| *** hieulq has joined #openstack-sdks | 00:14 | |
| *** thrash is now known as thrash|g0ne | 00:26 | |
| *** zhurong has joined #openstack-sdks | 00:36 | |
| *** kevo has quit IRC | 01:06 | |
| *** hoangcx has joined #openstack-sdks | 01:18 | |
| *** chlong has joined #openstack-sdks | 01:18 | |
| *** yanyanhu has joined #openstack-sdks | 01:27 | |
| *** hoangcx has quit IRC | 01:31 | |
| *** salv-orlando has joined #openstack-sdks | 01:34 | |
| *** hockeynut has quit IRC | 01:43 | |
| *** salv-orlando has quit IRC | 01:44 | |
| *** aohuanxuan has joined #openstack-sdks | 01:58 | |
| *** hoangcx has joined #openstack-sdks | 02:09 | |
| *** gildub has quit IRC | 02:10 | |
| *** jamielennox|away is now known as jamielennox | 02:41 | |
| *** salv-orlando has joined #openstack-sdks | 02:43 | |
| *** gouthamr has quit IRC | 02:49 | |
| *** salv-orlando has quit IRC | 02:50 | |
| *** amotoki has joined #openstack-sdks | 02:51 | |
| *** salv-orlando has joined #openstack-sdks | 02:52 | |
| *** salv-orlando has quit IRC | 02:57 | |
| *** hoangcx has quit IRC | 02:58 | |
| *** aohuanxuan has quit IRC | 03:06 | |
| *** amotoki has quit IRC | 03:08 | |
| *** kevo has joined #openstack-sdks | 03:12 | |
| *** amotoki has joined #openstack-sdks | 03:21 | |
| *** amotoki has quit IRC | 03:30 | |
| openstackgerrit | Qiming Teng proposed openstack/python-openstacksdk: Rebase network resources onto resource2 (1) https://review.openstack.org/364043 | 03:31 |
|---|---|---|
| *** amotoki has joined #openstack-sdks | 03:35 | |
| openstackgerrit | Qiming Teng proposed openstack/python-openstacksdk: Rebase network resources onto resource2 (2) https://review.openstack.org/364052 | 03:53 |
| *** salv-orlando has joined #openstack-sdks | 03:53 | |
| *** salv-orlando has quit IRC | 03:56 | |
| *** amotoki has quit IRC | 04:01 | |
| *** jamielennox is now known as jamielennox|away | 04:05 | |
| *** amotoki has joined #openstack-sdks | 04:16 | |
| *** kevo has quit IRC | 04:19 | |
| *** amit213 has quit IRC | 04:37 | |
| *** NobodyCam has quit IRC | 04:38 | |
| *** scottda has quit IRC | 04:38 | |
| *** etoews has quit IRC | 04:38 | |
| *** nikhil has quit IRC | 04:38 | |
| *** GheRivero has quit IRC | 04:38 | |
| *** scottda has joined #openstack-sdks | 04:40 | |
| *** amit213 has joined #openstack-sdks | 04:40 | |
| *** GheRivero has joined #openstack-sdks | 04:40 | |
| *** nikhil has joined #openstack-sdks | 04:40 | |
| *** etoews has joined #openstack-sdks | 04:40 | |
| *** NobodyCam has joined #openstack-sdks | 04:41 | |
| *** dtroyer has quit IRC | 04:44 | |
| *** dtroyer has joined #openstack-sdks | 04:44 | |
| *** zacksh has quit IRC | 04:45 | |
| *** zacksh has joined #openstack-sdks | 04:47 | |
| *** salv-orlando has joined #openstack-sdks | 05:03 | |
| *** meteorfox has quit IRC | 05:03 | |
| *** jamielennox|away is now known as jamielennox | 05:04 | |
| *** meteorfox has joined #openstack-sdks | 05:04 | |
| *** salv-orlando has quit IRC | 05:10 | |
| *** salv-orlando has joined #openstack-sdks | 05:14 | |
| *** kevo has joined #openstack-sdks | 05:33 | |
| *** adriant has quit IRC | 05:49 | |
| *** hoangcx has joined #openstack-sdks | 05:57 | |
| *** edmondsw has joined #openstack-sdks | 05:59 | |
| *** edmondsw has quit IRC | 06:00 | |
| *** reedip has quit IRC | 06:03 | |
| *** salv-orl_ has joined #openstack-sdks | 06:08 | |
| *** kzaitsev_ws has quit IRC | 06:08 | |
| *** salv-orlando has quit IRC | 06:11 | |
| openstackgerrit | Qiming Teng proposed openstack/python-openstacksdk: Rebase network resources onto resource2 (2) https://review.openstack.org/364052 | 06:15 |
| *** kzaitsev_ws has joined #openstack-sdks | 06:15 | |
| openstackgerrit | Qiming Teng proposed openstack/python-openstacksdk: Rebase network resources to resource2 (3) https://review.openstack.org/364084 | 06:16 |
| *** kzaitsev_ws has quit IRC | 06:41 | |
| *** salv-orl_ has quit IRC | 06:50 | |
| *** hoangcx has quit IRC | 06:52 | |
| *** hoangcx_ has joined #openstack-sdks | 06:54 | |
| *** kzaitsev_ws has joined #openstack-sdks | 06:58 | |
| *** jamielennox is now known as jamielennox|away | 07:17 | |
| *** jpena|off is now known as jpena | 07:18 | |
| *** kevo has quit IRC | 07:29 | |
| *** hoangcx_ has quit IRC | 07:35 | |
| *** salv-orlando has joined #openstack-sdks | 07:51 | |
| *** hoangcx has joined #openstack-sdks | 07:53 | |
| *** salv-orlando has quit IRC | 08:00 | |
| *** hoangcx has quit IRC | 08:05 | |
| openstackgerrit | Qiming Teng proposed openstack/python-openstacksdk: Rebase network resources to resource2 (3) https://review.openstack.org/364084 | 08:10 |
