Tuesday, 2019-10-01

*** goldyfruit_ has joined #openstack-sdks00:14
*** dave-mccowan has joined #openstack-sdks01:22
*** dave-mccowan has quit IRC03:04
*** gmann_afk is now known as gmann03:33
*** gkadam has joined #openstack-sdks03:55
*** gkadam has quit IRC04:08
*** ricolin has joined #openstack-sdks04:16
*** Luzi has joined #openstack-sdks05:02
*** ttsiouts has joined #openstack-sdks06:35
*** gtema has joined #openstack-sdks06:38
bverschueren_if someone could have a look at https://review.opendev.org/#/c/682909/ ? thx06:47
*** slaweq has joined #openstack-sdks06:56
*** zbr is now known as zbr|ruck07:08
*** tosky has joined #openstack-sdks07:16
*** dtantsur|afk is now known as dtantsur07:23
*** cdent has joined #openstack-sdks07:35
*** ttsiouts has quit IRC07:43
*** jpena|off is now known as jpena07:43
*** ralonsoh has joined #openstack-sdks07:45
*** jpich has joined #openstack-sdks07:52
*** jangutter has quit IRC07:56
*** ttsiouts has joined #openstack-sdks07:57
*** Jeffrey4l has quit IRC07:59
*** Jeffrey4l has joined #openstack-sdks08:00
*** ttsiouts has quit IRC08:12
*** jangutter has joined #openstack-sdks08:14
*** ttsiouts has joined #openstack-sdks08:18
*** yolanda has joined #openstack-sdks08:23
*** e0ne has joined #openstack-sdks08:29
mordredrm_work: yay!08:38
cdentI first read that as "rm work ; yay!" which would be pretty yay.08:41
dtantsurrm -rf08:48
mordredcdent: does the system continue working overall with that? like - if I just rm work, do I still get paychecks?08:51
cdentmordred: in my pretty little mind it does08:52
mordredcdent: I like the workings of that mind08:53
cdentit's a very generous mind08:54
dtantsurjust copy paycheck somewhere before rm -rf08:54
mordreddtantsur: ++08:54
cdentit's a fifo that you can read from as needed08:54
mordreddtantsur: wow. vendor_data2.json08:56
mordreddtantsur: because that's a filename08:56
mordredheaven forbid we put any sort of versioning into the file content itself.08:57
openstackgerritMonty Taylor proposed openstack/keystoneauth master: Fetch discovery documents with auth when needed  https://review.opendev.org/68504209:12
mordredefried, kmalloc, cmurphy: ^^ that's a simpler version and I think should be backportable09:13
*** ricolin has quit IRC09:44
*** ttsiouts has quit IRC10:08
*** ttsiouts has joined #openstack-sdks10:09
*** ttsiouts has quit IRC10:13
openstackgerritMerged openstack/openstacksdk master: Support vendor data in configdrive building  https://review.opendev.org/68384210:58
openstackgerritMerged openstack/openstacksdk master: Add a non-voting ironic-inspector job  https://review.opendev.org/68472911:06
*** dave-mccowan has joined #openstack-sdks11:11
*** ttsiouts has joined #openstack-sdks11:16
*** jpena is now known as jpena|lunch11:38
*** enriquetaso has joined #openstack-sdks12:06
*** gtema has quit IRC12:09
*** cdent has quit IRC12:31
*** ricolin has joined #openstack-sdks12:32
*** cdent has joined #openstack-sdks12:32
*** jpena|lunch is now known as jpena12:38
*** gtema has joined #openstack-sdks12:41
*** gtema has quit IRC12:41
*** gtema has joined #openstack-sdks12:42
*** Luzi has quit IRC12:55
*** dave-mccowan has quit IRC13:13
*** dave-mccowan has joined #openstack-sdks13:14
*** mriedem has joined #openstack-sdks13:15
*** dave-mccowan has quit IRC13:19
*** ttsiouts has quit IRC14:25
*** ttsiouts has joined #openstack-sdks14:25
*** ttsiouts has quit IRC14:30
*** gtema_ has joined #openstack-sdks14:45
*** ttsiouts has joined #openstack-sdks14:49
mordredefried: you'll be happy to know that the 401 fix for keystoneauth, in additional to uncovering a different issue with swift (easy to fix) also uncovered the fact that, since we weren't ACTUALLY doing version discovery for nova before, we were falling back to 2.1 in sdk in all cases (which was actually our intention - we wanted to opt-in to newer behavior as we supported it) - but we were falling back15:15
mordredto it accidentally- so with discovery fixed, sdk is getting newest rather than oldest microversion by default - and the change to flavor broke sdk15:15
mordredefried: yay for testing15:15
mordredgtema: didn't we have code in sdk that was supposed to cause sdk to skip trying to do version discovery for swift?15:19
gtemayes - skip_discovery in proxy15:19
mordredlike - I feel like we had it - then I forgot about it and wrote it again15:20
mordredgtema: then why is https://07f03003afd3ca1f2387-1738d3a5d5585f54500f5c8600ed53a6.ssl.cf1.rackcdn.com/685042/4/check/openstacksdk-functional-devstack-tips/64dae12/testr_results.html.gz15:20
mordreddoing version discovery gets?15:20
gtemathat's a different story :D15:21
mordred(I know why they're broken - that's the ksa patch - but upon reflection, why are we doing them in the first place?)15:21
gtemaperhaps something else in your patches? lemme try in my cloud fast15:21
mordredgtema: oh!15:22
mordredgtema: it's the test itself15:22
*** ttsiouts has quit IRC15:22
mordredgtema: the functional base test has a "require_service" method which we have a todo to replace15:22
mordredwhich is doing an explicit get_endpoint_data15:22
gtemaaah, ok. In real life discovery is not done for swift15:22
mordredso - I think ksa should throw a better error in the case that someone tries this - but I also think we should fix these tests15:23
mordredthis is a fun rabbit hole15:23
* mordred hands efried the prize for finding the funnest bug15:23
