Earlier  
Posted Nick Remark
#openstack-sdks - 2018-04-10
22:38:46 openstackgerrit Merged openstack/os-client-config master: add lower-constraints job https://review.openstack.org/556042
#openstack-sdks - 2018-04-11
04:03:56 openstackgerrit Merged openstack/openstacksdk master: Strip the version prefix from the next link for pagination https://review.openstack.org/558461
08:53:44 frickler amotoki: mordred: dtroyer_zz: these three would seem to close the last gap in DNS support between neutron and OSC https://review.openstack.org/516701 https://review.openstack.org/558820 https://review.openstack.org/558824
09:24:34 jpich dtroyer_zz: Hi! I submitted a couple of release patches for osc stable/ocata as well, if you would have time to give them a look over at some point I would be grateful :) Thank you
11:38:45 openstackgerrit Jacky Hu proposed openstack/openstacksdk master: Allow cascade deletion of load balancer https://review.openstack.org/553404
12:25:45 dtroyer_zz jpich: I'll have a look this afternoon, thanks!
13:01:46 jpich dtroyer_zz: Thank you!
13:05:43 pooja_jadhav mordred: Hi
13:17:39 openstackgerrit Jens Harbott (frickler) proposed openstack/openstacksdk master: WIP: Add functional tests for Neutron DNS extension https://review.openstack.org/559068
13:40:42 pooja_jadhav hi all, any body knows about split logging?
13:41:10 pooja_jadhav I need some help
15:22:36 frickler mugsie: do you happen to know why there is a broken functional recordset test in the sdk? might that be due to the ovo changes? http://logs.openstack.org/68/559068/4/check/openstacksdk-functional-devstack-tips/796096e/testr_results.html.gz
15:22:52 mugsie I freaking hope not
15:22:56 mugsie let me look
15:27:17 frickler mugsie: might also be imported from shade and never tested before. in that patch I enable designate for the functional test for the first time in an attempt to work around an issue with neutron integration
15:27:46 mugsie shade was pretty well tested against designate afaik
15:27:58 frickler mugsie: just a bit disappointed that I got the original issue resolved and now things are still red because of this failure
15:28:04 mugsie for sure
15:28:30 frickler anyway, I'm off to enjoy the remainder of the beautiful weather here, will continue tomorrow
15:28:51 mugsie frickler: enjoy :)
15:32:29 mugsie frickler: I think I found it - let me see if I can get a fix together
15:45:32 openstackgerrit Graham Hayes proposed openstack/openstacksdk master: Fix DNS Recordset CRUD to use IDs https://review.openstack.org/560477
15:45:43 mugsie frickler: ^ should fix it
17:00:49 frickler mugsie: good job, I had only looked at the broken test so far and not yet noticed that the implemented function indeed doesn't live up to what it promises. added comments
17:01:44 mugsie frickler: https://review.openstack.org/#/c/560545/1 was added to test the fix as well :)
17:03:12 frickler mugsie: ah, I didn't see that, marvelous
17:16:13 mugsie gah - that is failing unit tests anyway
#openstack-sdks - 2018-04-12
06:44:57 frickler mugsie: seems the unit test will need added mocking for the new calls you inserted, let me know if you want help with that
07:02:00 mhen Hello!
07:03:17 mhen This channel is also related to the OpenStack client, correct?
07:08:28 openstackgerrit Jens Harbott (frickler) proposed openstack/openstacksdk master: WIP: Add functional tests for Neutron DNS extension https://review.openstack.org/559068
07:11:51 frickler mhen: yes, feel free to ask your real question. we should also consider making the channel topic more inviting
07:12:22 mhen okay
07:20:34 mhen Is there no IRC meeting for the client anymore? I found a schedule and agendas online but the most recent seems to be mid 2017.
07:24:04 mhen background is: I'm trying to implement image signing functionality into the OpenStack client (which currently has to be done manually using 'openssl' commands). I'd like to discuss if this would be worth a contribution and if so, where it would be best to put related crypto code (openstack-sdk or cursive library?).
07:29:58 frickler mhen: I haven't seen an IRC meeting and the schedule on eavesdrop doesn't seem to list one. probably outdated information in the wiki?