| *** rvasilets__ has joined #openstack-sdks | 08:36 | |
| *** lucas-dinner is now known as lucasagomes | 08:38 | |
| *** e0ne has joined #openstack-sdks | 08:44 | |
| *** HenryG_ has joined #openstack-sdks | 08:54 | |
| *** amotoki_ has joined #openstack-sdks | 08:55 | |
| openstackgerrit | Qiming Teng proposed openstack/python-openstacksdk: Rebase network resources to resource2 (3) https://review.openstack.org/364084 | 08:56 |
| openstackgerrit | Qiming Teng proposed openstack/python-openstacksdk: Rebase network proxy to proxy2 https://review.openstack.org/364164 | 08:58 |
| *** amotoki has quit IRC | 08:58 | |
| *** tangchen has quit IRC | 08:58 | |
| *** bapalm has quit IRC | 08:58 | |
| *** HenryG has quit IRC | 08:58 | |
| *** malloc83 has quit IRC | 08:58 | |
| *** eliqiao has quit IRC | 08:58 | |
| *** salv-orlando has joined #openstack-sdks | 08:58 | |
| *** bapalm has joined #openstack-sdks | 08:59 | |
| *** tangchen has joined #openstack-sdks | 09:00 | |
| *** salv-orlando has quit IRC | 09:01 | |
| *** malloc83 has joined #openstack-sdks | 09:03 | |
| *** eliqiao has joined #openstack-sdks | 09:03 | |
| *** cdent has joined #openstack-sdks | 09:05 | |
| *** HenryG_ is now known as HenryG | 09:17 | |
| *** namnh has joined #openstack-sdks | 09:21 | |
| *** HenryG has quit IRC | 09:29 | |
| *** HenryG has joined #openstack-sdks | 09:29 | |
| *** salv-orlando has joined #openstack-sdks | 09:45 | |
| *** sdague has joined #openstack-sdks | 09:46 | |
| *** yanyanhu has quit IRC | 09:52 | |
| *** zhurong has quit IRC | 10:03 | |
| *** namnh has quit IRC | 10:15 | |
| *** thrash|g0ne is now known as thrash | 10:19 | |
| *** sigmavirus|awa is now known as sigmavirus | 10:37 | |
| *** Qiming has quit IRC | 10:40 | |
| *** Qiming has joined #openstack-sdks | 10:45 | |
| *** salv-orlando has quit IRC | 10:58 | |
| *** reedip has joined #openstack-sdks | 11:15 | |
| *** rtheis has joined #openstack-sdks | 11:18 | |
| *** lucasagomes is now known as lucas-hungry | 11:33 | |
| *** Kevin_Zheng has quit IRC | 11:42 | |
| *** jpena is now known as jpena|lunch | 11:56 | |
| *** Kevin_Zheng has joined #openstack-sdks | 11:56 | |
| *** fzdarsky has joined #openstack-sdks | 12:01 | |
| *** lucas-hungry is now known as lucasagomes | 12:21 | |
| *** dtantsur has joined #openstack-sdks | 12:31 | |
| mordred | dtroyer, stevemar: just had a gate-breaking issue with the latest osc - filed a bug: https://bugs.launchpad.net/python-openstackclient/+bug/1619274 | 12:32 |
| openstack | Launchpad bug 1619274 in python-openstackclient "openstack complete triggers password prompt" [Undecided,New] | 12:32 |
| mordred | we also pushed up a devstack level workaround ... | 12:32 |
| *** zhurong has joined #openstack-sdks | 12:34 | |
| *** _elmiko is now known as elmiko | 12:40 | |
| dtantsur | hi everyone! what's the OSC analog of "neutron router-gateway-set"? | 12:41 |
| cdent | etoews, elmiko: I would be best served by skipping api-wg meeting today, cuz feature freeze and such | 12:46 |
| elmiko | cdent: is it that time of season again? | 12:46 |
| cdent | rather | 12:46 |
| elmiko | i don't have any objections to that | 12:46 |
| *** markvoelker has joined #openstack-sdks | 12:50 | |
| *** gouthamr has joined #openstack-sdks | 12:51 | |
| *** jpena|lunch is now known as jpena | 12:55 | |
| rtheis | dtantsur: there isn't an OSC equivalent yet ... it is in progress here: https://review.openstack.org/#/c/357973/ | 13:02 |
| dtantsur | got it, thanks. I guess, not in Newton? | 13:03 |
| dtantsur | I wonder why these "commands" are called "advances".. floating IPs don't seem to work without a gateway. maybe it's only for me though :) | 13:07 |
| *** salv-orlando has joined #openstack-sdks | 13:07 | |
| *** markvoelker has quit IRC | 13:07 | |
| *** salv-orlando has quit IRC | 13:13 | |
| *** amotoki_ has quit IRC | 13:14 | |
| openstackgerrit | Merged openstack/python-openstackclient: Rearrange integration tests https://review.openstack.org/363920 | 13:17 |
| *** markvoelker has joined #openstack-sdks | 13:23 | |
| *** markvoelker has quit IRC | 13:28 | |
| *** amotoki has joined #openstack-sdks | 13:31 | |
| rtheis | They aren't really advanced commands. We should broke them down into different categories for development | 13:37 |
| dtantsur | +1 | 13:42 |
| *** amotoki has quit IRC | 13:43 | |
| *** clouddig has joined #openstack-sdks | 13:45 | |
| *** cleong has joined #openstack-sdks | 13:48 | |
| *** openstackgerrit has quit IRC | 13:49 | |
