there, commented | 18:42 |
| JudeC | Johnsom: Yeah they are great, made fixes so incredibly easy. | 18:42 |
| *** JudeC has quit IRC | 18:47 | |
| *** JudeC has joined #openstack-lbaas | 18:47 | |
| openstackgerrit | Merged openstack/python-octaviaclient master: Updated from global requirements https://review.openstack.org/471182 | 18:47 |
| *** sshank has quit IRC | 18:49 | |
| johnsom | rm_work for list commands, also --pool? | 18:54 |
| johnsom | I think we would need to | 18:54 |
| *** pcaruana has joined #openstack-lbaas | 18:56 | |
| rm_work | hmmmmmm | 19:00 |
| rm_work | would that then have no positional | 19:00 |
| rm_work | but i guess that's standard for lists | 19:00 |
| johnsom | Yeah, it's a bit odd, but consistent with the other commands | 19:00 |
| *** aojea has joined #openstack-lbaas | 19:02 | |
| johnsom | SOB, we have some kind of bug. Another project_id is now NULL | 19:04 |
| johnsom | hmmm | 19:05 |
| johnsom | I think it is related to going through neutron though | 19:05 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add loadbalancer commands to client https://review.openstack.org/454516 | 19:08 |
| rm_work | i seriously don't know how | 19:11 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add l7policy commands to client https://review.openstack.org/462431 | 19:13 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add Listener commands to client https://review.openstack.org/461231 | 19:13 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add member commands to client https://review.openstack.org/463035 | 19:13 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add pool commands to client https://review.openstack.org/461992 | 19:13 |
| JudeC | ok | 19:13 |
| JudeC | that should fix the merge conflict | 19:13 |
| johnsom | JudeC ok, done with my first pass over the patches. Will circle back and start at the top after a lunch break. | 19:27 |
| JudeC | Awesome ok :) | 19:27 |
| JudeC | Im going to get food too | 19:27 |
| openstackgerrit | Michael Johnson proposed openstack/python-octaviaclient master: Add loadbalancer commands to client https://review.openstack.org/454516 | 19:29 |
| *** harlowja has quit IRC | 19:29 | |
| johnsom | Fixed a conflicting requirement | 19:29 |
| *** yamamoto has joined #openstack-lbaas | 19:31 | |
| *** pcaruana has quit IRC | 19:33 | |
| *** yamamoto has quit IRC | 19:36 | |
| *** harlowja has joined #openstack-lbaas | 19:50 | |
| *** kobis has joined #openstack-lbaas | 19:54 | |
| *** gcheresh_ has joined #openstack-lbaas | 19:57 | |
| rm_work | augh flake8 is driving me crazy | 19:58 |
| *** armax has quit IRC | 20:07 | |
| *** sshank has joined #openstack-lbaas | 20:15 | |
| *** leitan_ has joined #openstack-lbaas | 20:25 | |
| *** leitan has quit IRC | 20:25 | |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add pool commands to client https://review.openstack.org/461992 | 20:30 |
| *** gcheresh_ has quit IRC | 20:31 | |
| *** aojea has quit IRC | 20:32 | |
| *** yamamoto has joined #openstack-lbaas | 20:32 | |
| *** aojea has joined #openstack-lbaas | 20:35 | |
| *** aojea has quit IRC | 20:36 | |
| *** yamamoto has quit IRC | 20:37 | |
| *** aojea has joined #openstack-lbaas | 20:37 | |
| *** armax has joined #openstack-lbaas | 20:48 | |
| *** leitan_ has quit IRC | 20:51 | |
| *** sshank has quit IRC | 20:53 | |
| *** sshank has joined #openstack-lbaas | 20:53 | |
| johnsom | JudeC Did you test that set and delete aren't returning values now? I'm testing loadbalnacer but I am still getting output. It could just be my environment didn't update right, so double checking | 21:06 |
| JudeC | Yeah I removed the return, although I admit I hadn't explicitly tested it yet. | 21:06 |
| johnsom | Ok, give it a go and let me know if it is just my venv being strange or not | 21:07 |
| JudeC | stack@ubuntu-linux:~$ openstack loadbalancer set f6aa66bc-5eff-485f-a587-96769e5aef15 --name gg | 21:07 |