gtemaagree. But I like the approach of passing auth to endpoint if we got exception15:25
gtemaI have it in my cloud everywhere15:25
gtemaand this is exactly what I was pointing in the API-SIG rework about version discovery15:26
gtemathose clear requirements were lost15:26
efriedmordred: Is sdk *supposed* to be getting the newest rather than the oldest microversion by default? (by design)15:32
efriedokay, reread your first message, got it.15:32
mordredso - you know - bugs all around :)15:32
efriedThough I would imagine specific operations would explicitly use newer microversions under the covers.15:32
mordredyes - absolutely!15:32
mordredand for those it should absolutely use them15:33
efriedbut, like, something where /foo changes payload formats with different microversions, and we have a foos() method, I guess we would need you to tell us which shape you wanted the response to take.15:33
mordredwe just might need to take additional action sdk-side to keep the sdk api consistent, so we want to opt-in to microversion changes (which is one of the reasons I've become an ardent supporter of microversions)15:33
efriedunless we obscure all of that by objectifying the response15:33
efriedbut there again, some fields might be present/absent depending15:34
mordredyah- in the normal shade and sdk layer, we obscure by objectifying. if you want the REST layer and know what you're doing, requesting like 2.latest should be fine15:34
efriedpresumably the microversion kwarg passes through request() seamlessly still?15:35
efriedyeah, it must, because we're using it for placement from nova15:35
mordredfields being different, specifically flavor, is what caused the tests to catch that the ksa 401 fix had unbroken a discovery thing that was accidentally keeping sdk working properly :)15:35
efried...unless we're getting the latest, which would also be working without us knowing...15:35
efriedyeesh, so we have to fix ksa and then blacklist the fix from sdk until we fix sdk?15:36
mordredno - we can fix sdk before we fix ksa15:36
efriedI had a mentor early in my career who called this RBB15:36
efriedRevoked Bug Benefit15:36
mordredit's a good term15:36
efriedmordred: well, in other news, I think I got compute versioned discovery unauthed working: https://review.opendev.org/68518115:37
efriedquestion is whether the change in NoAuthMiddleware is going to be acceptable. I would hope that's not being used outside of our own testing, but it seems technically supported by our conf, so I'm not sure.15:37
mordredI'm glad it was broken though - look at all the other stuff we get to fix15:37
efriedUnfortunately, people who would know or care seem to be sparse.15:38
*** jpena is now known as jpena|brb15:42
openstackgerritMonty Taylor proposed openstack/keystoneauth master: Fetch discovery documents with auth when needed  https://review.opendev.org/68504215:52
openstackgerritMonty Taylor proposed openstack/openstacksdk master: Use has_service in functional test's require_service  https://review.opendev.org/68599515:54
mordredefried, gtema: I *think* that should handle all the things except for the nova microversion issue15:59
openstackgerritMonty Taylor proposed openstack/keystoneauth master: Fetch discovery documents with auth when needed  https://review.opendev.org/68504216:04
*** jpich has quit IRC16:10
mordredgtema, efried: BWAHAHAHAHA. it's "not a bug" that we're requesting a later microversion from nova - we actually bumped the sdk layer microversion for the server resource16:11
*** gtema_ has quit IRC16:12
mordredbut - because of the other bug, we never actually were getting 2.72 payloads in devstack and didn't notice we needed to update our flavor support16:12
gtemalife is perfect16:12
gtema(unless one touches anything)16:12
gtemadiscovery and microversions are voodoo things in OpenStack16:14
*** e0ne has quit IRC16:18
*** jpena|brb is now known as jpena16:19
openstackgerritMonty Taylor proposed openstack/openstacksdk master: Fix server for later microversion  https://review.opendev.org/68599916:23
mordredgtema: yah. although they wouldn't have been if the client libraries hadn't hardcoded around them and if people hadn't put versioned endpoints in the catalog for the last 9 years. like - there was this whole system that literally nobody actually used :(16:24
gtemahope we will fix it now with Nova switched to SDK and others using it deeper16:25
mordredgtema: or we've just opened ourselves to a new kind of pain16:27
*** ricolin has quit IRC16:33
gtemathat's of course as well16:35
*** dtantsur is now known as dtantsur|afk16:40
*** jpena is now known as jpena|off17:16
*** enriquetaso has quit IRC17:21
*** ralonsoh has quit IRC17:27
*** gtema_ has joined #openstack-sdks17:34
*** e0ne has joined #openstack-sdks17:35
*** gtema has quit IRC17:37
*** e0ne has quit IRC18:17
*** gtema_ has quit IRC19:06
*** e0ne has joined #openstack-sdks19:08
*** slaweq has quit IRC19:59
*** enriquetaso has joined #openstack-sdks20:09
*** enriquetaso has quit IRC20:17
*** e0ne has quit IRC20:27
*** enriquetaso has joined #openstack-sdks20:41
*** cdent has quit IRC20:59
*** openstackstatus has quit IRC21:32
*** openstackstatus has joined #openstack-sdks21:35
*** ChanServ sets mode: +v openstackstatus21:35
*** mriedem has quit IRC21:56
*** enriquetaso has quit IRC23:01
*** tosky has quit IRC23:02

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!