| stevemar | mordred: wheres the devstack workaround? | 13:49 |
| *** openstackgerrit has joined #openstack-sdks | 13:50 | |
| mordred | stevemar: https://review.openstack.org/#/c/364266/ | 13:50 |
| *** amotoki has joined #openstack-sdks | 13:50 | |
| stevemar | mordred: cool | 13:50 |
| mordred | stevemar: giving an --os-cloud will get it find a password, so it won't trigger the prompt | 13:50 |
| stevemar | mordred: cool, let me see if i can dig into this one and give dtroyer a break | 13:51 |
| mordred | coolio | 13:51 |
| mordred | I started poking a little earlier but then something else caught fire so I haven't gotten all the way there | 13:51 |
| stevemar | mordred: isnt that always the case? :) | 13:52 |
| mordred | stevemar: zomg | 13:52 |
| mordred | dtantsur: simple solution - stop using floating ips. problem solved! | 13:52 |
| dtantsur | mordred, well... maybe I don't know enough about openstack network, but will I be able to access the VMs from outside then? | 13:53 |
| mordred | dtantsur: if the cloud is configured with a shared public network, yes. I'm mostly trolling, as a user you have no control over whether your cloud has configured itself to require that you put your servers behind NAT or not | 13:54 |
| mordred | which is annoying | 13:54 |
| dtantsur | true | 13:55 |
| mordred | but as a deployer, you can absolutely configure neutron to actually put IPs onto vms directly ... just in case you're a deployer I figured I'd be my normal annoying anti-FIP voice | 13:55 |
| dtantsur | right now I'm acting as a deployer, but my neutron knowledge is pretty shallow :) | 13:56 |
| mordred | dtantsur: this is true for most of us I think ;) | 13:58 |
| dtantsur | fair :D | 13:58 |
| *** salv-orlando has joined #openstack-sdks | 13:59 | |
| *** amotoki has quit IRC | 13:59 | |
| dtroyer | ugh… mordred this break is partly due to moving the auth waaaay up in the sequence, we used to defer it until we were actually processing a command and had things like if the command required auth or not. | 14:03 |
| dtroyer | stevemar: the fix is going to be to defer the prompting until when we used to do it in ClientManager.setup_auth() | 14:04 |
| mordred | dtroyer: I'm not sure if I've ever asked this or not - but is there any reason why passing validate=False fo get_one_cloud doesn't work for you? | 14:04 |
| mordred | dtroyer: that completely skips all auth plugin interactions | 14:05 |
| dtroyer | I did that at first, and then had to duplicate the rest of the plugin arg handling anyway | 14:05 |
| *** fguillot has joined #openstack-sdks | 14:06 | |
| dtroyer | that is where I am headed though, to break down get_one_cloud() into enough pieces that OSC doesn't need to use it | 14:06 |
| dtroyer | directly, that is, call the pieces | 14:06 |
| mordred | yah | 14:06 |
| *** dtantsur has left #openstack-sdks | 14:06 | |
| mordred | I hear you - and agree. I think having the thing that's protected by validate==True currently would make a very clear and obvious function that returns an auth plugin, for instance | 14:06 |
| dtroyer | I did do that a couple of weeks ago, and soething in the exception handling screwed me up so I aborted that one | 14:07 |
| mordred | nod | 14:07 |
| dtroyer | due to timing and needing to fix stuff first | 14:07 |
| *** fguillot has quit IRC | 14:10 | |
| *** amotoki has joined #openstack-sdks | 14:14 | |
| stevemar | dtroyer: you can take this one again, i'm working on the keystone release. i thought i could sneak in something in the morning but no such luck | 14:16 |
| stevemar | i don't want to hold up the train with my slowness | 14:17 |
| dtroyer | will do. fortunately it's all in OSC still so just one thing | 14:18 |
| *** fguillot has joined #openstack-sdks | 14:27 | |
| mordred | dtroyer: also - any idea what's causing this to get triggered: http://logs.openstack.org/46/363846/2/check/gate-shade-dsvm-functional-libs-nv/623c5f3/logs/devstacklog.txt.gz#_2016-09-01_12_39_33_225 - is that an osc plugin perhaps? | 14:46 |
| dtroyer | it is, IIRC ironicclient needs a release to clear it, or if tripleoclient is loaded, it too | 14:46 |
| mordred | likely ironicclient - we load that in that test | 14:47 |
| dtroyer | that's what it was when I last looked on Tuesday | 14:47 |
| mordred | awesome. I imagine that should get taken care of today /me crosses fingers | 14:47 |
| *** singhj has joined #openstack-sdks | 14:57 | |
| *** zhurong has quit IRC | 14:58 | |
| sdague | dtroyer: the other option, can you make the pass prompt timeout after 60s or something? | 15:00 |
