| mnewby | sdague: Tempest exposes the dark underbelly of openstack rest api's and python client api's. Some projects are good, others are terrible, consistency is shockingly absent. | 00:03 |
|---|---|---|
| mnewby | (or perhaps not-so-shockingly) | 00:04 |
| *** anniec has quit IRC | 00:05 | |
| jaypipes | mordred: I've checked out both testr and testresources and I haven't been able to figure out how to apply fixtures in a functional integration test way... it works great for unit tests, not so much for this. | 00:06 |
| *** esp has quit IRC | 00:06 | |
| *** sacharya has quit IRC | 00:07 | |
| *** mtreinish has quit IRC | 00:10 | |
| *** jaypipes has quit IRC | 00:13 | |
| *** yidclare has joined #openstack-dev | 00:14 | |
| sdague | mnewby: honestly, tempest has it's own dark underbelly | 00:19 |
| mnewby | sdague: I agree, but trying to work with openstack client python api's has been it's own brand of terrible. | 00:19 |
| mnewby | sdague: I can't believe these are for public consumption. | 00:20 |
| mnewby | sdague: I mean, fine we have messy internals in various projects. But to have 3rd party programmers subjected to such terrible design is really unconscionable to me. | 00:21 |
| mnewby | sdague: and sadly, no amount of code review or other infrastructure can make up for poor design sense. | 00:21 |
| sdague | ok, well new attempt, less sledge hammer, up https://review.openstack.org/#/c/18098/ | 00:22 |
| mnewby | sdague: I've talked with jaypipes about the need for a sync flag for api clients... | 00:24 |
| mnewby | sdague: Everybody needs the capability at some point or another, and it would be great if it could live in the clients instead of having to be implemented by everyone else. | 00:25 |
| mnewby | sdague: I only mention that because I was implementing the same thing as your change for the base class for smoke testing. | 00:26 |
| *** nati_ueno has joined #openstack-dev | 00:27 | |
| *** yidclare has quit IRC | 00:28 | |
| *** bearovercloud has joined #openstack-dev | 00:29 | |
| *** juice_ has quit IRC | 00:33 | |
| *** alunduil has joined #openstack-dev | 00:34 | |
| *** lloydde has joined #openstack-dev | 00:36 | |
| *** jergerber has joined #openstack-dev | 00:40 | |
| *** dwalleck has joined #openstack-dev | 00:41 | |
| *** maxiz has joined #openstack-dev | 00:47 | |
| *** dwalleck has quit IRC | 00:48 | |
| *** dwalleck has joined #openstack-dev | 00:49 | |
| *** mrodden1 has quit IRC | 00:50 | |
| *** ndipanov has quit IRC | 00:53 | |
| *** dolphm has joined #openstack-dev | 00:56 | |
| *** dani4571 has quit IRC | 00:59 | |
| *** dolphm_ has joined #openstack-dev | 01:00 | |
| *** dolphm has quit IRC | 01:00 | |
| *** n000 has joined #openstack-dev | 01:00 | |
| *** gugl has joined #openstack-dev | 01:01 | |
| *** alexpilotti has joined #openstack-dev | 01:01 | |
| *** n000 is now known as koolhead11 | 01:04 | |
| cyeoh | can anyone tell me if I run tempest against devstack whether it should by default pass all the tests? If not are there certain config settings I need to get them to all pass or will some just always fail? | 01:04 |
| clarkb | cyeoh: I think there are two levels full and smoke. smoke should always pass and full passes most of the time | 01:05 |
| *** dolphm_ has quit IRC | 01:05 | |
| clarkb | because of this smoke is used in the gate | 01:05 |
| cyeoh | clarkb: thanks! | 01:06 |
| gatuus | boys!! | 01:07 |
| *** mrodden has joined #openstack-dev | 01:07 | |
| gatuus | boys!!! guess what!!!! I found a Cloudinit for WINDOWS!!!! weee!!!! =) =) =) | 01:07 |
| *** vipuls is now known as vipuls|away | 01:08 | |
| *** dwalleck has quit IRC | 01:10 | |
| *** vipuls|away is now known as vipuls | 01:11 | |
| gatuus | Its not the "cloudinit.net" it is a Cloudinit for windows for openstack!!! =) =) =) | 01:11 |
| *** gyee has quit IRC | 01:13 | |
| *** anniec has joined #openstack-dev | 01:16 | |
| *** anniec_ has joined #openstack-dev | 01:16 | |
| *** anniec has quit IRC | 01:21 | |
| *** anniec_ is now known as anniec | 01:21 | |
| *** zeriouz has joined #openstack-dev | 01:21 | |
| *** anniec has quit IRC | 01:23 | |
| *** anniec has joined #openstack-dev | 01:24 | |
| *** sacharya has joined #openstack-dev | 01:25 | |
| *** anniec has quit IRC | 01:26 | |
| *** ahardy_ has quit IRC | 01:29 | |
| *** novas0x2a|laptop has quit IRC | 01:31 | |
| *** zul has quit IRC | 01:32 | |
| *** gugl has quit IRC | 01:33 | |
| *** sacharya has quit IRC | 01:35 | |
| *** sacharya has joined #openstack-dev | 01:35 | |
| *** dwalleck has joined #openstack-dev | 01:39 | |
| *** crandquist has joined #openstack-dev | 01:39 | |
| *** dwalleck_ has joined #openstack-dev | 01:40 | |
| *** dwalleck_ has joined #openstack-dev | 01:41 | |
| *** markwash has quit IRC | 01:42 | |
| *** xchu has joined #openstack-dev | 01:43 | |
| *** crandquist has quit IRC | 01:43 | |
| *** dwalleck has quit IRC | 01:44 | |
| *** maxiz has quit IRC | 01:44 | |
| *** andrewbogott is now known as andrewbogott_afk | 01:48 | |
| *** dwalleck_ has quit IRC | 01:50 | |
| *** lloydde has quit IRC | 01:51 | |
| openstackgerrit | A change was merged to openstack/cinder: Add *.swp to gitignore. https://review.openstack.org/18074 | 01:51 |
| *** hillyu has joined #openstack-dev | 01:54 | |
| *** winston-d_ has joined #openstack-dev | 01:54 | |
| *** alexpilotti has quit IRC | 01:54 | |
| *** dwalleck has joined #openstack-dev | 01:58 | |
| *** dprince has joined #openstack-dev | 01:59 | |
| *** hillyu has quit IRC | 01:59 | |
| *** harlowja has quit IRC | 02:02 | |
| *** dims has quit IRC | 02:09 | |
| *** dolphm has joined #openstack-dev | 02:11 | |
| *** dolphm has quit IRC | 02:15 | |
| *** bing_bu has joined #openstack-dev | 02:15 | |
| *** dims has joined #openstack-dev | 02:23 | |
| *** gongysh has quit IRC | 02:31 | |
| dims | mordred, ah! will try that | 02:35 |
| *** tserong_ has joined #openstack-dev | 02:39 | |
| *** tserong_ has quit IRC | 02:39 | |
| *** tserong_ has joined #openstack-dev | 02:39 | |
| *** markmcclain has joined #openstack-dev | 02:40 | |
| *** jcooley is now known as jcooley|away | 02:40 | |
| *** Mandell has quit IRC | 02:41 | |
| *** tserong has quit IRC | 02:41 | |
| *** stevebake has quit IRC | 02:43 | |
| *** stevebake has joined #openstack-dev | 02:43 | |
| *** gongysh has joined #openstack-dev | 02:43 | |
| *** mnewby has quit IRC | 02:49 | |
| *** hillyu has joined #openstack-dev | 02:51 | |
| *** nati-ueno has joined #openstack-dev | 02:52 | |
| *** nati-ueno has quit IRC | 02:55 | |
| sdague | cyeoh: tempest should pass on a clean devstack | 02:55 |
| sdague | the volumes parts can sometimes be flakey, but the rest should be good. | 02:56 |
| *** zul has joined #openstack-dev | 02:56 | |
| *** aeperezt has quit IRC | 02:59 | |
| *** dprince has quit IRC | 03:00 | |
| *** adjohn has quit IRC | 03:02 | |
| *** aeperezt has joined #openstack-dev | 03:04 | |
| *** imsplitbit has joined #openstack-dev | 03:07 | |
| *** bearovercloud has left #openstack-dev | 03:08 | |
| *** imsplitbit has quit IRC | 03:10 | |
| *** hazmat has joined #openstack-dev | 03:12 | |
| *** hazmat has joined #openstack-dev | 03:12 | |
| *** stevebake has quit IRC | 03:14 | |
| *** stevebake has joined #openstack-dev | 03:14 | |
| *** stevebake has quit IRC | 03:19 | |
| *** stevebake has joined #openstack-dev | 03:19 | |
| *** almaisan-away is now known as al-maisan | 03:20 | |
| *** al-maisan is now known as almaisan-away | 03:20 | |
| *** stevebake has quit IRC | 03:22 | |
| *** stevebake has joined #openstack-dev | 03:22 | |
| *** davidlenwell has quit IRC | 03:22 | |
| *** vipuls is now known as vipul | 03:22 | |
| *** tserong_ is now known as tserong | 03:22 | |
| *** pixelbeat has quit IRC | 03:25 | |
| *** davidlenwell has joined #openstack-dev | 03:30 | |
| *** stevebake has quit IRC | 03:32 | |
| *** stevebake has joined #openstack-dev | 03:33 | |
| *** kpg2012 has joined #openstack-dev | 03:34 | |
| *** ajc_ has joined #openstack-dev | 03:35 | |
| openstackgerrit | A change was merged to openstack/nova: Fix for broken switch for config_drive https://review.openstack.org/18092 | 03:35 |
| *** dubsquared1 has joined #openstack-dev | 03:41 | |
| *** blamar has joined #openstack-dev | 03:41 | |
| *** dubsquared1 has quit IRC | 03:41 | |
| *** aeperezt has quit IRC | 03:45 | |
| openstackgerrit | A change was merged to openstack/nova: Imported Translations from Transifex https://review.openstack.org/17759 | 03:45 |
| *** markmcclain has quit IRC | 03:48 | |
| *** aeperezt has joined #openstack-dev | 03:49 | |
| *** kspear has quit IRC | 03:51 | |
| *** jergerber has quit IRC | 03:51 | |
| *** tacy_lee has joined #openstack-dev | 03:56 | |
| *** tacy_lee has left #openstack-dev | 03:56 | |
| openstackgerrit | A change was merged to openstack/nova: Add general mechanism for testing api coverage. https://review.openstack.org/18106 | 04:01 |
| *** martine has joined #openstack-dev | 04:01 | |
| *** kspear has joined #openstack-dev | 04:03 | |
| *** jimfehlig has quit IRC | 04:03 | |
| *** mlavalle has joined #openstack-dev | 04:06 | |
| *** mlavalle has left #openstack-dev | 04:07 | |
| openstackgerrit | A change was merged to openstack/quantum: Replaces assertEquals to assertEqual https://review.openstack.org/18104 | 04:07 |
| *** brich has joined #openstack-dev | 04:09 | |
| openstackgerrit | A change was merged to openstack/quantum: Completes coverage of quantum.api.v2.resource https://review.openstack.org/17841 | 04:09 |
| openstackgerrit | A change was merged to openstack/quantum: Fixed the unit tests using SQLite do not check foreign keys. https://review.openstack.org/17820 | 04:12 |
| *** sunilthaha has joined #openstack-dev | 04:12 | |
| *** adjohn has joined #openstack-dev | 04:13 | |
| *** zaneb has quit IRC | 04:15 | |
| *** wu_wenxiang has joined #openstack-dev | 04:17 | |
| *** adjohn has quit IRC | 04:17 | |
| *** martine_ has joined #openstack-dev | 04:24 | |
| *** martine has quit IRC | 04:24 | |
| *** brich has quit IRC | 04:27 | |
| *** stevebake has quit IRC | 04:27 | |
| *** roaet is now known as roaet-away | 04:27 | |
| *** stevebake has joined #openstack-dev | 04:27 | |
| *** jcooley|away is now known as jcooley | 04:28 | |
| *** stevemar has joined #openstack-dev | 04:29 | |