| JudeC | stack@ubuntu-linux:~$ | 21:07 |
| JudeC | lgtm | 21:07 |
| JudeC | let me try delete | 21:08 |
| johnsom | Ok, probably just a messed up venv on my side. I will build a fresh one | 21:08 |
| JudeC | yeah tested delete as well looks good on my end. | 21:10 |
| johnsom | Yeah, nuked the venv, recreated, now it picked up the new code | 21:11 |
| rm_work | IHOUWBC&#VRGCO#&ynex23ufnw78ug36w978br | 21:12 |
| rm_work | flake8 is going to make me absolutely batshit crazy | 21:12 |
| johnsom | Time to change that password... Grin, JK | 21:12 |
| rm_work | that'd be a pretty ridiculous password to ever type :P | 21:12 |
| rm_work | but i guess that's what password managers are for lol | 21:13 |
| johnsom | Yep | 21:13 |
| johnsom | Looks like some of mine | 21:13 |
| *** kobis has quit IRC | 21:14 | |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add l7policy commands to client https://review.openstack.org/462431 | 21:16 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add Listener commands to client https://review.openstack.org/461231 | 21:16 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add member commands to client https://review.openstack.org/463035 | 21:16 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add pool commands to client https://review.openstack.org/461992 | 21:16 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add l7rule commands to client https://review.openstack.org/462808 | 21:16 |
| JudeC | ^sorry lol | 21:16 |
| rm_work | this is stupid and doesn't matter and i don't need to be working on it, but at this point if i don't get it working i may as well give up for the day anyway because i'm so pissed off | 21:16 |
| JudeC | rm_work: whats wrong buddy | 21:17 |
| *** blogan_ has joined #openstack-lbaas | 21:17 | |
| JudeC | oh | 21:17 |
| JudeC | i need to learn to read | 21:17 |
| johnsom | rm_work Do you want us to look at something? | 21:18 |
| *** blogan has quit IRC | 21:18 | |
| johnsom | Other than the push that just happened (another quick review) I am going to start +2's on OSC. It looks good to me now | 21:18 |
| johnsom | Well, the fixed parts | 21:19 |
| rm_work | k | 21:19 |
| rm_work | ugh maybe in a sec | 21:20 |
| *** cpuga has quit IRC | 21:20 | |
| rm_work | augh ok f*@$ the flake8 docs | 21:24 |
| openstackgerrit | Merged openstack/python-octaviaclient master: Add install information to readme.rst https://review.openstack.org/447527 | 21:30 |
| *** yamamoto has joined #openstack-lbaas | 21:33 | |
| *** yamamoto has quit IRC | 21:37 | |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add healthmonitor commands to client https://review.openstack.org/463830 | 21:49 |
| rm_work | GOT IT | 21:55 |
| rm_work | ok | 21:55 |
| *** sshank has quit IRC | 21:57 | |
| *** aojea has quit IRC | 22:03 | |
| openstackgerrit | Adam Harwell proposed openstack/python-octaviaclient master: Add hacking checks https://review.openstack.org/471168 | 22:03 |
| openstackgerrit | Adam Harwell proposed openstack/python-octaviaclient master: Add check O346 to disallow backslash line continuation https://review.openstack.org/471499 | 22:03 |
| rm_work | suck it, backslashes | 22:03 |
| *** sshank has joined #openstack-lbaas | 22:04 | |
| xgerman_ | looks like our namespace driver has a bug and returns 409 bt syill creates the pool | 22:05 |
| *** blogan_ has quit IRC | 22:10 | |
| *** blogan__ has joined #openstack-lbaas | 22:10 | |
| *** leitan has joined #openstack-lbaas | 22:17 | |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Add check O346 to disallow backslash line continuation https://review.openstack.org/471503 | 22:17 |
| * rm_work shakes his fist at backslashes | 22:19 | |
| *** leitan_ has joined #openstack-lbaas | 22:21 | |
| *** leitan has quit IRC | 22:21 | |
| JudeC | I think that rebase made me delete all my changes to memebers saldkjfsadfsadflsadkjf | 22:21 |