| sdague | because I'm a little concern that it waits forever | 15:00 |
| sdague | which makes someone discovering that *it* is the issue, kind of hard | 15:00 |
| dtroyer | I can look in to that, also though it is supposed to skip the whole thing if stdin is not a tty | 15:00 |
| sdague | especially given that the prompt goes to stdout, not stderr | 15:01 |
| dtroyer | that was how we attempted to not block like this | 15:01 |
| sdague | so if you are redirecting the output to process it (which we were), you never see there is a prompt | 15:01 |
| dtroyer | why though does this evaluate True? hasattr(sys.stdin, 'isatty') and sys.stdin.isatty() | 15:02 |
| sdague | dtroyer: well, in a devstack run with ansible ssh, who knows what the tty situation is | 15:02 |
| sdague | mordred: you know? | 15:03 |
| dtroyer | right… it used to not be a tty (most of the time, and bad assumption to make, I know, but that was once the case) | 15:03 |
| *** annegentle has joined #openstack-sdks | 15:04 | |
| *** fzdarsky has quit IRC | 15:04 | |
| dtroyer | also, not sure how to timeout getpass.getpass() in py2.7 | 15:04 |
| sdague | yeh, no idea | 15:05 |
| sdague | you'd probably need to spawn a timer and kill it? | 15:05 |
| dtroyer | I would be happy to remove prompting altogether, but that doesn't go over well with some ;) | 15:06 |
| sdague | dtroyer: ++ on removing prompting | 15:06 |
| sdague | dtroyer: who does that not go over well with? | 15:06 |
| dtroyer | heh, getpass() is supposed to write to stderr, and if you don't see the prompt there, then thet confirms the process sees a tty on stdin | 15:06 |
| dtroyer | *stderr if no tty | 15:07 |
| sdague | right, which means there definitely is a tty | 15:07 |
| sdague | otherwise we'd have the output | 15:07 |
| stevemar | sdague: eh, people like the password prompt | 15:08 |
| sdague | stevemar: what people? | 15:08 |
| stevemar | sdague: just speaking from past experiences when i've shown folks the cli | 15:08 |
| dtroyer | I can't quantify them, but it was long before clouds.yaml was a thing… | 15:08 |
| stevemar | i realize you can pass it in as --os-password, and in clouds.yaml | 15:09 |
| sdague | I actually can't imagine anyone liking that. And if so, make another front end that does that part | 15:09 |
| stevemar | eh, this is the second bug i've seen related to it in many years | 15:09 |
| sdague | a normal interaction session is going to be like at least a dozen osc commands | 15:09 |
| dtroyer | so how about this? we make -q also skip prompting (now it just sets log levels to error or debug (can't recall)) | 15:10 |
| sdague | and typing password every time is just crazy | 15:10 |
| sdague | dtroyer: or, make there be a 'prompt me' flag | 15:10 |
| *** fzdarsky has joined #openstack-sdks | 15:12 | |
| sdague | because being hung, for an infinite amount of time, is a pretty critical bug | 15:12 |
| dtroyer | ok, so I understand, in the failing case stack.sh is spawned (indirectly?) from ansible and not another shell or jenkins? | 15:15 |
| sdague | yes | 15:15 |
| sdague | stack.sh is always spawned by ansible now | 15:16 |
| sdague | there is no longer jenkins | 15:16 |
| sdague | zuul doing ansible is the replacement | 15:16 |
| dtroyer | gotcha, so this is going to be everywhere in the gate | 15:16 |
| sdague | yes | 15:16 |
| sdague | but... | 15:16 |
| sdague | the issue only shows up on the subnode in multinode | 15:16 |
| sdague | because apparently it was never really configured for osc | 15:17 |
| sdague | which is fine, the only osc command we ran is the generate bash completion | 15:17 |
| sdague | which used to work | 15:17 |
| dtroyer | because that's the only place a no-auth command is run without something like env vars or OS_CLOUD set I'm guessing | 15:17 |
| sdague | now prompts for a password | 15:17 |
| sdague | so, we worked around that thing with a devstack patch | 15:17 |
| dtroyer | setting auth-type to token should bypass it. auth-type == password needs the password | 15:17 |
| dtroyer | but we really can't change the default, as that is also horrible ux | 15:18 |
| sdague | sure, but the point is previously if things were misconfigured you'd get a fail | 15:18 |
| sdague | or maybe not | 15:18 |
| sdague | anyway, the point is if you are misconfigured, you can hang the shell forever | 15:18 |
| dtroyer | no, we never hit the auth code previously because it was deferred until we were processing the command | 15:18 |
| sdague | ok | 15:19 |
| dtroyer | o-c-c moved all of that waaaaay up in the process, well before we have even parsed the comamnd | 15:19 |