| cyeoh | sdague: thanks - I get a bunch of failures/errors even when using what I think are the same localrc settings. Just working through why now... | 04:31 |
| *** zaneb has joined #openstack-dev | 04:33 | |
| *** arata has joined #openstack-dev | 04:33 | |
| *** Mandell has joined #openstack-dev | 04:40 | |
| *** stevebake has quit IRC | 04:43 | |
| *** stevebake has joined #openstack-dev | 04:46 | |
| *** zeriouz has quit IRC | 04:46 | |
| *** danwent has joined #openstack-dev | 04:47 | |
| *** winston-d_ has quit IRC | 04:49 | |
| *** navid_ has joined #openstack-dev | 04:52 | |
| *** stevebake has quit IRC | 04:53 | |
| *** stevebake has joined #openstack-dev | 04:54 | |
| *** lloydde has joined #openstack-dev | 04:57 | |
| *** esp has joined #openstack-dev | 04:57 | |
| *** esp has left #openstack-dev | 04:59 | |
| *** zeriouz has joined #openstack-dev | 04:59 | |
| *** cfarquhar has joined #openstack-dev | 04:59 | |
| *** stevemar has quit IRC | 05:08 | |
| *** aeperezt has quit IRC | 05:09 | |
| *** aeperezt has joined #openstack-dev | 05:13 | |
| *** tacy_lee has joined #openstack-dev | 05:16 | |
| *** nati_ueno has quit IRC | 05:19 | |
| *** darjeeling has quit IRC | 05:21 | |
| *** nati-ueno has joined #openstack-dev | 05:24 | |
| *** nati-ueno has quit IRC | 05:29 | |
| *** nati-ueno has joined #openstack-dev | 05:30 | |
| *** nati-ueno is now known as nati_ueno | 05:31 | |
| *** nati_ueno_2 has joined #openstack-dev | 05:32 | |
| *** nati_ueno_2 has quit IRC | 05:32 | |
| *** nati_ueno has quit IRC | 05:32 | |
| *** nati_ueno has joined #openstack-dev | 05:33 | |
| *** nati_ueno has quit IRC | 05:34 | |
| *** mrodden1 has joined #openstack-dev | 05:37 | |
| *** mrodden has quit IRC | 05:39 | |
| *** maurosr has quit IRC | 05:39 | |
| *** nati_ueno has joined #openstack-dev | 05:40 | |
| *** mlavalle has joined #openstack-dev | 05:44 | |
| *** martine_ has quit IRC | 05:46 | |
| *** mlavalle has left #openstack-dev | 05:51 | |
| *** nati_ueno has quit IRC | 05:52 | |
| *** nati_ueno has joined #openstack-dev | 05:53 | |
| *** aeperezt has quit IRC | 05:56 | |
| *** Mandell has quit IRC | 06:01 | |
| *** tacy_lee has quit IRC | 06:02 | |
| *** Mandell has joined #openstack-dev | 06:04 | |
| *** nati_ueno has quit IRC | 06:05 | |
| *** sacharya has quit IRC | 06:06 | |
| *** nati_ueno has joined #openstack-dev | 06:06 | |
| *** agorodnev has joined #openstack-dev | 06:08 | |
| *** danwent has quit IRC | 06:12 | |
| *** tacy_lee has joined #openstack-dev | 06:15 | |
| *** nati_ueno has quit IRC | 06:16 | |
| *** nati_ueno has joined #openstack-dev | 06:20 | |
| *** nati_ueno has quit IRC | 06:28 | |
| *** colinmcnamara has joined #openstack-dev | 06:29 | |
| *** colinmcnamara has left #openstack-dev | 06:31 | |
| *** mnewby has joined #openstack-dev | 06:43 | |
| *** wenjianhn has joined #openstack-dev | 06:43 | |
| *** sthaha has quit IRC | 06:49 | |
| *** tacy_lee has left #openstack-dev | 06:50 | |
| *** nunosantos has quit IRC | 06:55 | |
| *** koolhead11 has quit IRC | 07:00 | |
| *** darjeeling has joined #openstack-dev | 07:03 | |
| *** jgriffith has quit IRC | 07:03 | |
| *** arata has quit IRC | 07:09 | |
| *** nati_ueno has joined #openstack-dev | 07:11 | |
| *** mindpixel has joined #openstack-dev | 07:14 | |
| *** juice_ has joined #openstack-dev | 07:18 | |
| *** mrunge has joined #openstack-dev | 07:19 | |
| *** nati_ueno has quit IRC | 07:20 | |
| *** CaptTofu has joined #openstack-dev | 07:26 | |
| *** maxiz has joined #openstack-dev | 07:27 | |
| *** zaitcev has quit IRC | 07:39 | |
| *** renier has joined #openstack-dev | 07:40 | |
| *** renier_ has quit IRC | 07:40 | |
| *** Mandell has quit IRC | 07:40 | |
| *** yolanda has joined #openstack-dev | 07:41 | |
| *** juice_ is now known as juice | 07:44 | |
| *** uvg has joined #openstack-dev | 07:45 | |
| *** ndipanov has joined #openstack-dev | 07:46 | |
| *** mnewby has quit IRC | 07:47 | |
| *** EmilienM has joined #openstack-dev | 07:47 | |
| *** EmilienM has quit IRC | 07:47 | |
| *** EmilienM has joined #openstack-dev | 07:48 | |
| *** juice has quit IRC | 07:51 | |
| *** EmilienM has quit IRC | 07:51 | |
| *** corXi has joined #openstack-dev | 07:52 | |
| *** henrynash_ has joined #openstack-dev | 07:53 | |
| *** jcooley is now known as jcooley|away | 07:55 | |
| *** henrynash has quit IRC | 07:56 | |
| *** henrynash_ is now known as henrynash | 07:56 | |
| ttx | notmyname: pong? | 07:56 |
| *** EmilienM has joined #openstack-dev | 07:58 | |
| *** EmilienM has quit IRC | 07:59 | |
| *** fc__ has joined #openstack-dev | 07:59 | |
| *** reidrac has joined #openstack-dev | 07:59 | |
| *** EmilienM_ has joined #openstack-dev | 08:00 | |
| *** amotoki has joined #openstack-dev | 08:02 | |
| *** rafaduran has joined #openstack-dev | 08:04 | |
| *** brucer has quit IRC | 08:05 | |
| *** nati_ueno has joined #openstack-dev | 08:07 | |
| openstackgerrit | A change was merged to openstack/cinder: Raise NotImplemented for drivers that don't support images https://review.openstack.org/18108 | 08:08 |
| *** mmagr has joined #openstack-dev | 08:12 | |
| *** renier_ has joined #openstack-dev | 08:18 | |
| *** renier has quit IRC | 08:18 | |
| *** maxiz has quit IRC | 08:19 | |
| *** halj has quit IRC | 08:22 | |
| *** zeriouz has quit IRC | 08:23 | |
| *** jcooley|away is now known as jcooley | 08:25 | |
| *** akscram1 is now known as akscram | 08:32 | |
| *** darjeeling has quit IRC | 08:35 | |
| *** m4xmr has joined #openstack-dev | 08:35 | |
| *** jcooley is now known as jcooley|away | 08:38 | |
| *** renier_ has quit IRC | 08:41 | |
| *** gael__ has joined #openstack-dev | 08:42 | |
| *** jpich has joined #openstack-dev | 08:43 | |
| *** dwalleck has quit IRC | 08:45 | |
| *** darjeeling has joined #openstack-dev | 08:48 | |
| *** nati_ueno_2 has joined #openstack-dev | 08:48 | |
| *** pixelbeat has joined #openstack-dev | 08:48 | |
| *** nati_ueno_2 has quit IRC | 08:50 | |
| *** e1mer has quit IRC | 08:50 | |
| *** nati_ueno_2 has joined #openstack-dev | 08:51 | |
| *** nati_ueno has quit IRC | 08:52 | |
| *** henrynash has quit IRC | 08:59 | |
| *** nati_ueno has joined #openstack-dev | 08:59 | |
| *** jgriffith has joined #openstack-dev | 09:01 | |
| *** nati_ueno_2 has quit IRC | 09:03 | |
| *** mindpixel has quit IRC | 09:04 | |
| *** jakedahn_zz has joined #openstack-dev | 09:07 | |
| *** renier has joined #openstack-dev | 09:08 | |
| *** boden has joined #openstack-dev | 09:12 | |
| *** lloydde has quit IRC | 09:19 | |
| *** jcooley|away is now known as jcooley | 09:19 | |
| *** lloydde has joined #openstack-dev | 09:20 | |
| *** jakedahn_zz has quit IRC | 09:20 | |
| *** lloydde has quit IRC | 09:24 | |
| *** nati_ueno has quit IRC | 09:25 | |
| *** afazekas has joined #openstack-dev | 09:25 | |
| *** nati_ueno has joined #openstack-dev | 09:25 | |
| *** alobbs has quit IRC | 09:25 | |
| *** alobbs has joined #openstack-dev | 09:26 | |
| *** derekh has joined #openstack-dev | 09:27 | |
| *** jcooley is now known as jcooley|away | 09:29 | |
| *** trapni has joined #openstack-dev | 09:34 | |
| *** jgriffith has quit IRC | 09:35 | |
| zykes- | asalkeld: continue here insteard i guess | 09:40 |
| asalkeld | where ever | 09:40 |
| zykes- | asalkeld: yeah, but what about where the image is stored, if it's not on shared storage, you're borked ? | 09:40 |
| zykes- | and heat, I know heat can do monitoring / ha | 09:40 |
| asalkeld | so your storage ha is a different beast | 09:40 |
| asalkeld | layer your ha | 09:41 |
| asalkeld | storage/instance | 09:41 |
| asalkeld | does xen do all that for you? | 09:41 |
| zykes- | instance ha is pretty simple, but atm the only ha storage alternatives are ceph or gluster or nfs | 09:41 |
| zykes- | asalkeld: i guess it has pretty kneat support for FC already | 09:42 |
| asalkeld | fc? | 09:42 |
| zykes- | fiberchannel | 09:42 |
| *** openpercept has joined #openstack-dev | 09:43 | |
| *** thimble has joined #openstack-dev | 09:44 | |
| *** CaptTofu has quit IRC | 09:48 | |
| zykes- | how else do you do that? | 09:52 |
| *** trapnii has joined #openstack-dev | 09:52 | |
| *** nati_ueno has quit IRC | 09:54 | |
| *** trapni has quit IRC | 09:56 | |
| *** xchu has quit IRC | 10:04 | |
| *** johngarbutt has joined #openstack-dev | 10:07 | |
| zykes- | johngarbutt: hey | 10:08 |
| *** trapnii has quit IRC | 10:10 | |
| johngarbutt | zykes: hi! | 10:10 |
| zykes- | johngarbutt: are you in the eu timezone or ? | 10:10 |
| *** kpg2012 has quit IRC | 10:11 | |
| johngarbutt | I am in the UK, near Cambridge | 10:11 |
| johngarbutt | where are u based? | 10:12 |
| zykes- | .no | 10:12 |
| *** radez_g0n3 has quit IRC | 10:12 | |
| *** uvg has quit IRC | 10:12 | |
| zykes- | johngarbutt: how good would the xenapi stuff be for FiberChannel support ? | 10:12 |
| *** trapnii has joined #openstack-dev | 10:13 | |
| johngarbutt | XenAPI has some good tested support for fiberchannel, however, it only really works with pools if XenServers/XCPs today | 10:13 |
| *** jcooley|away is now known as jcooley | 10:13 | |
| johngarbutt | why do you not want VMs on local disk? | 10:13 |
| zykes- | johngarbutt: HA | 10:14 |
| johngarbutt | openstack doesn't really do HA like that right now though, although we could go in that direction | 10:14 |
| zykes- | johngarbutt: yes it can, with heat & shared storage | 10:15 |
| johngarbutt | that is more server virt than cloud, so you are fighting what OpenStack is doing a little bit | 10:15 |
| johngarbutt | right heat certainly gets there | 10:15 |
| zykes- | :) | 10:15 |
| johngarbutt | is it using boot from volume? | 10:15 |
| zykes- | yeah, it is kinda like standard virtualization | 10:15 |
| *** radez_g0n3 has joined #openstack-dev | 10:15 | |
| zykes- | but you can make it easier for companies that already has shared storage in place :p | 10:15 |
| johngarbutt | right, if you use boot from volume, then OpenStack at least knows the disk is persistent, so that is good | 10:16 |
| zykes- | tons of companies here that has SANs from EMC, HP or Dell | 10:16 |
| johngarbutt | agreed | 10:16 |
| johngarbutt | cinder guys will know more on the current state of things | 10:16 |
| zykes- | and I highly doubt they'll be wanting to kill of those "cheap" investments : p | 10:16 |
| zykes- | johngarbutt: yeah, but would be cool if you can set a flag in nova that means it should use a specific pool in conder or something | 10:17 |