| JudeC | members* | 22:21 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Update help text for all commands https://review.openstack.org/466441 | 22:24 |
| openstackgerrit | Merged openstack/python-octaviaclient master: Add loadbalancer commands to client https://review.openstack.org/454516 | 22:25 |
| openstackgerrit | Merged openstack/python-octaviaclient master: Add Listener commands to client https://review.openstack.org/461231 | 22:25 |
| openstackgerrit | Merged openstack/python-octaviaclient master: Add pool commands to client https://review.openstack.org/461992 | 22:26 |
| rm_work | it might have | 22:32 |
| rm_work | check your reflog | 22:32 |
| *** yamamoto_ has joined #openstack-lbaas | 22:34 | |
| *** yamamoto_ has quit IRC | 22:40 | |
| *** leitan_ has quit IRC | 22:41 | |
| *** catintheroof has quit IRC | 22:41 | |
| *** leitan has joined #openstack-lbaas | 22:43 | |
| *** KeithMnemonic has joined #openstack-lbaas | 22:44 | |
| *** cpuga has joined #openstack-lbaas | 22:46 | |
| rm_work | johnsom: https://review.openstack.org/#/c/471168/ | 22:46 |
| rm_work | JudeC: you find your missing changes? :/ | 22:47 |
| JudeC | nah just doing them again. lol | 22:48 |
| rm_work | T_T | 22:48 |
| JudeC | I rebased and then didnt commit becuase I am dumb | 22:48 |
| JudeC | I didnt miss too much so it NBD | 22:48 |
| rm_work | ok | 22:48 |
| johnsom | Ah good, it was Alex that ignored our own hacking rule. Thought it might have been some brain dead thing I did | 22:53 |
| rm_work | no it was ACTUALLY a braindead thing I did, as much as I'd like to pin it on him | 22:53 |
| rm_work | I was standing over his shoulder and only half paying attention, and told him to just add the ignore | 22:53 |
| rm_work | didn't notice until now that it was our own check :P | 22:53 |
| rm_work | (in the HP office in Seattle during a midcycle) | 22:54 |
| rm_work | I remember the exact moment, and it haunts me | 22:54 |
| johnsom | Yeah, it was totally during a midcycle in Seattle. The only time we let him touch the code... Grin | 22:54 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Add check O346 to disallow backslash line continuation https://review.openstack.org/471503 | 22:54 |
| rm_work | lol | 22:54 |
| rm_work | i forgot to add the test | 22:54 |
| *** leitan has quit IRC | 22:57 | |
| rm_work | I'm tempted to propose it as an optional / default-disabled check to hacking proper | 22:57 |
| johnsom | rm_work what is up with the releasenotes change here: https://review.openstack.org/#/c/471168/ | 22:59 |
| rm_work | https://review.openstack.org/#/c/471168/2/octaviaclient/hacking/checks.py | 23:00 |
| rm_work | see line 98 | 23:00 |
| rm_work | this matches the same file in octavia | 23:00 |
| johnsom | ha | 23:00 |
| johnsom | That is all templated | 23:00 |
| rm_work | we already made the change there | 23:00 |
| rm_work | :P | 23:00 |
| rm_work | well, it's our own check | 23:01 |
| rm_work | so that makes sense | 23:01 |
| johnsom | yeah | 23:01 |
| rm_work | just checking, now that I spent practically my whole day fighting flake8 -- you're going to +2 a "no backslash" check, right? :P | 23:01 |
| * johnsom Wonders if he should push rm_work over the edge or not..... | 23:02 | |
| rm_work | if not, i might be done for the day, lol | 23:02 |
| * rm_work throws laptop out window | 23:02 | |
| johnsom | I'm looking at it. Of course want to see the test results. | 23:02 |
| rm_work | you can pull it down and fake a fail quickly | 23:02 |
| rm_work | and see what it shows | 23:02 |
| rm_work | since you will only see passes and that doesn't prove anything :P | 23:03 |
| johnsom | Yeah, I probably will, though the regex is pretty simple too | 23:03 |
| rm_work | yep | 23:03 |
| rm_work | so, i was going to check whitespace after the slash | 23:03 |
| rm_work | but | 23:03 |
| rm_work | that gets caught by TWO other pep checks | 23:03 |