| sdague | well, then we exposed this issue that we hang the prompt forever | 15:19 |
| sdague | there is no other unix command that does that | 15:19 |
| sdague | sudo times out | 15:19 |
| dtroyer | I'm not arguing with you, this shouldn't be happening | 15:19 |
| sdague | ok | 15:19 |
| dtroyer | it's the damn o-c-c rewrite that as f*cked everything up | 15:19 |
| dtroyer | ruined my summer at least | 15:20 |
| *** annegentle has quit IRC | 15:32 | |
| *** fzdarsky has quit IRC | 15:35 | |
| *** salv-orlando has quit IRC | 15:43 | |
| *** markvoelker has joined #openstack-sdks | 15:51 | |
| *** fzdarsky has joined #openstack-sdks | 16:00 | |
| etoews | cdent: elmiko: you available for the api wg meeting? | 16:01 |
| cdent | etoews: sorry, said above, feature freezing :( | 16:02 |
| etoews | sorry i missed it. | 16:02 |
| rtheis | briancurtin: how would OSC pin its SDK requirement? Doesn't that have to be done globally in openstack/requirements? | 16:07 |
| etoews | cdent: sorry to be a bother but can you please workflow+1 https://review.openstack.org/#/c/354266/ since you were the one who froze it? | 16:08 |
| cdent | etoews: done | 16:09 |
| etoews | thx! | 16:09 |
| openstackgerrit | Merged openstack/api-wg: Change Cinder api-wg liason https://review.openstack.org/363163 | 16:09 |
| openstackgerrit | Merged openstack/api-wg: Clear the case if the version string isn't parsable https://review.openstack.org/346846 | 16:11 |
| *** salv-orlando has joined #openstack-sdks | 16:12 | |
| openstackgerrit | Corentin Ardeois proposed openstack/js-openstack-lib: Create common service version abstraction https://review.openstack.org/363304 | 16:13 |
| openstackgerrit | Merged openstack/api-wg: A guideline for links https://review.openstack.org/354266 | 16:15 |
| openstackgerrit | Merged openstack/api-wg: Add the beginning of a set of guidlines for URIs https://review.openstack.org/322194 | 16:15 |
| *** annegentle has joined #openstack-sdks | 16:17 | |
| *** e0ne has quit IRC | 16:21 | |
| *** lucasagomes is now known as lucas-dinner | 16:21 | |
| briancurtin | rtheis: im not sure, i don’t really pay attention to global anything. i would hope there’s a way for a project to just not take certain versions of something | 16:31 |
| briancurtin | which is how python works | 16:31 |
| rtheis | briancurtin: osc is pinned to global requirements so cannot deviate | 16:32 |
| briancurtin | if it has to wait because that’s the process, then i guess it has to wait. it will not wait very long, though, breakage or not | 16:32 |
| rtheis | briancurtin: I understand. Will try to work on this as soon as I can | 16:38 |
| briancurtin | rtheis: what amount of work will it take on the OSC side of things? is it one patch or a major thing? so far nothing in the other refactorings have changed usage | 16:39 |
| rtheis | briancurtin: I'm not sure the extent yet. I need to revisit my initial analysis and run some tests against qiming's patches | 16:40 |
| *** amotoki has quit IRC | 16:51 | |
| openstackgerrit | Michael McCune proposed openstack/api-wg: add a warning about json expectations https://review.openstack.org/364460 | 16:52 |
| elmiko | etoews, cdent ^^ | 16:52 |
| openstackgerrit | Michael McCune proposed openstack/api-wg: add a warning about json expectations https://review.openstack.org/364460 | 16:53 |
| *** markvoelker has quit IRC | 16:58 | |
| cdent | elmiko: | 17:04 |
| cdent | and etoews the gerrit+launchpad connection seems to be working | 17:04 |
| elmiko | was it out for awhile? | 17:04 |
| *** kevo has joined #openstack-sdks | 17:08 | |
| openstackgerrit | Michael Krotscheck proposed openstack/js-openstack-lib: Added more verbose logging to keystone functional tests. https://review.openstack.org/363301 | 17:17 |
| *** kzaitsev_ws has quit IRC | 17:31 | |
| *** kzaitsev_ws has joined #openstack-sdks | 17:31 | |
| *** e0ne has joined #openstack-sdks | 17:32 | |
| dtroyer | sdague: FWIW, I don't know why that osc complete command ever worked previously. The complete command wants to know if Neutron is available, which we get from the Service catalog, which, you know, requires auth. | 17:42 |
| *** jpena is now known as jpena|away | 17:46 | |
| *** e0ne has quit IRC | 17:53 | |
| dtroyer | ah, it didn't work and hasn't for a loooong time: http://logs.openstack.org/46/357446/1/check/gate-tempest-dsvm-neutron-multinode-full/39bd9ea/logs/subnode-2/devstacklog.txt.gz#_2016-08-18_21_24_59_718 | 18:00 |
| openstackgerrit | Michael Krotscheck proposed openstack/js-openstack-lib: Added loglevel https://review.openstack.org/364456 | 18:00 |