| johngarbutt | certainly if they have "free" storage around | 10:17 |
| zykes- | or volume type | 10:17 |
| johngarbutt | OK, so the problem is that OpenStack currently assumes you can attach a volume anywhere in your cloud | 10:17 |
| zykes- | johngarbutt: they may have SAN for OS disks but ISCSI / Ceph for something else | 10:17 |
| johngarbutt | yes, that does make sense | 10:17 |
| johngarbutt | unified orchestration | 10:17 |
| johngarbutt | kinda | 10:17 |
| zykes- | cinder will do this I think already with support for multiple storage backends :p | 10:17 |
| johngarbutt | volume types should give that soon, once the scheduler hits trunk | 10:18 |
| zykes- | jgriffith or kmartin does maybe know more about the state of this | 10:18 |
| johngarbutt | not 100% sure if that is in yet, I want to see it for some stuff I want to try too | 10:18 |
| zykes- | :) | 10:18 |
| johngarbutt | yes for sure jgriffith is the man on that stuff | 10:18 |
| zykes- | kmartin: has the pin on fc san stuff | 10:18 |
| johngarbutt | ah, col | 10:18 |
| johngarbutt | cool | 10:18 |
| johngarbutt | So, XenAPI current thinks it creates all the LUNs | 10:19 |
| johngarbutt | and that is basically the issue | 10:19 |
| zykes- | johngarbutt: basically the cool stuff would be to have the possibility to go to a customer and say "Hey you can deploy this within a few days / weeks and without the pain of new investments in hw immediately" | 10:19 |
| johngarbutt | right, got you | 10:19 |
| zykes- | that atm isn't that easy :p | 10:19 |
| johngarbutt | well unless you use Alamo or Piston right? | 10:20 |
| zykes- | they still need local storage don't they ? | 10:20 |
| johngarbutt | probably... I guess the existing server virt machines don't have big local disks then? | 10:20 |
| zykes- | johngarbutt: many people i think only have the smallest possible local disk and FC attached storage | 10:21 |
| johngarbutt | Alamo I think can use NFS for VM disks, but you would have to check with them on that | 10:21 |
| zykes- | like Blades f ex | 10:21 |
| johngarbutt | right | 10:21 |
| zykes- | some companies here gots tons of blade servers from dell / ibm / hp | 10:21 |
| johngarbutt | I heard sometimes the local disks on blades have lower IOPs that the connected SAN | 10:21 |
| zykes- | johngarbutt: really they might not even have local disk, just a SD card or similar | 10:22 |
| johngarbutt | right, I remember dell selling something like that with XenServer installed on a USB stick or something | 10:22 |
| zykes- | : o | 10:22 |
| johngarbutt | thanks for the background - very useful context | 10:22 |
| *** jcooley is now known as jcooley|away | 10:23 | |
| johngarbutt | so I think the current issue is along the lines of making sure XenAPI can attach to LUNs created by Cinder | 10:23 |
| *** alexpilotti has joined #openstack-dev | 10:23 | |
| johngarbutt | given those LUNs are connected with HBA or something like that? | 10:23 |
| *** mindpixel has joined #openstack-dev | 10:24 | |
| johngarbutt | zykes: are you expecting RAW LUNS or a LUN that contains VM disk files? | 10:26 |
| *** rushiagr has joined #openstack-dev | 10:27 | |
| *** bing_bu has quit IRC | 10:27 | |
| *** rushiagr is now known as rushiagr2 | 10:28 | |
| *** odyssey4me has joined #openstack-dev | 10:32 | |
| *** brucer has joined #openstack-dev | 10:35 | |
| *** enikanorov has quit IRC | 10:36 | |
| *** enikanorov has joined #openstack-dev | 10:37 | |
| *** boden_ has joined #openstack-dev | 10:39 | |
| *** boden has quit IRC | 10:39 | |
| *** Robbin has joined #openstack-dev | 10:39 | |
| *** irimi has joined #openstack-dev | 10:40 | |
| zykes- | johngarbutt: I guess for VM disks I could just make luns that holds a OS disk | 10:40 |
| Robbin | Is there any way to direct interact to OpenStack swift 'SWAUTH' middleware and get an account detail? Trying with curl with no luck. | 10:41 |
| *** darjeeling has quit IRC | 10:45 | |
| *** timello has joined #openstack-dev | 10:47 | |
| *** boden_ has quit IRC | 10:47 | |
| *** boden has joined #openstack-dev | 10:48 | |
| *** boden_ has joined #openstack-dev | 10:51 | |
| *** boden has quit IRC | 10:53 | |
| *** rushiagr2 has left #openstack-dev | 10:54 | |
| *** alexpilotti has quit IRC | 10:57 | |
| *** alexpilotti has joined #openstack-dev | 10:58 | |
| *** alexpilotti_ has joined #openstack-dev | 11:00 | |
| *** alexpilotti has quit IRC | 11:03 | |
| *** alexpilotti_ is now known as alexpilotti | 11:03 | |
| *** rushiagr_ has joined #openstack-dev | 11:03 | |
| *** alexpilotti has quit IRC | 11:06 | |
| *** alexpilotti has joined #openstack-dev | 11:06 | |
| *** jcooley|away is now known as jcooley | 11:07 | |
| *** timello has quit IRC | 11:08 | |
| *** hillyu has left #openstack-dev | 11:14 | |
| *** jcooley is now known as jcooley|away | 11:17 | |
| *** arbrandes has joined #openstack-dev | 11:18 | |
| *** maurosr has joined #openstack-dev | 11:18 | |
| *** m4xmr has left #openstack-dev | 11:18 | |
| *** trapnii has quit IRC | 11:28 | |
| *** yjiang5 is now known as yjiang5_away | 11:30 | |
| *** obondarev has joined #openstack-dev | 11:32 | |
| *** rushiagr_ has left #openstack-dev | 11:33 | |
| *** hillyu has joined #openstack-dev | 11:33 | |
| *** maurosr has quit IRC | 11:36 | |
| *** maurosr has joined #openstack-dev | 11:42 | |
| *** navid_ has left #openstack-dev | 11:43 | |
| *** markmc has joined #openstack-dev | 11:44 | |
| *** darjeeling has joined #openstack-dev | 11:44 | |
| sdague | cyeoh: you still kicking around? | 11:45 |
| *** mindpixel has quit IRC | 11:45 | |
| *** rafaduran1 has joined #openstack-dev | 11:49 | |
| *** rafaduran has quit IRC | 11:50 | |
| *** wu_wenxiang has quit IRC | 11:56 | |
| *** wenjianhn has left #openstack-dev | 11:59 | |
| zykes- | johngarbutt: whaet you think then ? : ) | 12:00 |
| *** jcooley|away is now known as jcooley | 12:02 | |
| *** tdumont has joined #openstack-dev | 12:05 | |
| johngarbutt | sorry, didn't spot the message | 12:09 |
| *** wu_wenxiang has joined #openstack-dev | 12:11 | |
| johngarbutt | it could work with luns for each disk | 12:11 |
| *** al1thomas has joined #openstack-dev | 12:11 | |
| johngarbutt | there will need to be a new storage type in XenAPI to support that though, I think | 12:11 |
| *** jcooley is now known as jcooley|away | 12:11 | |
| johngarbutt | zykes: going to run away for lunch I am afraid | 12:12 |
| zykes- | johngarbutt: ok :) | 12:12 |
| zykes- | johngarbutt: would you be willing to help push this with the cinder team and get the appropriate stuff into nova ? | 12:12 |
| *** wu_wenxiang has quit IRC | 12:14 | |
| *** al1thomas has left #openstack-dev | 12:15 | |
| *** martine has joined #openstack-dev | 12:16 | |
| johngarbutt | I fear I don't have the time at the moment :-( but I am happy to help with designs and things | 12:18 |
| johngarbutt | the real issue is changes in XCP that are needed | 12:18 |
| zykes- | ;p | 12:19 |
| zykes- | doesn't xcp or xen already support using FC stuff ? | 12:19 |
| zykes- | I thought they did like esxi with vcenter legacy virtualization setup | 12:19 |
| johngarbutt | XCP does, but not in a way openstack can use it right now | 12:19 |
| johngarbutt | don't think it is big, and its in python, and you can extend a running system, in XCP that is | 12:20 |
| johngarbutt | just not sure exactly what need to be done without taking some time out to look into into more | 12:20 |
| zykes- | why can't openstack use xenserver with fc support as it is ? | 12:20 |
| zykes- | it would be just like in vcenter you add more datastores that you store os disks on ? | 12:21 |
| johngarbutt | it requires XenServer pools to be useful, and XenServer assumes it is creating the LUNs, and most support uses a special LVHD mode to allow snapshots and things | 12:21 |
| zykes- | what's bad with pools ? | 12:22 |
| johngarbutt | does really match a real cloud deployment, the pool is quite small and wants to own the storage | 12:23 |
| johngarbutt | OpenStack doesn't really support the pool that well for that reason | 12:23 |
| zykes- | ah | 12:23 |
| zykes- | hmmm | 12:23 |
| zykes- | that kinda sucks ;p | 12:23 |
| johngarbutt | well in the cloud, you are best using lots of individual hypervisors, and local storage, just like Rackspace does with XenServer and OpenStack right? | 12:24 |
| zykes- | for public clouds yes... | 12:25 |
| zykes- | but as I said, I rarely think enterprises will re-invest and throw out existing storage.. | 12:25 |
| johngarbutt | well they don't have cloud enabled apps yet, they probably still want server virtualization to stay | 12:25 |
| *** tdumont has quit IRC | 12:26 | |
| johngarbutt | anyways, boot from volume gets you close | 12:26 |
| zykes- | johngarbutt: take this then, what about having support for multiple "instance storage" folders ? :) | 12:26 |
| zykes- | then you could just make a LUN and add it to a server as usual | 12:27 |
| zykes- | but it's still shared storage | 12:27 |
| zykes- | or boot from vol | 12:27 |
| johngarbutt | don't quite get you? XenServer does that today right, has loads of shared storage on a single server | 12:27 |
| johngarbutt | but you can only attach one shared storage thing to one host (or one pool) | 12:27 |
| johngarbutt | otherwise the locking goes crazy | 12:27 |
| zykes- | you mean shared storage for instances no ? | 12:28 |
| johngarbutt | with iSCSI we add a new storage repository for every volume/LUN, and that works great | 12:28 |
| johngarbutt | I guess when I say shared storage, I really mean "remote storage" | 12:28 |
| *** arbrandes has quit IRC | 12:28 | |
| zykes- | johngarbutt: I mean as you do with vmware today, you create multiple datastores where you typically store your instances, you can allocate instances to any of those | 12:29 |
| johngarbutt | unfortunately that doesn't work with XCP/XenServer quite like that today :-( | 12:29 |
| zykes- | oh really? I thought that you could make multipel storage repositories and store stuff on them ? | 12:30 |
| zykes- | a datastore being just 1 lun | 12:30 |
| *** ajc_ has quit IRC | 12:30 | |
| johngarbutt | you can make multiple, and it can just be 1 lun | 12:30 |
| zykes- | 1 lun pr store yes ? | 12:30 |
| johngarbutt | but you can only attach it to one hypervisor at once, in the general case | 12:30 |
| *** vkmc has joined #openstack-dev | 12:30 | |
