Earlier  
Later
New Search
Posted Nick Remark
#openstack-sdks - 2018-07-27
16:02:56 mordred yah. I hear you
16:03:20 mordred corvus: openstacksdk makes more sense than shade did - mainly because it actually is used inside of some of the openstack services
16:03:50 corvus mordred: where's the release going to come from?
16:04:02 Shrews mordred: +2/+A and +1 (can't +2 stable)
16:04:09 mordred oh - I should fix that
16:04:41 mordred corvus, Shrews: you've both been added to openstacksdk-stable-maint
16:04:57 mordred corvus: the 0.17.1 release will get cut from the stable/rocky branch
16:05:21 corvus mordred: when do you start cutting releases from master again?
16:05:44 corvus (also, if this is documented somewhere, feel free to tell me to rtfm)
16:05:54 mordred once all of the server projects have their stable/rocky branches cut and the freeze is lifted
16:06:10 corvus ok. kinda sorta makes sense.
16:06:24 corvus i'll stop asking questions to reduce the risk of that changing. :)
16:06:33 mordred corvus: smcginnis sent out an email with the subject "Release countdown for week R-4" if you want to read more
16:07:12 mordred corvus: it's definitely taken some getting used to - I still don't have my head fully wrapped around it - but I'm doing better this cycle than last cycle
16:11:56 mordred corvus, Shrews: also - I think I want to remove the auto-creation of container from the sdk create_object call - corvus' work here shows it's actually not a great idea
16:12:08 mordred but for backwards compat, we'll keep it in the shade version
16:12:17 mordred sound ok to y'all?
16:12:48 Shrews mordred: yeah, though i'm curious why it was there in there in the 1st place. didn't i comment that bit?
16:13:19 Shrews # On some clouds this is not necessary. On others it is. I'm confused.
16:13:33 Shrews i guess i'm still confused
16:13:35 Shrews :)
16:13:37 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Add method for returning a raw response for an object https://review.openstack.org/586262
16:13:43 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Add support for streaming object responses https://review.openstack.org/586250
16:14:06 mordred Shrews: so say we all
16:34:03 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Remove the auto-creation of containers in create_object https://review.openstack.org/586616
16:51:57 openstackgerrit Akihiro Motoki proposed openstack/osc-lib master: Revert "Fixes a bug with issubclass() being called on an instance" https://review.openstack.org/586617
16:53:28 amotoki dtroyer: mordred: osc-lib 1.11.0 release breaks OSC plugins (at least neutronclient OSC plugin). the above revert will fix it. could you check it as you two reviewed the reverted patch?
17:22:14 openstackgerrit Dmitry Tantsur proposed openstack/openstacksdk master: Add a simple baremetal functional job https://review.openstack.org/586574
19:37:07 openstackgerrit Merged openstack/openstacksdk master: Update create_object to handled chunked data https://review.openstack.org/585532
20:44:42 openstackgerrit Hongbin Lu proposed openstack/python-openstackclient master: [WIP] Support enable/disable uplink status propagation https://review.openstack.org/586684
21:04:55 openstackgerrit Hongbin Lu proposed openstack/openstacksdk master: [WIP] Add propagate_uplink_status to port https://review.openstack.org/586687
23:12:02 mordred amotoki: whoops! the issubclass/isinstance swap was drivenby python 3.7 being more strict about using isinstance. but I agree - this case looks like we read the code wrong and that does look like it's a class
23:12:43 mordred now - need to look at why that functional test failed
23:15:52 openstackgerrit Merged openstack/openstacksdk master: Support for microversions in base Resource https://review.openstack.org/583491
23:39:13 mordred amotoki: it looks like osc functional tests got broken by glance v1 not being available
23:39:26 mordred http://logs.openstack.org/17/586617/1/check/osc-functional-devstack/7218285/testr_results.html.gz
23:50:59 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Add method for returning a raw response for an object https://review.openstack.org/586262
23:50:59 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Add support for streaming object responses https://review.openstack.org/586250
23:51:00 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Remove the auto-creation of containers in create_object https://review.openstack.org/586616
23:51:00 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Update config doc url to point to openstacksdk https://review.openstack.org/586726
23:51:01 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Handle image and object key metadata for shade transition https://review.openstack.org/586727
23:52:39 openstackgerrit Monty Taylor proposed openstack-infra/shade master: Defer object interactions to openstacksdk https://review.openstack.org/586728
23:57:10 openstackgerrit Monty Taylor proposed openstack-infra/shade master: Defer object interactions to openstacksdk https://review.openstack.org/586728
#openstack-sdks - 2018-07-28
15:23:14 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Remove deprecated profile and authenticator parameters https://review.openstack.org/586822
15:56:09 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Remove the auto-creation of containers in create_object https://review.openstack.org/586616
15:56:10 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Update config doc url to point to openstacksdk https://review.openstack.org/586726
15:56:10 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Handle image and object key metadata for shade transition https://review.openstack.org/586727
19:31:29 openstackgerrit Vieri proposed openstack/osc-lib master: Update image link https://review.openstack.org/586837
#openstack-sdks - 2018-07-30
02:14:54 chenyb4 hi, mordred
02:25:44 chenyb4 mordred, openstacksdk already supports microversions?