| openstackgerrit | Michael Krotscheck proposed openstack/js-openstack-lib: Added more verbose logging to keystone functional tests. https://review.openstack.org/363301 | 18:00 |
| openstackgerrit | Michael Krotscheck proposed openstack/js-openstack-lib: Use clouds.yaml credentials for admin functional tests. https://review.openstack.org/364484 | 18:00 |
| *** e0ne has joined #openstack-sdks | 18:01 | |
| sdague | dtroyer: ok, but it never hung before | 18:02 |
| sdague | and pipe hid the error | 18:02 |
| dtroyer | understood, I was off looking in a place I didn't seed to be | 18:02 |
| dtroyer | *need | 18:03 |
| dtroyer | trying to fix the why-is-complete-wanting-auth question | 18:03 |
| dtroyer | so the devstack workaround is actually the correct fix, even if osc didn't hang, because the complete output cannot be correct without auth | 18:07 |
| *** kzaitsev_ws has quit IRC | 18:07 | |
| *** salv-orl_ has joined #openstack-sdks | 18:08 | |
| *** singhj has quit IRC | 18:09 | |
| *** salv-orlando has quit IRC | 18:11 | |
| *** openstackgerrit has quit IRC | 18:18 | |
| *** cdent has quit IRC | 18:18 | |
| *** openstackgerrit has joined #openstack-sdks | 18:19 | |
| dtroyer | mordred: do you remember why o-c-c's _validate_auth_ksc() checks if the options are required but _validate_auth() does not? | 18:20 |
| mordred | dtroyer: gimme a sec to stare | 18:21 |
| mordred | dtroyer: also - I was this morning looking to see if we could delete _validate_auth_ksc .. it was only there because osc wasn't on ksa yet originally | 18:21 |
| dtroyer | yes, I've done that in osc's workaournds | 18:21 |
| dtroyer | which is why I now lost the required opt checking ;) | 18:22 |
| mordred | :) | 18:22 |
| mordred | well - I found another thing you'll love | 18:22 |
| mordred | when I delete that chunk of code ... a set of tests start failing | 18:22 |
| mordred | but they start failing not because they depended on ksc | 18:22 |
| mordred | but because they are somehow executing get_one_cloud a _second_time_ with validate=True | 18:22 |
| mordred | I have never been as baffled by python code before | 18:22 |
| dtroyer | heh, I'm doing it 3 times now | 18:23 |
| mordred | like, I've got prints everywhere and I cannot see where the second time is happening | 18:23 |
| mordred | it's happening for sure | 18:23 |
| mordred | but I have _no_clue_ where | 18:23 |
| *** singhj has joined #openstack-sdks | 18:24 | |
| mordred | I also got this: | 18:24 |
| mordred | File "os_client_config/tests/test_config.py", line 736, in test_register_argparse_service_type | 18:24 |
| mordred | self.assertRaises(AttributeError, opts.os_network_service_type) | 18:24 |
| mordred | AttributeError: 'Namespace' object has no attribute 'os_network_service_type' | 18:24 |
| mordred | wait - that's not the funy one | 18:24 |
| mordred | oh - yes it is. ... assertRaises AtributeError ... and then it raises AttributeError and the unittest dies | 18:25 |
| mordred | wth? | 18:25 |
| dtroyer | the attributeerror is being raised by the opts.os_network_service_type arg itself, likely before the function is even called? | 18:26 |
| mordred | aha! look at you with the smart | 18:26 |
| openstackgerrit | Monty Taylor proposed openstack/os-client-config: Split auth plugin loading into its own method https://review.openstack.org/364493 | 18:27 |
| openstackgerrit | Monty Taylor proposed openstack/os-client-config: Remove validate_auth_ksc https://review.openstack.org/364494 | 18:27 |
| dtroyer | no, just tired of looking at get_one_cloud() ;) | 18:27 |
| mordred | anywho - 364494 fails spectacularly in strange ways | 18:27 |
| mordred | I pushed it up just to share :) | 18:27 |
| dtroyer | k | 18:27 |
| * mordred now looks about why the reuqired options thing ... | 18:27 | |
| *** singhj has quit IRC | 18:28 | |
| dtroyer | what you wind up with in load_auth_plugin() after both of those is exactly what I have, except called get_auth_plugin() | 18:29 |
| dtroyer | maybe go straight to that in 364493? | 18:29 |
| dtroyer | well, ok I see why not | 18:29 |
| dtroyer | it'll work fine once the tests are sorted | 18:30 |
| mordred | yah. also, I can rename ot get_auth_plugin if you'd like | 18:30 |
| dtroyer | doesn't matter, I just like we came up with the same solution…means I'm not totally off | 18:30 |
| mordred | woot | 18:31 |
| *** singhj has joined #openstack-sdks | 18:36 | |
| *** annegentle has quit IRC | 18:39 | |
| *** cdent has joined #openstack-sdks | 18:39 | |