07:30:59 frickler mhen: regarding the other question I have no idea, maybe dtroyer_zz or mordred have some idea. but that topic may also better be suited for a discussion on the dev-ml
07:37:04 mhen For reference, this is the meeting schedule I'm referring to: http://eavesdrop.openstack.org/#OpenStackClient_Team_Meeting
07:39:51 mhen frickler, you referring to this I guess? http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
07:41:45 openstackgerrit Jens Harbott (frickler) proposed openstack/openstacksdk master: Fix DNS Recordset CRUD to use IDs https://review.openstack.org/560477
07:42:19 frickler mugsie: fixed a small glitch in your patch first ^^
07:44:41 frickler mhen: you are correct about th mailing list
07:45:46 frickler mhen: and no idea why I missed the meeting in the schedule. probably should get dropped there if it doesn't happen anyway
07:46:12 mhen frickler, okay. Thanks for your guidance! I will try there :)
08:57:50 openstackgerrit Jens Harbott (frickler) proposed openstack/openstacksdk master: Add designate to the services deployed for functional tests https://review.openstack.org/560831
09:01:18 frickler mugsie: FYI ^^ waiting to see what others think about that now
09:17:02 openstackgerrit Graham Hayes proposed openstack/openstacksdk master: Fix DNS Recordset CRUD to use IDs https://review.openstack.org/560477
09:19:02 openstackgerrit Graham Hayes proposed openstack/openstacksdk master: DNM: Fix DNS Recordset CRUD to use IDs https://review.openstack.org/560545
09:19:43 mugsie frickler: cool. ^^ should fix the unit tests, and I reup'd the testing patch on top as well
13:11:57 openstackgerrit Nick Jones proposed openstack/os-client-config master: Remove DataCentred from list of vendors https://review.openstack.org/560894
13:47:22 mordred mugsie, frickler: I'd be fine just adding designate to the normal functional test config for openstacksdk as long as doing so doesn't increase the memory load on the devstack such that stuff starts failing
13:49:02 mordred mugsie, frickler: however, if adding it to the default doesn't work, the senlin folks have a patch up adding a senlin-enabled job, we could also add designate to that (and name it somethin gdifferent)
14:01:10 gtmanfred mordred: have you had a chance to look into the floating ip thing? This is basically all we are doing right now, is create_server with wait=True, and auto_ip=True https://github.com/saltstack/salt/blob/2018.3/salt/cloud/clouds/openstack.py#L644
14:04:09 mugsie mordred: cool - I don't think we will add a huge amount of extra pressure, and chewing up an entire extra node for 3 tests feels like a bad thing :/ - lets hope it fits :)
16:00:46 elmiko cdent around?
16:00:53 cdent dag nabbit
16:14:34 openstackgerrit Merged openstack/api-wg master: Break up the HTTP guideline into smaller documents https://review.openstack.org/554234
16:37:28 edleafe cdent: elmiko: https://etherpad.openstack.org/p/api-sig-newsletter is ready for your review
16:37:37 elmiko edleafe: ack, looking
16:38:19 elmiko edleafe: lgtm
16:38:31 cdent edleafe: yeah, agree, lgtm, short and sweet
16:38:46 edleafe ok, sending it to the clouds...
16:58:39 openstackgerrit Merged openstack/os-client-config master: Remove DataCentred from list of vendors https://review.openstack.org/560894
17:07:43 cdent elmiko, edleafe: the merge of the http split has broken the gate jobs: https://review.openstack.org/#/c/554921/
17:08:04 cdent I'm on the phone right now, but if nobody steps up I'll see if I can make it go
17:12:23 elmiko cdent: does it just need a kick?
17:14:47 elmiko ok, those errors look a little beyond my current knowledge of the system. i think you better check it when you can cdent
17:19:01 cdent elmiko: roger elmiko
17:19:49 elmiko cdent: sorry, wish i could be more helpful =(
17:19:54 cdent no problem
17:23:53 edleafe the error on that doesn't make sense, does it?
17:24:10 elmiko it didn't to me, but then i'm really out of touch with this stuff
17:24:28 edleafe it just says that the docs have to end in .rst or .json
17:26:48 cdent elmiko, edleafe: it is reading the dir 'http' as a file, yes? (still on phone)
17:27:57 elmiko hmm, maybe
17:29:44 edleafe cdent: how did the http patch ever make it through, then?
17:30:26 edleafe oh, it didn't run the py27 tests on that
17:32:03 edleafe Unknow directive type "toctree" - http://logs.openstack.org/21/554921/1/gate/openstack-tox-py27/f3ef842/job-output.txt.gz#_2018-04-12_17_00_26_938269
17:36:08 edleafe Yeah, I just checked that test. It's doing glob.glob("guidelines/*"), which pulls in the 'guidelines/http' entry
17:36:16 edleafe Want me to push a fix?
17:40:32 cdent edleafe: if you can that would be dandy
17:43:05 openstackgerrit Ed Leafe proposed openstack/api-wg master: Fix the checking of directories for extensions https://review.openstack.org/560984
17:43:16 edleafe cdent: ^^
17:43:37 edleafe Do we have to revert the HTTP merge first?
17:43:47 cdent I don't think so
17:44:41 cdent edleafe: I assume you tried it locally and was fine?
17:44:57 edleafe yep
17:46:56 openstackgerrit Jens Harbott (frickler) proposed openstack/openstacksdk master: WIP: Add functional tests for Neutron DNS extension https://review.openstack.org/559068
17:48:56 edleafe cdent: looking at that test, it bugs me that we use the builtin name 'file' as a local variable name. I'll push a fix for that, but no rush on that
17:49:23 cdent yeah, that stuff is super old, predates us I think
17:55:29 openstackgerrit Ed Leafe proposed openstack/api-wg master: Remove use of Python builtin name for variables https://review.openstack.org/560994
17:55:39 edleafe cdent: no rush on this ^^
17:55:50 cdent yay!
17:57:34 edleafe looks like etowes is the culprit, from Idf0dc2aa1fd1e10ed3dfb953838e6ea469c76933
17:58:26 edleafe but cdent +1'd that one! :-P
17:58:57 cdent oh the humanity
17:59:12 cdent but in a tool, it's a good opportunity for a newbie to come along and clean up later...newbie
18:00:42 edleafe I always hated giving interviewees live coding challenges. IMO, they are a terrible predictor of what they know and how they work. Instead, I wrote some code with varying degrees of wrongness buried within.
18:01:24 edleafe I then asked them to review the code, and that showed how well they understood Python better than pretty much anything else I could come up with
18:02:51 cdent that seems a reasonable too

Earlier   Later