Thursday, 2014-11-27

*** jeromatron has joined #magnetodb00:02
*** denis_makogon has quit IRC00:16
*** jeromatron has quit IRC00:27
*** charlesw has quit IRC02:16
*** charlesw has joined #magnetodb02:45
*** rushiagr_away is now known as rushiagr03:28
*** rushiagr is now known as rushiagr_away04:15
*** rushiagr_away is now known as rushiagr05:06
*** ajayaa has joined #magnetodb05:21
*** charlesw has quit IRC05:57
*** k4n0 has joined #magnetodb06:26
*** denis_makogon has joined #magnetodb08:25
*** romainh has joined #magnetodb09:08
rushiagrhello magnetoDB :)09:53
rushiagrI see a decorator on tests: @attr(type=['CreT-44'])09:54
rushiagrwhat does it mean?09:54
isviridovrushiagr hello10:55
isviridovit is test case code accroding to master test paln https://docs.google.com/a/mirantis.com/spreadsheets/d/1mZbJluIAUprA7tayiVe9owZXyEmAOipvIzPoY5Z44aE/edit#gid=192616503710:56
isviridovMeans Create Table functionality test case #4410:56
isviridovrushiagr such a process was used on first stages of development10:57
rushiagrisviridov: oh, cool10:57
rushiagrisviridov: maybe a mention of the spreadsheet somewhere in the code, or wiki would make it more obvious..10:58
isviridovominakov is it the latest version of this doc we have?10:58
ominakovisviridov, yep10:58
isviridovrushiagr I think yes, it should be. However currently process has changed and we are adding tests with feature at the same time10:59
isviridovominakov thx10:59
ominakovrushiagr, sounds good10:59
rushiagrisviridov: okay. I just want our code to make sense, otherwise wanderers like me will wonder what CreT-43 means. :)11:00
*** denis_makogon has quit IRC11:00
isviridovrushiagr great~11:03
openstackgerritDmitriy Ukhlov proposed stackforge/magnetodb-specs: Add cassandra-backend-implementation-redesign blueprint spec  https://review.openstack.org/13737911:29
openstackgerritAndrei V. Ostapenko proposed stackforge/magnetodb-specs: Adds table uuid in uri specification  https://review.openstack.org/13733612:25
*** k4n0 has quit IRC12:26
ajayaaikhudoshyn, aostapenko, Currently the code does not allow you to operate on projects which does not match context's project.13:00
ajayaaAnd that is right  because if you are not using keystone library to check existence of a project, you could potentially be creating a keyspace corresponding to a non-existent project in keystone.13:01
ajayaaAfter my patch, it would allow to operate on projects which exists in keystone but would never create a keyspace corresponding to a non existent project.13:03
ajayaaOnly problem is error code returned to the user.13:03
ajayaaAnd that should come in a different patch.13:03
ajayaaLine number 32 is never reached in current code.13:04
ajayaaIt reaches only if the url's project and context's project match.13:05
aostapenkoYou are right here, however if you run magnetodb with your code with policy allowing everything, and put valid token, that refers to tenant different from tenant in url, the tenant from token will be used for acting with resources. That is unexpected13:06
aostapenkoajayaa13:06
aostapenkoIn this case we need an additional check that tenant from url exists in keystone13:08
ajayaaYes. Exactly. And after that use project_id from url and perform operations on it.13:10
ajayaaAnd not user project_id from context.13:11
ajayaause*13:11
isviridovToday's agenda https://wiki.openstack.org/wiki/MagnetoDB/WeeklyMeetingAgenda#Nov_27.2C_2014.2C_14:00_UTC13:52
openstackgerritAndrei V. Ostapenko proposed stackforge/magnetodb-specs: Adds table uuid in uri specification  https://review.openstack.org/13733613:53
isviridovaostapenko you are fast. Please look at my comment about spec name13:57
isviridovHello everybody I believe we can start weekly meeting14:00
ikhudoshyno/14:00
isviridov#startmeeting magnetodb14:00