09:27:40 openstackgerrit tianhui proposed openstack/python-openstackclient master: Compute: Add tags support for server https://review.openstack.org/569386
10:46:11 openstackgerrit Dmitry Tantsur proposed openstack/openstacksdk master: Add a simple baremetal functional job https://review.openstack.org/586574
10:49:12 openstackgerrit yanpuqing proposed openstack/openstacksdk master: Add an error message when router add subnet https://review.openstack.org/586996
11:25:13 openstackgerrit Fan Zhang proposed openstack/python-openstackclient master: Remove image api v1 related tests. https://review.openstack.org/587005
11:34:18 openstackgerrit Fan Zhang proposed openstack/python-openstackclient master: Remove volume replication v1 support. https://review.openstack.org/587008
11:39:40 openstackgerrit Fan Zhang proposed openstack/python-openstackclient master: Remove volume replication v1 support. https://review.openstack.org/587008
12:30:02 openstackgerrit Dmitry Tantsur proposed openstack/openstacksdk master: Add a simple baremetal functional job https://review.openstack.org/586574
12:38:03 mordred chenyb4: yes. but we have not exposed microversion features in many places yet
12:38:22 mordred wow. that was bad timing
12:38:56 cdent mordred: you might care about https://review.openstack.org/#/c/586212/
12:42:06 mordred cdent: looks reasonable to me
12:42:41 cdent mordred: the reason I pinged you is because our comments on its antecedent didn't like the fact that it didn't mention designate
12:50:03 mordred cdent: yeah - I still don't like the fact that it doesn't mention designate
12:50:27 cdent mordred: nows your chance!!!!
12:50:28 cdent :)
12:51:27 mordred cdent: well, I think sean made a good point before - which is that there are many places, such as inside of corporatation, where deploying a dns server is unduly hard
12:51:45 mordred I think those places are sad making - but I'm willing to concede they exist
12:52:34 cdent My breain is slowly migrating towards "if this really is _starter_ kit, no cinder"
12:53:17 mordred yeah, I hear that - otoh, the description said 'persistent vm'
12:53:41 mordred and just doing a vm with no volumes is probaby not a great plan for persistent
12:53:56 mordred that said - sure - possibly for starter kit it wants to just be ephemeral vms
12:54:33 mordred my hunch is that people just getting their feet wet are more likely to have workloads that want a thing that looks lke a computer more than they want a thing that behaves like an ephemeral cloud-native workload
12:54:57 mordred especially in a world with docker and kubernetes
12:55:31 mordred but now I'm just getting close to complaining about the tragedy that are floating ips
12:56:12 mordred cdent: dude. I'm still first-coffee - all these words are almost certainly crap
12:56:48 cdent wave off wave off
12:57:00 cdent there is no discussion of floating ips here!
12:57:23 mordred PHEW
12:57:41 mordred cdent: although - if you let me rant about floating ips, it'll get back to dns
12:57:58 cdent Let's do that another time.
12:58:32 cdent I have covered my obligation to ping on this thing and the world is thus a slightly better place.
12:58:40 mordred since part of the problem with the design of openstack is that it's very hard to find the IP you should use to talk to a server - floating vs. fixed, ipv4 vs. ipv6 - and exposing that info via dns would make that much nicer ...
13:00:29 cdent https://www.youtube.com/watch?v=8v-JANIZVF8
13:20:51 mordred dtroyer: whence you awake, it seems we have 2 new osc tips test failures
13:20:58 mordred dtroyer: http://logs.openstack.org/62/586262/6/check/osc-functional-devstack-tips/3dc18eb/testr_results.html.gz
13:21:27 mordred dtroyer: I'm not sure the right way to fix either - the first seems to be that we're getting an os_hidden field back from glanceclient that we're not consuming
13:22:04 mordred the second is just that v1 has gone away - is there an appropriate 'only run this test if the cloud has XXX' cantrip?
13:55:48 samueldmq g'morning
13:55:58 samueldmq does -sdk have meetings in the PTG?
14:03:29 mordred samueldmq: we don't have any scheduled ahead of time - but we can use ptgbot to schedule some when we're there
14:03:57 samueldmq mordred: gotcha, thanks!
14:06:11 openstackgerrit Dmitry Tantsur proposed openstack/openstacksdk master: Add a simple baremetal functional job https://review.openstack.org/586574
14:54:28 mordred corvus, Shrews: if'n you get a few moments today: https://review.openstack.org/#/q/topic:swift-streaming
15:40:42 dtroyer mordred: I think we have to invent that catnip, it'll be a cousin to the nova-net/neutron selector
15:44:20 mordred dtroyer: nod
15:46:35 dtantsur mordred: sup! what's the difference between NotFoundException and ResourceNotFound? :)
15:48:56 dtantsur I'm asking because baremetal.get_node documents raising ResourceNotFound, but raises NotFoundException (its parent)
15:56:25 mordred dtantsur: I can come up with no reason for that - probably mistake from the sdk/shade merge
15:56:46 mordred dtantsur: we should probalby make ResourceNotFound an alias to NotFoundException instead of a subclass
15:57:37 dtantsur maybe, yeah. because the docstrings for at least baremetal are wrong

Earlier   Later