| *** singhj has quit IRC | 18:41 | |
| *** salv-orl_ has quit IRC | 18:42 | |
| *** annegentle has joined #openstack-sdks | 18:43 | |
| *** kzaitsev_ws has joined #openstack-sdks | 18:45 | |
| *** annegent_ has joined #openstack-sdks | 18:49 | |
| *** david-lyle has quit IRC | 18:49 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-openstackclient: Updated from global requirements https://review.openstack.org/364505 | 18:50 |
| *** david-lyle has joined #openstack-sdks | 18:50 | |
| *** annegentle has quit IRC | 18:52 | |
| *** singhj has joined #openstack-sdks | 18:56 | |
| *** eggshell has joined #openstack-sdks | 19:18 | |
| openstackgerrit | Dean Troyer proposed openstack/python-openstackclient: Defer auth prompting until it is actually needed https://review.openstack.org/364518 | 19:19 |
| dtroyer | mordred: ^^^ is my working solution that includes the required arg handling…turns out it is a bit more complicated with prompting also in there | 19:20 |
| mordred | dtroyer: that looks nice though | 19:21 |
| dtroyer | stevemar, rtheis: looks like I missed meeting start, still want to talk? | 19:21 |
| rtheis | dtroyer: I had a couple items, but see you are busy working another bug | 19:22 |
| *** salv-orlando has joined #openstack-sdks | 19:23 | |
| dtroyer | I just put it up for check queue, have mintes now | 19:23 |
| dtroyer | I also have minutes | 19:23 |
| rtheis | are we still holding new function merges? | 19:24 |
| dtroyer | I would like to until we branch stable/newton | 19:24 |
| dtroyer | otherwise things like today's fix gets complicated | 19:24 |
| rtheis | okay, will wait for that | 19:25 |
| rtheis | sdk updated in global requirements for network command fix | 19:25 |
| dtroyer | I saw that, +A it a minute ago, thanks! | 19:25 |
| rtheis | do we need a new release ? | 19:25 |
| rtheis | cool | 19:25 |
| openstackgerrit | Michael Krotscheck proposed openstack/js-openstack-lib: Passed tokens may now be promises. https://review.openstack.org/361431 | 19:26 |
| openstackgerrit | Michael Krotscheck proposed openstack/js-openstack-lib: Added Glance Service https://review.openstack.org/361415 | 19:26 |
| dtroyer | we will, so the prompt fix v2 and the g-r update, will be 3.3.0 it looks like | 19:26 |
| rtheis | ok | 19:26 |
| dtroyer | I may try more integ testing too if I have time. | 19:26 |
| dtroyer | so I suppose any other test-only reviews could merge as they don't change user visible stuff | 19:26 |
| rtheis | do we need to be concerned about https://bugs.launchpad.net/bugs/1618840 ? | 19:26 |
| openstack | Launchpad bug 1618840 in python-openstackclient "openstack client does not accept domain name in project list --domain" [Undecided,New] - Assigned to tamil vanan (tamilhce) | 19:26 |
| dtroyer | I haven't seen that one yet | 19:27 |
| dtroyer | ugh, the relationship between the three types of domain-id still trip me up… | 19:28 |
| dtroyer | I'll defer to stevemar about the severity of that bug, I really don't know if it is critical | 19:29 |
| rtheis | sounds good | 19:29 |
| dtroyer | maybe if the underlying problem affects more than just a list command? | 19:29 |
| *** annegent_ has quit IRC | 19:32 | |
| rtheis | dtroyer: and one more question ... I was wondering if it would be good to have a gate job to test osc with osc-lib master ? | 19:33 |
| dtroyer | yes, in osc-lib | 19:33 |
| dtroyer | oh, wait, you mean _both_ on master | 19:34 |
| dtroyer | yes to that too | 19:34 |
| rtheis | I was thinking both on master | 19:34 |
| openstackgerrit | Michael Krotscheck proposed openstack/js-openstack-lib: Create common service version abstraction https://review.openstack.org/363304 | 19:35 |
| openstackgerrit | Michael Krotscheck proposed openstack/js-openstack-lib: Created common service abstraction https://review.openstack.org/363185 | 19:35 |
| openstackgerrit | Michael Krotscheck proposed openstack/js-openstack-lib: Added imageList() method to glance API https://review.openstack.org/361442 | 19:35 |
| rtheis | I'll put that on my todo list. | 19:36 |
| *** sdague has quit IRC | 19:37 | |
| dtroyer | ok, thanks | 19:37 |
| dtroyer | mordred: at least part of the test issue in https://review.openstack.org/#/c/364493/ is testing things that ksc had and ksa doesn't, like NOVA_USERNAME | 19:47 |
| dtroyer | hmmmm…….I think | 19:47 |
| *** elmiko is now known as _elmiko | 20:00 | |
| rtheis | dtroyer: do you know if https://review.openstack.org/364518 may resolve https://bugs.launchpad.net/bugs/1619421? | 20:01 |