| johngarbutt | I just need to work through the details before I can clarify more I am afriad | 12:31 |
| zykes- | ;) | 12:31 |
| johngarbutt | lots of details in the exact way you attach | 12:31 |
| zykes- | I'll ask a dude here if he knows, used to work with xenserver before ;) | 12:31 |
| johngarbutt | we have done NFS that way and just added that into cinder | 12:31 |
| johngarbutt | but it gets tricker with fiberchannel | 12:32 |
| zykes- | I would think they have support for it johngarbutt, even RH has that in RHEV | 12:32 |
| *** alexpilotti_ has joined #openstack-dev | 12:32 | |
| johngarbutt | I think we are talking at corss purposes | 12:32 |
| zykes- | corss ? :) | 12:33 |
| johngarbutt | it is supported, just the way you access it does not really quite fit with how OpenStack wants to do things | 12:33 |
| irimi | Are there *any* plans to migrate OpenStack to a more recent Python version ? | 12:33 |
| johngarbutt | irimi: worth searching the mailing list archive for that discussion, I think we are stuck on 2.7 for a little while for a set of reasons I can't remember, but I could be wrong | 12:34 |
| *** alexpilotti has quit IRC | 12:34 | |
| *** alexpilotti_ is now known as alexpilotti | 12:34 | |
| johngarbutt | zkyes: cross, no corss, oops | 12:34 |
| zykes- | ;) | 12:35 |
| irimi | johngarbutt: hm, okay, yeah, it's just that I'm not comfortable with the current set of (compute) scheduler filters and would like to just write my own, however, learning Python 2.7 is kind of... *hmm* when Python 3.3 is already out and there're not even plans in actually porting :) | 12:35 |
| zykes- | johngarbutt: i guess one could create multiple luns and attach them to servers and have nova just pick 1 randomly that has free space no ? | 12:35 |
| *** alexpilotti_ has joined #openstack-dev | 12:37 | |
| *** alexpilotti has quit IRC | 12:37 | |
| *** alexpilotti_ is now known as alexpilotti | 12:37 | |
| johngarbutt | zkyes: it is a bit manual for how nova volume works, you want users to request the lun creation, so you need some way of automating that action in your infrastructure, generally that involves having a Cinder driver for that storage array, and configuring it correctly, on the XenServer side, if you have an pool of servers, it does all that for you, but when you join the two up they just end up fighting, so you nee | 12:38 |
| zykes- | hmmm | 12:39 |
| johngarbutt | zykes: my keyboad skills are declining, this means its really is time for lunch, hopefully catch you later | 12:39 |
| zykes- | :) | 12:39 |
| *** johngarbutt has quit IRC | 12:39 | |
| *** alexpilotti_ has joined #openstack-dev | 12:41 | |
| *** alexpilotti has quit IRC | 12:45 | |
| *** alexpilotti_ is now known as alexpilotti | 12:45 | |
| *** openpercept has quit IRC | 12:46 | |
| *** alexpilotti_ has joined #openstack-dev | 12:48 | |
| *** alexpilotti has quit IRC | 12:51 | |
| *** alexpilotti_ has quit IRC | 12:51 | |
| *** alexpilotti has joined #openstack-dev | 12:51 | |
| *** jcooley|away is now known as jcooley | 12:56 | |
| *** Ruetobas has quit IRC | 13:00 | |
| *** alexpilotti has quit IRC | 13:00 | |
| *** alexpilotti has joined #openstack-dev | 13:00 | |
| *** alexpilotti has quit IRC | 13:01 | |
| *** alexpilotti has joined #openstack-dev | 13:02 | |
| *** markvoelker has joined #openstack-dev | 13:02 | |
| *** alexpilotti has quit IRC | 13:04 | |
| *** alexpilotti has joined #openstack-dev | 13:04 | |
| *** alexpilotti_ has joined #openstack-dev | 13:05 | |
| *** jcooley is now known as jcooley|away | 13:05 | |
| *** e1mer has joined #openstack-dev | 13:05 | |
| *** odyssey4me has quit IRC | 13:06 | |
| *** uvg has joined #openstack-dev | 13:06 | |
| *** Ruetobas has joined #openstack-dev | 13:07 | |
| uvg | Reviews required for: https://review.openstack.org/#/c/17039/ (Glance) and https://review.openstack.org/#/c/17847/ (Nova). Thanks | 13:07 |
| *** aloga has quit IRC | 13:07 | |
| *** aloga has joined #openstack-dev | 13:09 | |
| *** alexpilotti has quit IRC | 13:09 | |
| *** alexpilotti_ is now known as alexpilotti | 13:09 | |
| zykes- | any quantum folks around ? | 13:10 |
| *** brucer is now known as brucer_out | 13:12 | |
| *** tmishler has joined #openstack-dev | 13:13 | |
| *** dims has quit IRC | 13:13 | |
| *** smartinelli has joined #openstack-dev | 13:14 | |
| *** rkukura has joined #openstack-dev | 13:15 | |
| *** roaet-away is now known as roaet | 13:17 | |
| *** brucer_out has quit IRC | 13:17 | |
| *** alexpilotti_ has joined #openstack-dev | 13:19 | |
| *** wu-wenxiang has joined #openstack-dev | 13:20 | |
| *** nunosantos has joined #openstack-dev | 13:20 | |
| *** alexpilotti has quit IRC | 13:22 | |
| *** alexpilotti_ is now known as alexpilotti | 13:22 | |
| *** alunduil has quit IRC | 13:25 | |
| *** Ruetobas has quit IRC | 13:25 | |
| sdague | uvg: put a comment on there | 13:28 |
| sdague | for nova | 13:28 |
| *** mrunge has quit IRC | 13:28 | |
| *** dims has joined #openstack-dev | 13:30 | |
| *** Ruetobas has joined #openstack-dev | 13:31 | |
| *** timello has joined #openstack-dev | 13:33 | |
| *** thimble has quit IRC | 13:33 | |
| *** dims has quit IRC | 13:34 | |
| uvg | sdague: ok, thanks | 13:37 |
| *** AnilV4 has quit IRC | 13:40 | |
| *** gongysh has quit IRC | 13:44 | |
| *** gongysh has joined #openstack-dev | 13:46 | |
| *** dolphm has joined #openstack-dev | 13:47 | |
| *** e1mer has quit IRC | 13:49 | |
| *** jcooley|away is now known as jcooley | 13:50 | |
| *** aabes_ has joined #openstack-dev | 13:52 | |
| *** sandywalsh has joined #openstack-dev | 13:53 | |
| *** johngarbutt has joined #openstack-dev | 13:57 | |
| *** maurosr has quit IRC | 13:57 | |
| *** jcooley is now known as jcooley|away | 13:59 | |
| openstackgerrit | A change was merged to openstack/nova: Fix errors in used_limits extension https://review.openstack.org/18031 | 14:02 |
| *** henrynash has joined #openstack-dev | 14:02 | |
| *** e1mer has joined #openstack-dev | 14:04 | |
| *** blamar has quit IRC | 14:04 | |
| *** jergerber has joined #openstack-dev | 14:05 | |
| *** jergerber has quit IRC | 14:06 | |
| *** jergerber has joined #openstack-dev | 14:08 | |
| *** arbrandes has joined #openstack-dev | 14:09 | |
| zykes- | i/win 36 | 14:09 |
| *** martine has quit IRC | 14:11 | |
| *** hattwick has joined #openstack-dev | 14:11 | |
| *** mrodden1 has quit IRC | 14:13 | |
| *** jaypipes has joined #openstack-dev | 14:14 | |
| *** bknudson has joined #openstack-dev | 14:14 | |
| *** Ruetobas has quit IRC | 14:22 | |
| *** Ruetobas has joined #openstack-dev | 14:24 | |
| sdague | jaypipes: https://review.openstack.org/#/q/status:open+project:openstack/tempest+branch:master+topic:bug/1079687,n,z should be a more sane approach if you can look when you get a chance | 14:25 |
| *** dprince has joined #openstack-dev | 14:25 | |
| *** zing has joined #openstack-dev | 14:26 | |
| *** kpg2012 has joined #openstack-dev | 14:27 | |
| *** danwent has joined #openstack-dev | 14:27 | |
| *** mpdehaan has quit IRC | 14:27 | |
| *** mrodden1 has joined #openstack-dev | 14:28 | |
| *** mpdehaan has joined #openstack-dev | 14:28 | |
| henrynash | ayoung: question on ldap if you are around... | 14:29 |
| *** uvg has quit IRC | 14:29 | |
| *** e1mer has quit IRC | 14:30 | |
| *** mmalesky has quit IRC | 14:31 | |
| *** mmalesky has joined #openstack-dev | 14:31 | |
| *** dims has joined #openstack-dev | 14:33 | |
| *** irimi has quit IRC | 14:34 | |
| jaypipes | sdague: yep, will try to get to that this morning. thx! | 14:35 |
| *** alexpilotti has quit IRC | 14:36 | |
| *** sunilthaha has quit IRC | 14:36 | |
| *** timello has quit IRC | 14:37 | |
| *** gatuus has quit IRC | 14:38 | |
| *** danwent has quit IRC | 14:42 | |
| *** jcooley|away is now known as jcooley | 14:44 | |
| *** olaph has joined #openstack-dev | 14:46 | |
| *** alunduil has joined #openstack-dev | 14:47 | |
| *** irimi has joined #openstack-dev | 14:47 | |
| *** hillyu has left #openstack-dev | 14:48 | |
| *** radez_g0n3 is now known as radez | 14:49 | |
| *** alunduil has quit IRC | 14:50 | |
| *** alunduil has joined #openstack-dev | 14:51 | |
| *** mtreinish has joined #openstack-dev | 14:52 | |
| *** jcooley is now known as jcooley|away | 14:53 | |
| *** mrodden has joined #openstack-dev | 14:54 | |
| henrynash | dolphm: do you want me to merge the api changes? | 14:55 |
| *** gatuus has joined #openstack-dev | 14:55 | |
| *** rpedde_away is now known as rpedde | 14:55 | |
| *** jimfehlig has joined #openstack-dev | 14:56 | |
| *** blamar has joined #openstack-dev | 14:56 | |
| *** mrodden1 has quit IRC | 14:57 | |
| *** imsplitbit has joined #openstack-dev | 14:57 | |
| *** aeperezt has joined #openstack-dev | 14:57 | |
| dolphm | henrynash: i just meant put them up for review, rather than linking to an isolated doc -- they need to go in, and *should* go in first | 14:58 |
| dolphm | henrynash: personally, i'd be cool with blueprints linking to WIP code reviews as their "spec" (or whatever the blueprint system calls that) | 14:58 |
| henrynash | dolphm: should I do that by submitting (for review) changes to the api_identity spec? | 14:59 |
| henrynash | dolpm: where does the api doc live? | 15:00 |
| *** brich has joined #openstack-dev | 15:00 | |
| henrynash | dolphm: (found it) | 15:01 |
| *** annegentle_ has joined #openstack-dev | 15:01 | |
| *** edmund has joined #openstack-dev | 15:01 | |
| *** alexpilotti has joined #openstack-dev | 15:01 | |
| *** CaptTofu has joined #openstack-dev | 15:02 | |
| dolphm | henrynash: yes, to your first question | 15:02 |
| *** radez is now known as radez_g0n3 | 15:03 | |
| wu-wenxiang | dolphm: Could you help to review: https://review.openstack.org/#/c/18111/? | 15:03 |
| *** aloga has quit IRC | 15:04 | |
| *** bknudson has quit IRC | 15:04 | |
| *** amotoki has quit IRC | 15:04 | |
| openstackgerrit | A change was merged to openstack/keystone: fixing bug 1046862 https://review.openstack.org/18054 | 15:04 |
| uvirtbot | Launchpad bug 1046862 in keystone "Test failures in py26 environment" [Low,In progress] https://launchpad.net/bugs/1046862 | 15:04 |
| zykes- | henrynash: / dolphm with the v3 api will I be able to say make a Domain + Projects and make Groups and give specific Groups access to specofic Projects ? | 15:05 |
| *** aloga has joined #openstack-dev | 15:05 | |