openstackMeeting started Thu Nov 27 14:00:19 2014 UTC and is due to finish in 60 minutes.  The chair is isviridov. Information about MeetBot at http://wiki.debian.org/MeetBot.14:00
openstackUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.14:00
openstackThe meeting name has been set to 'magnetodb'14:00
openstackgerritAndrei V. Ostapenko proposed stackforge/magnetodb-specs: Adds table uuid in uri specification  https://review.openstack.org/13733614:00
isviridovIt is holyday in US, so Symantec is not with us today14:00
isviridovikhudoshyn hi14:00
isviridovdukhlov are you with us?14:00
ominakovo/14:00
isviridovominakov o/14:01
dukhlov\o_14:01
* isviridov strange smile...14:01
isviridovLet us go througt action items from last meeting14:02
openstackgerritAndrei V. Ostapenko proposed stackforge/magnetodb-specs: Adds table uuid in uri specification  https://review.openstack.org/13733614:02
isviridov#topic Go through action items isviridov14:02
rushiagro/14:02
achuprin_o/14:02
aostapenkohi14:02
isviridovIt seems the only one was to review RBAC implementation14:02
isviridovrushiagr o/14:03
isviridovaostapenko ikhudoshyn ajayaa dukhlov anythin to discuss here?14:03
isviridov#link https://review.openstack.org/#/c/124391/14:03
ajayaaThere is a bug here. Unless we check the existence of a project with keystone, RBAC wouldn't work as expected.14:04
isviridovajayaa does it mean call to keystone on every request?14:04
ajayaaI am interested to know how other projects handle this situation, since most of other components work with project_ids in their URL.14:05
ajayaaisviridov, possibly!14:05
* isviridov surprised14:05
ajayaaisviridov, Anyway by using keystone tokens we are making a call to keystone.14:06
dukhlovwhy is current approach not good?14:06
isviridovWhat the thing is? Desn't PKI tiken contatin roles as well?14:06
dukhlovI mean checking that project_id == token's tenant14:06
isviridovajayaa it is expected to use PKI token and avoid call to ks14:06
ajayaaisviridov, PKI tokens are very long. Do you want the users to send an additional payload of 10 KB each time.14:07
ajayaaI think we discussed this sometime back.14:07
ajayaadukhlov, In that case you wouldn't be allowing an admin access to projects other than project embedded in his token.14:08
ajayaaBy admin, I mean someone who has access to all other projects.14:08
dukhlovajayaa, clear14:09
ajayaaAFAIK, most of the deployments use UUID tokens as of now, isviridov14:09
ajayaaThere was survey done by keystone-devs sometime back.14:09
dukhlovbut I thought that admin role allows user to get token for any tenant and then sent it to target service14:10
isviridovajayaa it was discussed before and we have decided that it is not bad, PKI token without catalog is not verly long and is about 1KB14:10
ikhudoshynajayaa: in fact we fought to avoid asking ks each time, so I think we do want PKI14:10
dukhlovmaybe your approach is good, but anyway we need to check how another services like nova handles this14:11
ajayaaikhudoshyn, sorry. perhaps I was not present in that discussion.14:11
ikhudoshynajayaa: np, its just fyi14:11
ajayaadukhlov, exactly. Let me see!14:11
ajayaamove on?14:12
isviridov#action ajayaa clarify auth in nova14:13
isviridovajayaa sounds good?14:13
ajayaayes14:15
ajayaaisviridov14:15
isviridovOk, move on14:15
isviridov#topic Authentication issues with monitoring API for third party services ominakov14:15
isviridovominakov?14:16
isviridovI believer I can start in behalh ominakov14:17
ominakovyep, as you know we have some issues with monitoring api14:17
isviridovYeap, please go on14:17
ominakovi describe issues and suggestions in https://blueprints.launchpad.net/magnetodb/+spec/api-uri-format-change14:17
isviridovThe thing is to make urls14:18
isviridovv1/data/<tenant_id>/...14:18
isviridovv1/monitoring/<tenant_id>/...14:18
isviridovthe different applications14:19