| openstack | Launchpad bug 1619421 in python-openstackclient "Help info for floating IPs fails without credentials" [Undecided,New] | 20:01 |
| dtroyer | probably? [looking] | 20:02 |
| dtroyer | no, it isn't going to | 20:03 |
| dtroyer | we created a problem with the 'look for network service' in those split network commands | 20:03 |
| dtroyer | that requires a service catalog, which requires auth | 20:03 |
| rtheis | ok | 20:04 |
| dtroyer | for the case of help, we _could_ show them both, but there are other places this pops up, like the complete command | 20:04 |
| *** annegentle has joined #openstack-sdks | 20:05 | |
| *** cdent has quit IRC | 20:09 | |
| *** fguillot has quit IRC | 20:16 | |
| *** fzdarsky has quit IRC | 20:23 | |
| *** eggshell has quit IRC | 20:28 | |
| *** e0ne has quit IRC | 20:30 | |
| *** gouthamr has quit IRC | 20:34 | |
| *** jpena|away is now known as jpena|off | 20:35 | |
| *** annegentle has quit IRC | 20:39 | |
| *** annegentle has joined #openstack-sdks | 20:47 | |
| *** rvasilets___ has joined #openstack-sdks | 21:10 | |
| *** jidar has joined #openstack-sdks | 21:28 | |
| jidar | is there no option on my python-openstackclient to create a server+cinder volume? `--block-device-mapping dev-name=foo:cinder:20:delete_on_terminate` seems to expect there to already be a volume named foo | 21:29 |
| *** rtheis has quit IRC | 21:29 | |
| *** cleong has quit IRC | 21:30 | |
| dtroyer | jidar: correct, the volume must exist | 21:31 |
| *** gouthamr has joined #openstack-sdks | 21:32 | |
| *** adriant has joined #openstack-sdks | 21:42 | |
| *** gouthamr_ has joined #openstack-sdks | 21:44 | |
| *** adriant has quit IRC | 21:45 | |
| openstackgerrit | Michael Krotscheck proposed openstack/js-openstack-lib: Create common service version abstraction https://review.openstack.org/363304 | 21:45 |
| openstackgerrit | Michael Krotscheck proposed openstack/js-openstack-lib: Added Glance Service https://review.openstack.org/361415 | 21:45 |
| openstackgerrit | Michael Krotscheck proposed openstack/js-openstack-lib: Created common service abstraction https://review.openstack.org/363185 | 21:45 |
| openstackgerrit | Michael Krotscheck proposed openstack/js-openstack-lib: Added imageList() method to glance API https://review.openstack.org/361442 | 21:46 |
| *** gouthamr has quit IRC | 21:47 | |
| *** adriant has joined #openstack-sdks | 21:59 | |
| *** singhj has quit IRC | 22:13 | |
| openstackgerrit | Merged openstack/python-openstackclient: Updated from global requirements https://review.openstack.org/364505 | 22:17 |
| *** thrash is now known as thrash|g0ne | 22:19 | |
| *** adriant has quit IRC | 22:25 | |
| *** clouddig has quit IRC | 22:27 | |
| *** adriant has joined #openstack-sdks | 22:45 | |
| *** gouthamr_ has quit IRC | 22:53 | |
| *** annegentle has quit IRC | 23:01 | |
| *** gouthamr has joined #openstack-sdks | 23:10 | |
| *** salv-orlando has quit IRC | 23:10 | |
| openstackgerrit | Monty Taylor proposed openstack/os-client-config: Add ability to configure Session constructor https://review.openstack.org/364589 | 23:10 |
| *** markvoelker has joined #openstack-sdks | 23:11 | |
| mordred | dtroyer: this ^^ is inspired by your KeystoneSession object. I could use a similar wrapper in shade ... but since I get my Session's from occ, I don't have a way to create the wrapper sanely | 23:11 |
| *** gouthamr_ has joined #openstack-sdks | 23:14 | |
| *** gouthamr has quit IRC | 23:14 | |
| *** rvasilets___ has quit IRC | 23:17 | |
| *** annegentle has joined #openstack-sdks | 23:27 | |
| *** chlong has quit IRC | 23:33 | |
| *** zhurong has joined #openstack-sdks | 23:48 | |
| *** markvoelker has quit IRC | 23:51 | |
| dtroyer | mordred: /me shakes head… OSC will never use it but it doesn't get in our way | 23:51 |
| mordred | dtroyer: yah - I didn't think it would be useful to you at all | 23:53 |
| mordred | dtroyer: much more concerned that it doens't bother you | 23:54 |
| dtroyer | it doesn't. too bad osc-lib didn't happen 2 years ago, you really could have just used ClientManager for most of this kind of stuff | 23:54 |
| dtroyer | dhellmann did a nice job structuring that for us | 23:54 |
| mordred | yah | 23:56 |
| *** markvoelker has joined #openstack-sdks | 23:56 | |
| openstackgerrit | Eric Brown proposed openstack/keystoneauth: Update reno for stable/newton https://review.openstack.org/362412 | 23:57 |
| openstackgerrit | Eric Brown proposed openstack/osc-lib: Update reno for stable/newton https://review.openstack.org/362399 | 23:58 |
| *** zhurong has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!