| henrynash | zykes: yes | 15:05 |
| zykes- | holy crap, that's NICE! | 15:05 |
| *** aabes_ has left #openstack-dev | 15:06 | |
| dolphm | zykes-: i hope so, but i haven't read the proposed api changes ;P | 15:06 |
| henrynash | zykes: not in a single api call…but yes | 15:06 |
| zykes- | been kinda a lacking feature before :) | 15:06 |
| dolphm | zykes-: groups were actually supported in diablo to an extent, but no one used them, so we removed them until we had time to get it right | 15:07 |
| *** vkmc has quit IRC | 15:07 | |
| zykes- | dolphm: yeh | 15:07 |
| *** danwent has joined #openstack-dev | 15:07 | |
| zykes- | it's a very important piece though that's lacking though atm in contrast to vcloud / cloudstack i think no ? | 15:07 |
| *** yidclare has joined #openstack-dev | 15:09 | |
| *** jcru has joined #openstack-dev | 15:10 | |
| *** radez_g0n3 is now known as radez | 15:10 | |
| dolphm | wu-wenxiang: this can already be done trivially client-side (get resource by name -> delete resource by ID) ... what makes tenants special such that they need a special API for this? | 15:10 |
| dolphm | zykes-: definitely | 15:11 |
| henrynash | dolphm: quick question - not familiar with doc format for the api - can I use any text editor to edit the .md file? | 15:11 |
| dolphm | henrynash: yep, it's just markdown | 15:11 |
| openstackgerrit | A change was merged to openstack/nova: Fix use of osapi_compute_extension option in api_samples. https://review.openstack.org/18073 | 15:11 |
| zykes- | god I want Grizzly! | 15:11 |
| zykes- | gonna rock all the previos releases :) | 15:11 |
| dolphm | zykes-: we need to make it exist first :P | 15:12 |
| zykes- | dolphm: I know, but I just tried like G1 Quantum with Folsom bits, it's sooo much better in many ways | 15:12 |
| zykes- | and that's just 1 project out the all the others | 15:12 |
| dolphm | zykes-: i have a feeling a lot of these v3 features are going to be available, but aren't going to get dashboard support, etc, until H* | 15:12 |
| zykes- | awwwww | 15:13 |
| zykes- | kick gabriel around then ;) | 15:13 |
| dolphm | zykes-: it's not his fault lol | 15:13 |
| zykes- | hehe :) | 15:13 |
| zykes- | who's ? : p | 15:13 |
| wu-wenxiang | dolphm: Thanks, in https://bugs.launchpad.net/keystone/+bug/1087234, the reason was: Although it is possible to query the tenant id by its name and then send the delete command. It requires 2 request-response operations overloading the network. | 15:13 |
| uvirtbot | Launchpad bug 1087234 in keystone "It should be possible to delete a tenant by its name" [Medium,In progress] | 15:13 |
| dolphm | zykes-: keystone's! ;) | 15:14 |
| openstackgerrit | A change was merged to openstack/python-novaclient: Add optional argument to include reservations in os-used-limits https://review.openstack.org/17925 | 15:14 |
| zykes- | will it be hard to add it dolphm ? | 15:14 |
| henrynash | dolpm: I certainly plan to groups to the dashboard…. | 15:14 |
| *** hillyu has joined #openstack-dev | 15:14 | |
| henrynash | (plan to add groups to...) | 15:14 |
| zykes- | i'll send you a beer if you do henrynash :) | 15:14 |
| dolphm | zykes-: no, it's just that a lot of the work has to happen in sequence | 15:15 |
| zykes- | hehe | 15:16 |
| wu-wenxiang | dolpm: So user suggest to improve this by providing an API to delete a tenant by its name. This can be implemented by internally querying for tenant id. | 15:16 |
| zykes- | basically dolphm there's so much goodies in grizzly it's kinda mind boggling | 15:17 |
| *** danwent has quit IRC | 15:18 | |
| *** markmcclain has joined #openstack-dev | 15:19 | |
| openstackgerrit | A change was merged to openstack/keystone: syncing run_tests to match tox https://review.openstack.org/16696 | 15:20 |
| dolphm | wu-wenxiang: i haven't seen this bug before now, but it should really be a blueprint and requires updating the v2 administrative api docs (OS-KSCRUD, I think). it's also not symmetric with the rest of the v2 api which explicitly operates on ID's (there's nothing special about tenants), furthermore, I don't see much reason to add features to v2 when we're moving to v3 in the same release | 15:20 |
| *** gongysh has quit IRC | 15:20 | |
| *** aeperezt has quit IRC | 15:20 | |
| *** danwent has joined #openstack-dev | 15:21 | |
| *** bknudson has joined #openstack-dev | 15:21 | |
| *** markmcclain1 has joined #openstack-dev | 15:21 | |
| *** markmcclain has quit IRC | 15:21 | |
| zykes- | dolphm: is there much left to do or with the v3 api ? | 15:21 |
| *** torandu has joined #openstack-dev | 15:23 | |
| wu-wenxiang | dolphm: fine, so, do you mean we could add this features to v3? | 15:23 |
| *** sacharya has joined #openstack-dev | 15:24 | |
| *** markmcclain1 has quit IRC | 15:27 | |
| *** KurtMartin has joined #openstack-dev | 15:28 | |
| *** arbrandes has quit IRC | 15:28 | |
| *** kpg2012 has quit IRC | 15:28 | |
| *** eharney has joined #openstack-dev | 15:28 | |
| *** markmcclain has joined #openstack-dev | 15:28 | |
| *** TerryH has joined #openstack-dev | 15:29 | |
| dolphm | wu-wenxiang: absolutely... would you like to proposie an API convention for this on v3? | 15:32 |
| dolphm | zykes-: at this point, there's not even a proposed spec for auth on v3... and i'm starting to wonder if we're going to continue doing auth on v2 while doing administrative crud on v3 (as of grizzly's release) | 15:33 |
| *** aeperezt has joined #openstack-dev | 15:33 | |
| dolphm | zykes-: that wouldn't necessarily be a bad thing, as we need to maintain backwards compatibility anyway | 15:34 |
| *** sacharya has quit IRC | 15:34 | |
| dolphm | zykes-: (especially on v2 auth) | 15:34 |
| zykes- | so, components will still use v2 ? | 15:34 |
| dolphm | zykes-: *can* still use v2, yes | 15:35 |
| dolphm | zykes-: i'd like to get auth_token middleware consuming the client, and let the client decide if it should use v2 or v3 | 15:35 |
| zykes- | ok :) | 15:35 |
| dolphm | zykes-: all the v2 administrative stuff will be supported but deprecated as of grizzly | 15:36 |
| wu-wenxiang | dolphm: Yes, I would. Could you tell me what do I have to do? A blue point? Modify API doc? Or just coding? Thanks | 15:37 |
| *** irimi has quit IRC | 15:37 | |
| *** jcooley|away is now known as jcooley | 15:38 | |
| dolphm | wu-wenxiang: a blueprint on the topic would be appreciated, and then start with modifying identity-api as appropriate https://github.com/openstack/identity-api/blob/master/openstack-identity-api/src/markdown/identity-api-v3.md | 15:39 |
| *** irimi has joined #openstack-dev | 15:39 | |
| *** jaustinpage|want has joined #openstack-dev | 15:40 | |
| ayoung | henrynash, I am now | 15:40 |
| *** jrodom has joined #openstack-dev | 15:41 | |
| *** maurosr has joined #openstack-dev | 15:41 | |
| *** larsks_ is now known as larsks | 15:42 | |
| *** CaptTofu has quit IRC | 15:43 | |
| dolphm | wu-wenxiang: definitely give consideration to the fact that we'd like to eventually make project "name" a domain-scoped attribute, rather than something that is guaranteed to be globally unique (i think that's going to be hard without using composite ID's for projects in the API) | 15:44 |
| dolphm | henrynash: ^ | 15:44 |
| dolphm | e.g. DELETE /domain/{domain_id}/projects?name={project_name} | 15:44 |
| dolphm | /domains/... <-- should have been plural | 15:45 |
| *** CaptTofu has joined #openstack-dev | 15:45 | |
| *** jcru is now known as jcru|away | 15:45 | |
| *** jcooley is now known as jcooley|away | 15:49 | |
| *** aloga has quit IRC | 15:49 | |
| *** jcru|away is now known as jcru | 15:50 | |
| *** aloga has joined #openstack-dev | 15:51 | |
| *** larsks has quit IRC | 15:52 | |
| *** kbringard has joined #openstack-dev | 15:54 | |
| *** smartinelli_ has joined #openstack-dev | 15:55 | |
| wu-wenxiang | dolphm: Thanks, I'll have a try :) | 15:55 |
| *** hillyu has left #openstack-dev | 15:58 | |
| *** smartinelli has quit IRC | 15:59 | |
| *** pabelanger has joined #openstack-dev | 15:59 | |
| *** darraghb has joined #openstack-dev | 16:00 | |
| *** datsun180b has joined #openstack-dev | 16:00 | |
| *** Guest8021 has joined #openstack-dev | 16:01 | |
| *** Guest8021 is now known as aabes_ | 16:01 | |
| *** mikal has quit IRC | 16:04 | |
| ayoung | dolphm, when you get a chance https://review.openstack.org/#/c/17782/ | 16:04 |
| *** irimi has quit IRC | 16:05 | |
| *** mikal has joined #openstack-dev | 16:06 | |
| wu-wenxiang | dolphm: I just wonder, why DELETE /projects/{project_id} but /domains/{domain_id}/projects?name={project_name}? Project ID is globally unique while Project name is just unique inside a domain? | 16:06 |
| dolphm | ayoung: today is code review day for me, and it's on my list | 16:07 |
| dolphm | wu-wenxiang: right now, they're both globally unique, and all collections are defined at the root of the API | 16:08 |
| dolphm | wu-wenxiang: all names are also globally unique, but we'd like to change that for at least project names, if not user names as well (which are also owned by a domain) | 16:08 |
| *** irimi has joined #openstack-dev | 16:09 | |
| wu-wenxiang | ayoung: Could you help to review https://review.openstack.org/#/c/17661/? A tiny change :) | 16:09 |
| *** adamguest has left #openstack-dev | 16:10 | |
| *** adjohn has joined #openstack-dev | 16:11 | |
| *** mrodden has quit IRC | 16:14 | |
| *** otherwiseguy has quit IRC | 16:15 | |
| wu-wenxiang | dolphm: OK, sound reasonable :) Have a funny code review day! I have to ZZZZZZZZZZZZZ | 16:15 |
| *** sacharya has joined #openstack-dev | 16:15 | |
| ayoung | wu-wenxiang, looking | 16:15 |
| *** reidrac has left #openstack-dev | 16:15 | |
| wu-wenxiang | ayoung: Thanks~ | 16:16 |
| *** markmcclain has quit IRC | 16:16 | |
| ayoung | wu-wenxiang, is that change to make it consistent with somewhere else? | 16:16 |
| *** sacharya1 has joined #openstack-dev | 16:17 | |
| wu-wenxiang | ayoung: yes | 16:19 |
| ayoung | wu-wenxiang, update your commit message and I'll merge | 16:20 |
| wu-wenxiang | fine ~ | 16:20 |
| *** sacharya has quit IRC | 16:20 | |
| ayoung | wu-wenxiang, heh, it is probably a culture thing, but when someone says "fine" like that, it usually implies "fine, go ahead, do it, you will be sorry." But I don't think you mean it that way. | 16:21 |
| *** KurtMartin has quit IRC | 16:21 | |