dukhlovlooks good!14:19
isviridovikhudoshyn?14:19
ikhudoshynagree14:19
ominakovi think, i can do this14:20
isviridovominakov bp has been approved. I believe documentaton should be also updated.14:20
ominakovisviridov, sure14:20
isviridovdukhlov ikhudoshyn do we need spec for this?14:21
ikhudoshynjust update existing docs14:21
dukhlovagree14:21
ikhudoshynBP with couple lines description would be enuff just to track activities14:22
isviridovOk, let us move on14:22
isviridovI see no ther topics in agenda except open discussion14:22
isviridov#topic  Open discussion isviridov14:22
aostapenkoLets see...14:23
ajayaaWhy do we want to create a separate application for dynamodb-api?14:23
isviridovaostapenko any progress with lookup table by uuid?14:23
aostapenkohttps://review.openstack.org/#/c/137336/ Here are specs14:24
isviridovajayaa in order to manage it separately during deployment. Balance requests, deploy on separate hardware so on.14:24
dukhlovwe have already created separate application for magnetoDB as far as I know14:24
isviridovajayaa but it is nice to have14:24
dukhlovI mean WSGI application14:25
isviridovdukhlov yes14:26
dukhlovand not we can deploy it with MagnetoDB API as composite application (using paste) or run as separate process14:26
ajayaadukhlov, okay!14:27
isviridovdukhlov with gunicorn deployment you are right, all port management is moved to higher level14:29
isviridovdukhlov that is why it is nice to have]14:29
isviridovdukhlov I mean to say, that separate process is not needed it this case, but separate WSGI app is14:31
isviridovaostapenko +2 to https://review.openstack.org/#/c/137336/14:31
dukhlovisviridov: at least we are providing deployment flexibility14:32
isviridovdukhlov agree14:32
isviridovDo we have anything else to discuss/highligt?14:33
aostapenkoisviridov thanks14:34
ikhudoshynisviridov: not from my side14:34
isviridovaostapenko ominakov ajayaa rushiagr?14:35
ajayaano from me. Thanks14:35
*** rushiagr has quit IRC14:35
ominakovnope14:35
isviridovThanks everybody for comming14:36
isviridov#endmeeting14:36
openstackMeeting ended Thu Nov 27 14:36:13 2014 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)14:36
openstackMinutes:        http://eavesdrop.openstack.org/meetings/magnetodb/2014/magnetodb.2014-11-27-14.00.html14:36
openstackMinutes (text): http://eavesdrop.openstack.org/meetings/magnetodb/2014/magnetodb.2014-11-27-14.00.txt14:36
openstackLog:            http://eavesdrop.openstack.org/meetings/magnetodb/2014/magnetodb.2014-11-27-14.00.log.html14:36
isviridovominakov bp has been approved (just added workitems)14:36
ominakovisviridov, ok14:37
openstackgerritAndrei V. Ostapenko proposed stackforge/magnetodb-specs: Adds table uuid in uri specification  https://review.openstack.org/13733614:51
openstackgerritAndrei V. Ostapenko proposed stackforge/magnetodb-specs: Adds table uuid in uri specification  https://review.openstack.org/13733614:53
*** rushiagr_away has joined #magnetodb15:04
*** rushiagr_away is now known as rushiagr15:11
rushiagrmy IRC bouncer was stuck by a DoS I think, so got disconnected :/15:13
*** charlesw has joined #magnetodb15:57
*** charlesw has quit IRC16:46
*** charlesw has joined #magnetodb17:06
*** rushiagr is now known as rushiagr_away17:11
*** charlesw has quit IRC17:38
*** ajayaa has quit IRC17:43
*** rushiagr_away is now known as rushiagr18:02
*** romainh has quit IRC18:15
*** miarmak has joined #magnetodb18:23
*** rushiagr is now known as rushiagr_away18:48
*** miarmak_ has joined #magnetodb18:51
*** miarmak has quit IRC18:51
*** rushiagr_away is now known as rushiagr18:56
*** charlesw has joined #magnetodb19:44
*** charlesw has quit IRC22:09
*** charlesw has joined #magnetodb22:12
*** denis_makogon has joined #magnetodb22:33
*** denis_makogon has quit IRC22:51

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!