| rm_work | so i think it's fine :P | 23:03 |
| johnsom | That should be caught by other errors | 23:03 |
| rm_work | yes | 23:03 |
| rm_work | W291 trailing whitespace | 23:04 |
| rm_work | E901 SyntaxError: unexpected character after line continuation character | 23:04 |
| rm_work | E113 unexpected indentation | 23:04 |
| rm_work | all of those happen if you add even a single space after it, lol | 23:04 |
| johnsom | I could ask that you fix the rest of the unit tests for hacking since you are adding it to dashboard | 23:04 |
| rm_work | so three actually | 23:04 |
| johnsom | Or, I mean client | 23:04 |
| rm_work | >_< | 23:04 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add member commands to client https://review.openstack.org/463035 | 23:04 |
| johnsom | But that has been broken for a long time | 23:04 |
| rm_work | yes | 23:04 |
| rm_work | copied directly from octavia | 23:05 |
| johnsom | garbage in, garbage out | 23:05 |
| rm_work | they're still in sync, too, assuming you approve the octavia patch for this | 23:05 |
| rm_work | note pep8 has already passed in zuul for the octavia one | 23:06 |
| johnsom | http://logs.openstack.org/03/471503/2/check/octavia-coverage-ubuntu-xenial/66884c6/cover/octavia_hacking_checks_py.html | 23:06 |
| *** leitan has joined #openstack-lbaas | 23:06 | |
| johnsom | Yeah, still 48% | 23:06 |
| rm_work | lol yes, thanks, i see that. | 23:06 |
| rm_work | T_T | 23:06 |
| rm_work | great work for someone looking for an easy task :P | 23:06 |
| *** leitan has quit IRC | 23:07 | |
| johnsom | Yeah | 23:07 |
| johnsom | It's still not as bad as the "controller_simulator" at 14% | 23:07 |
| *** leitan has joined #openstack-lbaas | 23:08 | |
| rm_work | >_> | 23:08 |
| rm_work | JudeC: you missed one | 23:09 |
| rm_work | so close | 23:09 |
| JudeC | eff | 23:09 |
| JudeC | I knew there was someing I missed | 23:10 |
| JudeC | I cant even type right now. | 23:10 |
| rm_work | lol | 23:10 |
| rm_work | today is ... a day | 23:10 |
| rm_work | i have accomplished almost nothing | 23:10 |
| johnsom | I hate those days, you start out with a plan, but at the end, nothing | 23:11 |
| JudeC | Question | 23:12 |
| JudeC | Do we want to make make all of the pool arguments no longer positional for memeber | 23:12 |
| rm_work | uhh | 23:12 |
| rm_work | what else uses it | 23:12 |
| rm_work | show... | 23:12 |
| JudeC | I know we wanted to do it on show | 23:12 |
| rm_work | list? | 23:12 |
| rm_work | create | 23:12 |
| rm_work | i guess they all use it, huh. | 23:13 |
| JudeC | set | 23:13 |
| johnsom | That is what I asked earlier. I think we need to for consistency | 23:13 |
| rm_work | yes, probably all of them | 23:13 |
| JudeC | set is the one that does the double positional | 23:13 |
| rm_work | ah right | 23:13 |
| rm_work | yeah | 23:13 |
| rm_work | i'd say everything | 23:13 |
| JudeC | kk | 23:13 |
| *** leitan has quit IRC | 23:14 | |
| johnsom | JudeC Ping me when you have the reset of the commands ready for re-review. | 23:16 |
| JudeC | ok | 23:16 |
| johnsom | Pretty happy we merged a few today, that is excellent | 23:16 |
| rm_work | were l7/hm not ready yet? | 23:16 |
| rm_work | i'm HOPING we can merge members today | 23:17 |
| rm_work | that'd make us able to actually create a working minimal LB | 23:17 |
| johnsom | reset, oye, yeah, I'm tired too. I meant rest. | 23:18 |
| rm_work | i think we read it the way you meant it :) | 23:19 |
| johnsom | rm_work are you going to be able to make the meeting tomorrow? | 23:19 |
| rm_work | I should be able to, yes | 23:19 |
| rm_work | and it's at... 1pm right? | 23:19 |
| rm_work | so I can plan to actually be home this time | 23:19 |
| johnsom | Cool, yeah. I have an agenda up. A few blueprints to talk about | 23:19 |