| *** KurtMartin has joined #openstack-dev | 16:22 | |
| *** markmcclain has joined #openstack-dev | 16:22 | |
| openstackgerrit | A change was merged to openstack/cinder: Support glance servers over https https://review.openstack.org/17894 | 16:24 |
| *** mlavalle has joined #openstack-dev | 16:25 | |
| dolphm | ayoung: lol i would take "fine" as "that would not be my first preference, but i find it to be agreeable" | 16:27 |
| *** mlavalle has left #openstack-dev | 16:27 | |
| ayoung | dolphm, BTW, anything of yours need reviewing? | 16:27 |
| *** mlavalle has joined #openstack-dev | 16:27 | |
| ayoung | Trying to keep my account current. | 16:27 |
| SpamapS | Hrm, why is nova booting an m1.xlarge when I asked for an m1.large? http://paste.ubuntu.com/1438302/ | 16:28 |
| dolphm | ayoung: this is my big change right now https://review.openstack.org/#/c/17226/ | 16:28 |
| ayoung | dolphm, well, you know I am fan of Normalization. This should be a gimmie/ | 16:29 |
| *** bencherian has joined #openstack-dev | 16:29 | |
| dolphm | ayoung: i need to check on guang-yee's question -- because i *think* he's right for the service catalog, but wrong for the administrative API (emphasis on *think*) | 16:29 |
| *** cp16net is now known as cp16net|away | 16:30 | |
| ayoung | dolphm, I'll give it a review, and hold off on any merge until we get that nailed down | 16:30 |
| ayoung | dolphm, as far as URL lenght | 16:31 |
| ayoung | for example | 16:31 |
| ayoung | echo https://review.openstack.org/#/c/17226/5/keystone/catalog/backends/sql.py | wc | 16:31 |
| ayoung | 74 chars | 16:31 |
| *** sacharya has joined #openstack-dev | 16:31 | |
| *** aabes_ is now known as aabes | 16:31 | |
| dolphm | ayoung: the idea of a base endpoint being longer than 255 chars is crazy to me lol | 16:31 |
| *** torandu has quit IRC | 16:32 | |
| ayoung | I think 255 should be enough for endpoints. The question is whether they need to be varchar or if text is sufficient | 16:32 |
| ayoung | are we going to index on them? | 16:32 |
| *** sacharya1 has quit IRC | 16:32 | |
| dolphm | ayoung: i don't know why we would? | 16:32 |
| aabes | hmm... dns allows 255 long names, add to that some prefix url.. and you might get bust | 16:33 |
| aabes | unless that's the name, not the actual endpoint... | 16:33 |
| ayoung | dolphm, I am not DB expert, but it is my understanding that fixed length strings have a fixed allocation in the table structure, and clob/blob/text strings are done as a pointer into a text table. You really don't want to indexd on clobs, but if they're are not going to be indexed, we can make them into text fields and avoid the length discussion | 16:33 |
| *** adjohn has quit IRC | 16:34 | |
| ayoung | aabes, these are likely to be: https://fqdn:1234/service/v3.0/ | 16:34 |
| dolphm | ayoung: would that affect performance? they'll be pulled very frequently | 16:34 |
| aabes | ayoung, yup - that's the prefix/suffix i meant ;) | 16:35 |
| ayoung | dolphm, shouldn't affect performance. THe whole string table gets pulled in together, so it should be a wash | 16:35 |
| ayoung | dolphm, and, if we really are worried about that level of performance, we have won | 16:35 |
| ayoung | aabes, right, I was just trying to make it explicit | 16:35 |
| *** halj has joined #openstack-dev | 16:36 | |
| aabes | siily q: what's interface? | 16:36 |
| henrynash | ayoung: ldap support for user-groups….is this a requirement? Should I be going and attacking it? | 16:36 |
| aabes | guess I should poke at the api doc ;) | 16:36 |
| ayoung | so we could go with 255 + 28 if we want to be safe, and then we know we are within DNS + expected decorations | 16:36 |
| ayoung | henrynash, I would say yes. We might have to punt on it if it is impossible, but I suspect we can make it happen | 16:37 |
| ayoung | henrynash, user assigned to group is a common part of LDAP, so we can assume that will work. It is the role assignments that are particular to Keystone that need investigation | 16:38 |
| aabes | from the spec: "url": "http://external:8776/v1/--project-id--", | 16:38 |
| ayoung | henrynash, the object class for roles has the member attribute assuming it is a user. | 16:38 |
| aabes | that 28 seems short, given the projec ID | 16:38 |
| aabes | "project_id": "c5271357278e4a2094a96e0e6856c5cf", | 16:39 |
| ayoung | aabes, I didn't think the project ID was part of the endpoint URL | 16:39 |
| henrynash | ayoung: OK, I need to read up on ldap and see what object types we should map to…. | 16:39 |
| ayoung | henrynash, one sec, I can speed up your research | 16:39 |
| dolphm | aabes: interface=[internal, public, admin] | 16:39 |
| aabes | thx | 16:39 |
| dolphm | aabes: although we're making room for additional interfaces, so i didn't want to do an enumeration | 16:39 |
| *** sacharya1 has joined #openstack-dev | 16:40 | |
| aabes | hmm ayoung - right, you'd probablt just have $(tenant_id) / $(project_id) in the db, not actually the expanded value.. | 16:40 |
| dolphm | aabes: +1 | 16:40 |
| ayoung | henrynash, do you have openldap installed? | 16:41 |
| dolphm | %(project_id)s ;) | 16:41 |
| aabes | still a fixed limit feels wrong... for indexing - use the ID or some other field that's not actually communicated to end users.... | 16:41 |
| henrynash | dolphm: identity api doc changes for user groups is up for review: https://review.openstack.org/#/c/18138 | 16:41 |
| aabes | and let operators have flexibility in thier deploymnet.. | 16:41 |
| ayoung | aabes, so maybe 300 char. | 16:41 |
| aabes | i.e. any sized (within reason) endpoint url. | 16:41 |
| aabes | I'd say 4k. | 16:41 |
| henrynash | ayoung: probably not….sounds like job #1 | 16:42 |
| dolphm | soo.... just change it to sql.Text() ? | 16:42 |
| ayoung | aabes, I'd say you were nuts | 16:42 |
| ayoung | henrynash, ok | 16:42 |
| aabes | some do ;) | 16:42 |
| dolphm | henrynash: awesome, thanks | 16:42 |
| ayoung | henrynash, do that, but I'll fpaste the schema | 16:42 |
| henrynash | ayoung: thx | 16:42 |
| ayoung | henrynash, http://www.fpaste.org/64DU/ | 16:42 |
| aabes | bits are cheap ;) limits are restrictive ;) | 16:43 |
| ayoung | that is what we are using for roleAssignments by defautl | 16:43 |
| ayoung | aabes, that is per line in the DB | 16:43 |
| *** gyee has joined #openstack-dev | 16:43 | |
| *** sacharya has quit IRC | 16:43 | |
| aabes | k... 2k, http://stackoverflow.com/questions/417142/what-is-the-maximum-length-of-a-url | 16:43 |
| ayoung | aabes, I'm tempted to just say "text" and be done with it. If we find we need to index on it, we'll migrate and fix the length. OTOH, we should not really have that many Endpoints | 16:44 |
| ayoung | aabes, yeah, but we won;t have a full URL | 16:44 |
| *** cp16net|away is now known as cp16net | 16:44 | |
| *** tmishler has quit IRC | 16:44 | |
| dolphm | it's currently stored in text, so just for the purposes of a reliable migration it should go into another text column | 16:45 |
| ayoung | henrynash, role assignement is done by putting an organizationalRole object under the tenant, and then adding user entries in the roleOccupant value | 16:45 |
| ayoung | however, we probably want to distinguish between users and groups | 16:45 |
| ayoung | dolphm, +1 | 16:45 |
| henrynash | ayoung: ok | 16:45 |
| ayoung | henrynash, but according to the spec, when we create a group, does that implicitly create a role in every project? | 16:46 |
| ayoung | say we create a group named superadmins | 16:46 |
| ayoung | and superadmins have the role admin in all projects | 16:46 |
| ayoung | now if I add a user to superadmins, they should get the role admin for every project in the domain, right? | 16:47 |
| henrynash | ayoung: according to my spec? | 16:47 |
| wu-wenxiang | ayoung: Really, OK, I will use OK instead fine from now on when I want to explain "I agree with you" | 16:47 |
| henrynash | ayoung: yes | 16:48 |
| henrynash | ayoung: ahh. hold on | 16:48 |
| ayoung | henrynash, now, I assume there is a scoped down version where we say "explicitly add these projects to the group?" | 16:48 |
| henrynash | ayoung: no,answered too fast | 16:48 |
| openstackgerrit | A change was merged to openstack/nova: Allow xenapi to work with empty image metadata https://review.openstack.org/17814 | 16:49 |
| henrynash | ayoung: creating a group and adding users to it does nothing | 16:49 |
| ayoung | henrynash, so a group needs an explicit list of projects, or vice versa | 16:49 |
| ayoung | a project needs an explicit list of groups | 16:49 |
| *** danwent has quit IRC | 16:49 | |
| henrynash | ayoung: you can then add a role to a group-project pair | 16:50 |
| henrynash | ayoung: just like individual suers | 16:50 |
| henrynash | ayoung: you can use group_id wherever you can user_id in terms of role assignments to domains and projects | 16:50 |
| wu-wenxiang | The distances between openstack and we east-guys is time-zone. It's 00:50 AM, I have to ZZZZZZZ, thanks all of you! | 16:51 |
| ayoung | henrynash, the thing is, checking to see if an id is a group id or a user if is expensive | 16:51 |
| ayoung | wu-wenxiang, thanks for your work. Sleep well. | 16:51 |
| ayoung | henrynash, so I would like to keep groups and users in different attributes | 16:52 |
| *** wu-wenxiang has quit IRC | 16:52 | |
| *** Robbin has left #openstack-dev | 16:52 | |
| henrynash | ayoung: the apis are separate….I didn't mean you pass it to the same api…they are separate attributes in the apis | 16:53 |
| *** sandywalsh has quit IRC | 16:53 | |
| henrynash | have a quick look at: https://review.openstack.org/#/c/18138 | 16:53 |
| ayoung | henrynash, I realize that. I meant in storage. | 16:53 |
| henrynash | ayoung: sorry, ok, storage in terms of….? | 16:53 |
| ayoung | henrynash, one way we could store it is to just put the groups in the roleOccupant attribute | 16:54 |
| ayoung | henrynash, but then we would have to figure out which ones were groups. | 16:54 |
| henrynash | ayoung: Ok, had not realised you were talking ldap specifically in the iestions | 16:54 |
| henrynash | questions | 16:54 |
| ayoung | yes | 16:54 |
| *** aloga has quit IRC | 16:55 | |
| *** torandu has joined #openstack-dev | 16:55 | |
| ayoung | trying to get you accelerated through the decision process | 16:55 |
| henrynash | ayoung: right, see your thought process | 16:55 |
| *** irimi has quit IRC | 16:55 | |
| ayoung | henrynash, the thing is, a group will span multiple projects, so we can use a different objectclass to represent the role. Or something, | 16:56 |
| *** sacharya has joined #openstack-dev | 16:56 | |
| henrynash | ayoung: ok , i need to go study it a bit | 16:56 |
| ayoung | henrynash, it all comes from this RFC | 16:57 |
| ayoung | http://www.ietf.org/rfc/rfc2256.txt | 16:57 |
| ayoung | typically, a group is done as 7.10. groupOfNames | 16:57 |
| ayoung | and I did roles as 7.9. organizationalRole | 16:57 |
| *** sandywalsh has joined #openstack-dev | 16:58 | |
| henrynash | ayoung: yes, I took a quick look…and tenants/projects are groupOfNames as well | 16:58 |
| ayoung | the goal was to come up with a schema from the RFCs, and not to invent our own | 16:58 |
| henrynash | ayoung: +2 | 16:58 |
| ayoung | yes, object classes can be reused | 16:58 |
| ayoung | the queries start with a subtree, and are not recursive | 16:59 |
| ayoung | but I would assume groups would be their own subtree, parallel with users and projects | 16:59 |
| henrynash | yes | 16:59 |
| *** sacharya1 has quit IRC | 17:00 | |
| ayoung | then user-> group assignment would be done via the members value of the group | 17:00 |
| ayoung | member attribute | 17:00 |
| henrynash | seems fine | 17:00 |
| ayoung | henrynash, so the only issue is how to do the role->groups mapping | 17:00 |
| henrynash | what I'll probably do is a stub implementation quickly, since right now my unit tests for ldap fail for the WIP server code | 17:01 |
| ayoung | henrynash, when I got advice from the FreeIPA developers, they suggested that assignments like these should be done in their own subtree. THe problem with that is there is not a standard way to do that in LDAP. | 17:01 |
| ayoung | Now, roleOccupants could be done with the full DN of the entry, and from there we could tell if it was a user or a group, but I think that is messy. | 17:02 |
| henrynash | yes, agreed, I | 17:03 |
| *** ahardy_ has joined #openstack-dev | 17:03 | |
| *** vkmc has joined #openstack-dev | 17:03 | |
| henrynash | ayoung: Ok, will be away from kbd for a bit | 17:04 |
| henrynash | ayoung: fyi, the WIP user group server side changes are up for early review (supports sql & kvs) | 17:05 |
| henrynash | ayoung: https://review.openstack.org/#/c/18097/ | 17:06 |
| *** mnewby has joined #openstack-dev | 17:06 | |
| *** markwash has joined #openstack-dev | 17:07 | |
| *** EmilienM_ has left #openstack-dev | 17:08 | |
| ayoung | henrynash, I know, I've been looking at them | 17:09 |
| *** rafaduran1 has quit IRC | 17:10 | |
| *** colinmcnamara has joined #openstack-dev | 17:11 | |
| *** corXi has quit IRC | 17:11 | |
| *** colinmcnamara has quit IRC | 17:15 | |
| *** colinmcnamara has joined #openstack-dev | 17:15 | |
| *** lloydde has joined #openstack-dev | 17:15 | |
| *** colinmcnamara has quit IRC | 17:16 | |
| *** colinmcnamara has joined #openstack-dev | 17:16 | |
| *** markmc has quit IRC | 17:16 | |
| *** mrodden has joined #openstack-dev | 17:20 | |
| *** markvoelker has quit IRC | 17:21 | |
| *** dprince has quit IRC | 17:22 | |
| *** john5223 has joined #openstack-dev | 17:23 | |
| *** CaptTofu has quit IRC | 17:23 | |
| *** CaptTofu has joined #openstack-dev | 17:25 | |
| *** CaptTofu has quit IRC | 17:25 | |
| *** CaptTofu has joined #openstack-dev | 17:25 | |
| *** jcooley|away is now known as jcooley | 17:25 | |
| *** imsplitbit has quit IRC | 17:26 | |
| *** otherwiseguy has joined #openstack-dev | 17:26 | |
| jaypipes | sdague: https://review.openstack.org/#/c/18102/ off to the gate. | 17:28 |
| *** cp16net is now known as cp16net|away | 17:28 | |
| *** cp16net|away is now known as cp16net | 17:29 | |
| *** CaptTofu has quit IRC | 17:29 | |
| *** hemna_ has joined #openstack-dev | 17:31 | |
| *** KurtMartin has quit IRC | 17:34 | |
| *** bknudson has quit IRC | 17:35 | |
| openstackgerrit | A change was merged to openstack/python-keystoneclient: modify ca-certificate default value https://review.openstack.org/17661 | 17:36 |
| *** bencherian has quit IRC | 17:37 | |
| *** darraghb has quit IRC | 17:40 | |
| zykes- | mnewby: you here for a few questions ? | 17:41 |
| *** gael__ has quit IRC | 17:42 | |
| *** koolhead11 has joined #openstack-dev | 17:42 | |
| dolphm | can anyone (CI?) explain why hudson would *unassign* me from a bug? https://bugs.launchpad.net/keystone/+bug/1089987 | 17:42 |
| uvirtbot | Launchpad bug 1089987 in keystone "Non-API specific 404 exposes traceback" [Medium,In progress] | 17:42 |
| *** m4xmr has joined #openstack-dev | 17:43 | |
| dolphm | henrynash: anything you can do to help kill identity kvs would be appreciated ;) | 17:44 |
| *** adjohn has joined #openstack-dev | 17:44 | |
| dolphm | henrynash: i'm trying to get any tests dependent on identity kvs running against identity sql in-memory instead | 17:44 |
| *** bencherian has joined #openstack-dev | 17:45 | |
| *** nati_ueno has joined #openstack-dev | 17:45 | |
| openstackgerrit | A change was merged to openstack/swift: Add a license header to functionalnosetests https://review.openstack.org/15883 | 17:45 |
| *** cp16net is now known as cp16net|away | 17:46 | |
| *** cp16net|away is now known as cp16net | 17:47 | |
| *** bknudson has joined #openstack-dev | 17:47 | |
| *** mrodden1 has joined #openstack-dev | 17:48 | |
| *** gatuus has quit IRC | 17:48 | |
| *** adjohn_ has joined #openstack-dev | 17:49 | |
| *** adjohn has quit IRC | 17:49 | |
| *** bknudson has quit IRC | 17:50 | |
| *** mrodden has quit IRC | 17:51 | |
| *** esp1 has joined #openstack-dev | 17:52 | |
| *** frankm has quit IRC | 17:52 | |
| *** zing has quit IRC | 17:52 | |
| *** epim_ has joined #openstack-dev | 17:53 | |
| *** epim has quit IRC | 17:54 | |
| *** epim_ is now known as epim | 17:54 | |
| *** smartinelli_ has quit IRC | 17:56 | |
| *** dtroyer_zz has joined #openstack-dev | 17:57 | |
| *** jcmartin has joined #openstack-dev | 17:57 | |
| *** Mandell has joined #openstack-dev | 18:01 | |
| *** cfarquhar has quit IRC | 18:01 | |
| *** gatuus has joined #openstack-dev | 18:01 | |
| *** jgriffith has joined #openstack-dev | 18:03 | |
| *** jgriffit1 has joined #openstack-dev | 18:06 | |
| *** jgriffith has quit IRC | 18:06 | |
| *** jgriffit1 has quit IRC | 18:06 | |
| *** jgriffith has joined #openstack-dev | 18:06 | |
| *** epim has quit IRC | 18:07 | |
| *** epim has joined #openstack-dev | 18:07 | |
| *** johngarbutt has quit IRC | 18:09 | |
| *** torandu has quit IRC | 18:19 | |
| *** CaptTofu has joined #openstack-dev | 18:20 | |
| *** colinmcnamara has quit IRC | 18:22 | |
| *** harlowja has joined #openstack-dev | 18:22 | |
| *** zaneb has quit IRC | 18:23 | |
| *** maurosr has quit IRC | 18:23 | |
| *** ev0ldave has joined #openstack-dev | 18:23 | |
| *** zaitcev has joined #openstack-dev | 18:27 | |
| *** colinmcnamara has joined #openstack-dev | 18:28 | |
| *** derekh has quit IRC | 18:29 | |
| *** jcooley is now known as jcooley|away | 18:29 | |
| *** zaneb has joined #openstack-dev | 18:32 | |
| openstackgerrit | A change was merged to openstack/nova: Fixes KeyError: 'sr_uuid' when booting from volume on xenapi https://review.openstack.org/17726 | 18:34 |
| *** mmagr has quit IRC | 18:35 | |
| *** dprince has joined #openstack-dev | 18:40 | |
| *** gatuus has quit IRC | 18:42 | |
| *** imsplitbit has joined #openstack-dev | 18:54 | |
| *** hghazal has joined #openstack-dev | 18:55 | |
| *** hghazal has quit IRC | 18:55 | |
| *** gatuus has joined #openstack-dev | 18:55 | |
| *** hghazal has joined #openstack-dev | 18:56 | |
| *** andrewbogott_afk is now known as andrewbogott | 18:59 | |
| *** esp1 has left #openstack-dev | 18:59 | |
| *** kpg2012 has joined #openstack-dev | 18:59 | |
| *** rnirmal has joined #openstack-dev | 19:00 | |
| *** yolanda has quit IRC | 19:00 | |
| arosen | Anyone know where you can download the 2012.2.1 debian packages off hand? | 19:03 |
| arosen | I don't seem them on http://ubuntu-cloud.archive.canonical.com/ubuntu | 19:03 |
| sdague | jaypipes: thanks | 19:07 |
| sdague | yeh, the personality -> person thing was to deal with pep8 line length | 19:07 |
| sdague | not great, but a reasonable compromise | 19:07 |
| jaypipes | sdague: ah! :) | 19:08 |
| *** alexpilotti has quit IRC | 19:08 | |
| *** maurosr has joined #openstack-dev | 19:13 | |
| *** jpich has quit IRC | 19:20 | |
| *** jcru has quit IRC | 19:22 | |
| *** CaptTofu has quit IRC | 19:22 | |
| *** datsun180b has quit IRC | 19:23 | |
| ayoung | dolphm, so we have identity, which has a bunch of stuff lumped into it, and then we have things like policy which contain a single table. We have a bunch of new entities coming in, like mapping and Trusts, and I am trying to determine if they hsould be in separate top level modules, or if we should consolidate. | 19:25 |
| ayoung | I could see policy and catalog being put together. I can see an argument that tokens should be separate, especially considering how big the tables get | 19:25 |
| ayoung | dolphm, http://www.sebastien-han.fr/blog/2012/12/12/cleanup-keystone-tokens/ seems to imply that size of the token store is going to be something we have to track over time | 19:26 |
| ayoung | Trusts don't feel like they need their own backend, and neither does the mapping stuff. | 19:26 |
| ayoung | I'm wondering if they belong in with Policy, or if we should have policy with catalog, and then mapping, trusts, and the rest of the Kent work for ABAC go into a single backend. | 19:27 |
| ayoung | It seems to me that the dividing line should be between stuff that we have to split up for storage reasons. I could see and argument that the current service catalog might go in LDAP at some point, so that may make sense to keep separate. | 19:29 |
| ayoung | Would we be better splitting up everything, such that Identity could go in LDAP but everything else goes into SQL? We would still need a way to make service accounts, so that would not be sufficient for the AD use cases... | 19:30 |
| *** lloydde has quit IRC | 19:32 | |
| *** lloydde has joined #openstack-dev | 19:32 | |
| openstackgerrit | A change was merged to openstack/glance: Use io.BufferedIOBase.read() instead of io.BytesIO.getvalue(). https://review.openstack.org/17932 | 19:33 |
| *** ndipanov has quit IRC | 19:35 | |
| *** lloydde has quit IRC | 19:37 | |