| rm_work | k | 23:19 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron-lbaas-dashboard master: Updated from global requirements https://review.openstack.org/471514 | 23:22 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-octaviaclient master: Updated from global requirements https://review.openstack.org/471515 | 23:26 |
| johnsom | Oye, now there are 5 GR patches on lbaas-dashboard | 23:27 |
| rm_work | I think maybe ONE has to land? | 23:30 |
| rm_work | let's merge one really quick... | 23:31 |
| johnsom | Github is down | 23:31 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add member commands to client https://review.openstack.org/463035 | 23:31 |
| *** marcin12345 has quit IRC | 23:31 | |
| *** cpuga has quit IRC | 23:33 | |
| rm_work | wait seriously, github is down? | 23:33 |
| johnsom | https://status.github.com/ | 23:33 |
| johnsom | Fall down, go boom | 23:33 |
| *** cpuga has joined #openstack-lbaas | 23:33 | |
| rm_work | uergh | 23:34 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add l7policy commands to client https://review.openstack.org/462431 | 23:34 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Update help text for all commands https://review.openstack.org/466441 | 23:34 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add healthmonitor commands to client https://review.openstack.org/463830 | 23:34 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add l7rule commands to client https://review.openstack.org/462808 | 23:34 |
| johnsom | Time to catch up on openstack-dev email and call it a day | 23:34 |
| rm_work | >_< | 23:35 |
| JudeC | I think everything is good now johnsom. I hope.... | 23:36 |
| JudeC | Feel a little bit like a chicken with its head cut off today. | 23:36 |
| *** yamamoto_ has joined #openstack-lbaas | 23:36 | |
| johnsom | Ok, will have a look (probably tomorrow) | 23:37 |
| JudeC | no problem! | 23:37 |
| JudeC | thank you again for the detailed comments. | 23:37 |
| JudeC | you to rm_work | 23:37 |
| JudeC | too | 23:37 |
| *** cpuga has quit IRC | 23:37 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron-lbaas-dashboard master: Updated from global requirements https://review.openstack.org/471516 | 23:38 |
| *** cpuga has joined #openstack-lbaas | 23:41 | |
| *** yamamoto_ has quit IRC | 23:42 | |
| rm_work | johnsom: ok can you +A this: https://review.openstack.org/#/c/471516/1 | 23:42 |
| rm_work | johnsom: github back up | 23:42 |
| rm_work | abandoned all the others | 23:43 |
| johnsom | Thing is, those changes are already in the repo | 23:43 |
| rm_work | some were | 23:43 |
| rm_work | these don't seem to be | 23:43 |
| rm_work | the first patch was a null-change | 23:43 |
| rm_work | but | 23:43 |
| rm_work | this seems to be valid | 23:43 |
| rm_work | no? | 23:43 |
| johnsom | It was null because it was rebased | 23:44 |
| rm_work | hmmmmmm | 23:44 |
| rm_work | so if we try to merge this, nothing will happen? | 23:44 |
| johnsom | I have no idea, but yeah, I guess.... | 23:45 |
| johnsom | I did for giggles | 23:45 |
| rm_work | it's proposing against the 3.0.0.0b1 tag | 23:46 |
| rm_work | which is the latest tag | 23:47 |
| rm_work | is there a setting where you added this, such that it proposes against the latest release or something? | 23:48 |
| johnsom | Yeah, but it should be master with a branch of master | 23:48 |
| johnsom | Not that I know of. To add the project you just add it to a text file. It's been there for a long time, it's not one I just added | 23:48 |
| rm_work | hmm | 23:48 |
| *** leitan has joined #openstack-lbaas | 23:55 | |
| rm_work | JudeC: posted a comment, get weird output on member show | 23:58 |
| rm_work | same on member set | 23:58 |
| rm_work | ahhh actually i realized a weirdness | 23:58 |
| rm_work | johnsom: so note that on "member set", everything that's a --arg is normally "to be set" | 23:59 |
| rm_work | ... now except "--pool" | 23:59 |
| rm_work | because obviously (to us) you can't change the member's pool | 23:59 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!