| *** Nachi has joined #openstack-dev | 19:39 | |
| *** Nachi has quit IRC | 19:41 | |
| *** renier has quit IRC | 19:41 | |
| *** renier_ has joined #openstack-dev | 19:41 | |
| *** nati_uen_ has joined #openstack-dev | 19:41 | |
| *** nati_uen_ has quit IRC | 19:41 | |
| *** bencherian has quit IRC | 19:42 | |
| *** adjohn_ has quit IRC | 19:43 | |
| *** Nachi has joined #openstack-dev | 19:47 | |
| *** todin has quit IRC | 19:52 | |
| *** todin has joined #openstack-dev | 19:56 | |
| *** alexpilotti has joined #openstack-dev | 19:56 | |
| *** annegentle_ has quit IRC | 19:57 | |
| *** brich has quit IRC | 20:02 | |
| *** Nachi has quit IRC | 20:03 | |
| *** Nachi has joined #openstack-dev | 20:04 | |
| *** aeperezt has quit IRC | 20:09 | |
| *** nati_uen_ has joined #openstack-dev | 20:09 | |
| *** alexpilotti has quit IRC | 20:11 | |
| *** aeperezt has joined #openstack-dev | 20:11 | |
| openstackgerrit | A change was merged to openstack/swift: document correct config in dispersion-report help https://review.openstack.org/17750 | 20:12 |
| openstackgerrit | A change was merged to openstack/swift: Fix permissions from last commit https://review.openstack.org/18142 | 20:12 |
| openstackgerrit | A change was merged to openstack/keystone: Test for content-type appropriate 404 (bug 1089987) https://review.openstack.org/18051 | 20:13 |
| dolphm | ayoung: +1 for isolation by storage backend ... but everyone has a different use case ... so if anything, i'm all for breaking it up even further (i definitely think we'll be splitting identity into 2 or 3 separate drivers at some point) | 20:13 |
| uvirtbot | Launchpad bug 1089987 in keystone "Non-API specific 404 exposes traceback" [Medium,In progress] https://launchpad.net/bugs/1089987 | 20:13 |
| openstackgerrit | A change was merged to openstack/keystone: Imported Translations from Transifex https://review.openstack.org/17018 | 20:13 |
| mnewby | jaypipes: *ping* | 20:13 |
| *** Nachi has quit IRC | 20:13 | |
| jaypipes | mnewby: little swamped right now, but go ahead :) | 20:14 |
| ayoung | dolphm, OK, works for me. What we are going to need, I think, is a way to specify multiple SQL sources, and then to link the various backends up with the appropriate ones. Same, potentially with LDAP, and possibly to split those by domain. SOme blueprints for 'H' | 20:14 |
| *** anniec has joined #openstack-dev | 20:14 | |
| mnewby | jaypipes: nmind, no worries. :) i'll work on getting a quantum dev to +1 in the interim | 20:15 |
| jaypipes | mnewby: thx :) | 20:15 |
| *** otherwiseguy has quit IRC | 20:16 | |
| mnewby | markmcclain: *ping* | 20:17 |
| SpamapS | also is there any way to get a better error when something goes wrong? | 20:23 |
| SpamapS | just getting back a big fat 500 on the cli is pretty useless :p | 20:24 |
| *** novas0x2a|laptop has joined #openstack-dev | 20:25 | |
| sgran | adding --debug helps | 20:25 |
| *** anniec has quit IRC | 20:26 | |
| sgran | but yeah, agreed | 20:26 |
| SpamapS | oww | 20:26 |
| *** anniec has joined #openstack-dev | 20:26 | |
| *** otherwiseguy has joined #openstack-dev | 20:26 | |
| SpamapS | big giant server-side backtrace | 20:26 |
| sgran | that's usually the cause of the not very helpful 500 :) | 20:26 |
| *** yidclare1 has joined #openstack-dev | 20:30 | |
| ayoung | dolphm, know you've been bust. I'm ok if there are nits on https://review.openstack.org/#/c/17782/ but I keep rebasing the trust work on top of this, and I'd like to know if I should do so again, or if there are going to other shifts first. | 20:30 |
| dolphm | ayoung: *just* started digging into it actually | 20:31 |
| *** anniec has left #openstack-dev | 20:31 | |
| *** anniec has joined #openstack-dev | 20:32 | |
| *** zhiteng has joined #openstack-dev | 20:32 | |
| ayoung | dolphm, much appreciated | 20:32 |
| *** chalupaul has joined #openstack-dev | 20:34 | |
| *** john5223 has quit IRC | 20:34 | |
| *** jaustinpage|want has quit IRC | 20:35 | |
| *** aabes has quit IRC | 20:35 | |
| *** yidclare has quit IRC | 20:35 | |
| *** winston-d has quit IRC | 20:35 | |
| *** aabes has joined #openstack-dev | 20:35 | |
| *** chalupau1 has quit IRC | 20:35 | |
| *** jaustinpage|want has joined #openstack-dev | 20:36 | |
| *** anniec has left #openstack-dev | 20:37 | |
| *** anniec has joined #openstack-dev | 20:37 | |
| *** obondarev has quit IRC | 20:37 | |
| *** anniec has quit IRC | 20:38 | |
| *** ogelbukh has quit IRC | 20:38 | |
| *** obondarev has joined #openstack-dev | 20:39 | |
| *** aabes has quit IRC | 20:49 | |
| *** otherwiseguy has quit IRC | 20:49 | |
| *** ron-slc has quit IRC | 20:49 | |
| *** vipul has quit IRC | 20:49 | |
| *** JordanRi1ke has quit IRC | 20:49 | |
| *** srini_g has quit IRC | 20:49 | |
| *** otherwiseguy has joined #openstack-dev | 20:50 | |
| *** afazekas has quit IRC | 20:53 | |
| *** anniec has joined #openstack-dev | 21:00 | |
| *** anniec has quit IRC | 21:00 | |
| *** Ruetobas has quit IRC | 21:03 | |
| *** sacharya1 has joined #openstack-dev | 21:06 | |
| dolphm | ayoung: made lots of comments, mostly nits | 21:08 |
| dolphm | ayoung: my biggest gripe is where version & extensions stuff ended up -- neither are specific to a backend and it makes the most sense (to me) to put them in keystone.controllers and keystone.routers | 21:09 |
| *** sacharya has quit IRC | 21:09 | |
| dolphm | ayoung: i like most of it though :) | 21:09 |
| *** Ruetobas has joined #openstack-dev | 21:11 | |
| *** stevebake has quit IRC | 21:13 | |
| *** dubsquared1 has joined #openstack-dev | 21:16 | |
| *** dolphm has quit IRC | 21:17 | |
| *** stevebake has joined #openstack-dev | 21:18 | |
| *** nati_ueno has quit IRC | 21:21 | |
| *** nati_ueno has joined #openstack-dev | 21:22 | |
| *** danwent has joined #openstack-dev | 21:23 | |
| *** eharney has quit IRC | 21:24 | |
| *** sandywalsh has quit IRC | 21:28 | |
| *** sacharya has joined #openstack-dev | 21:35 | |
| *** sacharya1 has quit IRC | 21:35 | |
| *** blamar has quit IRC | 21:38 | |
| *** dubsquared1 has quit IRC | 21:42 | |
| *** rpedde is now known as rpedde_away | 21:42 | |
| *** melwitt has joined #openstack-dev | 21:50 | |
| *** ogelbukh has joined #openstack-dev | 21:50 | |
| *** aabes has joined #openstack-dev | 21:50 | |
| *** ron-slc has joined #openstack-dev | 21:50 | |
| *** vipul has joined #openstack-dev | 21:50 | |
| *** JordanRi1ke has joined #openstack-dev | 21:50 | |
| *** srini_g has joined #openstack-dev | 21:50 | |
| *** melwitt has quit IRC | 21:50 | |
| *** melwitt has joined #openstack-dev | 21:50 | |
| *** danwent has quit IRC | 21:52 | |
| *** boden_ has quit IRC | 21:54 | |
| *** cmagina has quit IRC | 21:55 | |
| *** cmagina has joined #openstack-dev | 21:57 | |
| *** bencherian has joined #openstack-dev | 21:59 | |
| *** adjohn has joined #openstack-dev | 22:02 | |
| *** brianr-a1ay has quit IRC | 22:07 | |
| *** radez is now known as radez_g0n3 | 22:18 | |
| *** anniec has joined #openstack-dev | 22:19 | |
| *** jaustinpage|want has quit IRC | 22:19 | |
| *** imsplitbit has quit IRC | 22:21 | |
| *** lloydde has joined #openstack-dev | 22:26 | |
| *** sacharya has quit IRC | 22:32 | |
| *** gongysh has joined #openstack-dev | 22:33 | |
| *** pabelanger has quit IRC | 22:36 | |
| *** roaet is now known as roaet-away | 22:39 | |
| *** halj has left #openstack-dev | 22:40 | |
| mnewby | jaypipes: +1 from a quantum core member, ready for your approval! | 22:43 |
| *** dprince has quit IRC | 22:45 | |
| *** adjohn has quit IRC | 22:48 | |
| *** martine has joined #openstack-dev | 22:49 | |
| openstackgerrit | A change was merged to openstack/quantum: Add restproxy.ini to config_path in setup.py. https://review.openstack.org/18135 | 22:51 |
| *** gyee has quit IRC | 22:55 | |
| *** henrynash has quit IRC | 22:56 | |
| *** jergerber has quit IRC | 23:00 | |
| *** otherwiseguy has quit IRC | 23:00 | |
| *** alunduil has quit IRC | 23:01 | |
| *** mtreinish has quit IRC | 23:01 | |
| *** enikanorov has quit IRC | 23:02 | |
| *** bencherian has quit IRC | 23:04 | |
| *** jrodom has quit IRC | 23:05 | |
| *** enikanorov has joined #openstack-dev | 23:06 | |
| *** almaisan-away is now known as al-maisan | 23:07 | |
| *** mmalesky has quit IRC | 23:07 | |
| *** anniec has quit IRC | 23:08 | |
| *** mmalesky has joined #openstack-dev | 23:08 | |
| *** kbringard has quit IRC | 23:09 | |
| gongysh | Thierry Carrez: ping | 23:10 |
| *** markwash has quit IRC | 23:16 | |
| *** markwash has joined #openstack-dev | 23:17 | |
| *** henrynash has joined #openstack-dev | 23:18 | |
| *** al-maisan is now known as almaisan-away | 23:23 | |
| *** TerryH has quit IRC | 23:24 | |
| *** kmartin has quit IRC | 23:25 | |
| *** kmartin has joined #openstack-dev | 23:25 | |
| *** kmartin has quit IRC | 23:25 | |
| *** kmartin has joined #openstack-dev | 23:26 | |
| *** jimfehlig has quit IRC | 23:27 | |
| *** imsplitbit has joined #openstack-dev | 23:28 | |
| *** hemna_ has quit IRC | 23:31 | |
| *** hemna_ has joined #openstack-dev | 23:31 | |
| *** bencherian has joined #openstack-dev | 23:32 | |
| *** mrodden1 has quit IRC | 23:32 | |
| *** mjfork has quit IRC | 23:32 | |
| *** rnirmal has quit IRC | 23:33 | |
| *** nati_uen_ has quit IRC | 23:33 | |
| openstackgerrit | A change was merged to openstack/glance: Refactor multi-tenant swift store. https://review.openstack.org/17652 | 23:33 |
| *** Nachi has joined #openstack-dev | 23:33 | |
| openstackgerrit | A change was merged to openstack/nova: Add a developer trap for api samples https://review.openstack.org/18146 | 23:33 |
| *** gyee has joined #openstack-dev | 23:33 | |
| *** annegentle_ has joined #openstack-dev | 23:34 | |
| *** TerryH has joined #openstack-dev | 23:38 | |
| bcwaldon | mordred: https://bugs.launchpad.net/bugs/1090592 | 23:38 |
| uvirtbot | Launchpad bug 1090592 in python-glanceclient "Glance client version incorrect" [Undecided,New] | 23:38 |
| *** edmund has quit IRC | 23:40 | |
| *** mjfork has joined #openstack-dev | 23:42 | |
| *** halj1 has joined #openstack-dev | 23:43 | |
| *** halj1 has left #openstack-dev | 23:43 | |
| *** zaitcev has quit IRC | 23:47 | |
| *** mrodden has joined #openstack-dev | 23:49 | |
| *** martine_ has joined #openstack-dev | 23:52 | |
| *** martine has quit IRC | 23:53 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!