| *** SlickNik has quit IRC | 00:13 | |
| *** SlickNik has joined #openstack-trove | 00:13 | |
| *** matsuhashi has joined #openstack-trove | 00:25 | |
| *** nosnos has joined #openstack-trove | 00:56 | |
| *** SlickNik has quit IRC | 01:08 | |
| *** SlickNik has joined #openstack-trove | 01:09 | |
| *** nosnos has quit IRC | 01:10 | |
| *** nosnos has joined #openstack-trove | 01:11 | |
| *** erkules_ has joined #openstack-trove | 01:18 | |
| *** erkules has quit IRC | 01:21 | |
| *** juice has quit IRC | 01:23 | |
| *** juice has joined #openstack-trove | 01:25 | |
| *** juice has joined #openstack-trove | 01:26 | |
| *** adrian_otto has joined #openstack-trove | 01:37 | |
| *** esp has joined #openstack-trove | 01:40 | |
| *** esp has left #openstack-trove | 01:40 | |
| *** adrian_otto has quit IRC | 02:12 | |
| *** simonmcc has quit IRC | 02:36 | |
| *** simonmcc has joined #openstack-trove | 02:37 | |
| *** shakayumi has quit IRC | 02:59 | |
| *** jasonb365 has joined #openstack-trove | 03:47 | |
| *** adrian_otto has joined #openstack-trove | 03:59 | |
| *** adrian_otto has quit IRC | 04:01 | |
| *** yogeshmehra has joined #openstack-trove | 04:10 | |
| *** shakayumi has joined #openstack-trove | 04:23 | |
| *** shakayumi has quit IRC | 04:27 | |
| *** shakayumi has joined #openstack-trove | 04:29 | |
| *** haomaiwang has quit IRC | 04:30 | |
| *** shakayumi has quit IRC | 04:39 | |
| *** jasonb365 has quit IRC | 04:40 | |
| *** dmakogon has joined #openstack-trove | 05:03 | |
| *** erkules_ is now known as erkules | 05:06 | |
| *** adrian_otto has joined #openstack-trove | 05:12 | |
| *** matsuhashi has quit IRC | 05:20 | |
| *** matsuhashi has joined #openstack-trove | 05:21 | |
| *** adrian_otto has quit IRC | 05:34 | |
| *** nosnos has quit IRC | 06:00 | |
| *** nosnos_ has joined #openstack-trove | 06:00 | |
| *** matsuhas_ has joined #openstack-trove | 06:01 | |
| *** matsuhashi has quit IRC | 06:01 | |
| *** SnowDust has joined #openstack-trove | 06:06 | |
| SnowDust | dmakogon: hiiii | 06:06 |
|---|---|---|
| SnowDust | dmakogon: can u review plz .. https://review.openstack.org/#/c/51836/ | 06:09 |
| dmakogon | SnowDust, dobe | 06:15 |
| dmakogon | SnowDust, done | 06:15 |
| SnowDust | tx ! | 06:15 |
| SnowDust | let me see | 06:15 |
| *** krow has joined #openstack-trove | 06:21 | |
| *** ashestakov has joined #openstack-trove | 06:27 | |
| *** matsuhas_ has quit IRC | 06:44 | |
| *** matsuhashi has joined #openstack-trove | 06:45 | |
| *** nosnos has joined #openstack-trove | 06:47 | |
| *** nosnos_ has quit IRC | 06:48 | |
| *** matsuhashi has quit IRC | 06:49 | |
| *** SushilKM has joined #openstack-trove | 06:51 | |
| SushilKM | Denis around | 06:51 |
| *** matsuhashi has joined #openstack-trove | 07:02 | |
| dmakogon | SushilKM, yes | 07:16 |
| *** SlickNik has quit IRC | 07:24 | |
| *** SlickNik has joined #openstack-trove | 07:24 | |
| *** dmakogon is now known as away_guy | 07:31 | |
| *** away_guy has quit IRC | 07:36 | |
| *** ashestakov has quit IRC | 08:13 | |
| *** SnowDust has quit IRC | 08:15 | |
| *** __dedi__ has joined #openstack-trove | 08:22 | |
| __dedi__ | hi folks | 08:22 |
| __dedi__ | is it possible the securitygroup manipulation with trove-cli? | 08:26 |
| *** matsuhashi has quit IRC | 08:27 | |
| *** matsuhashi has joined #openstack-trove | 08:28 | |
| __dedi__ | oops I find it. | 08:30 |
| *** dmakogon has joined #openstack-trove | 08:32 | |
| __dedi__ | I have a running instance but I cannot get is security group information with 'trove-cli secgroup list' : -> SecurityGroupInstanceAssociation Not Found (HTTP 404) | 08:32 |
| *** SushilKM has quit IRC | 08:45 | |
| *** yogeshmehra has quit IRC | 08:50 | |
| *** __dedi__ has quit IRC | 08:57 | |
| *** nosnos has quit IRC | 09:01 | |
| *** nosnos has joined #openstack-trove | 09:01 | |
| *** ashestakov_ has quit IRC | 09:05 | |
| *** ashestakov has joined #openstack-trove | 09:16 | |
| openstackgerrit | Denis M. proposed a change to openstack/trove-integration: Initial support for single instance Cassandra Database https://review.openstack.org/52666 | 09:24 |
| *** ppenjoy has joined #openstack-trove | 09:26 | |
| *** ppenjoy has left #openstack-trove | 09:26 | |
| *** SushilKM has joined #openstack-trove | 09:28 | |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Initial support for single instance Cassandra Database https://review.openstack.org/51884 | 09:30 |
| dmakogon | gate Jenkins is broken | 09:36 |
| *** aignatov has joined #openstack-trove | 09:43 | |
| *** krow has quit IRC | 10:00 | |
| *** krow has joined #openstack-trove | 10:16 | |
| *** matsuhashi has quit IRC | 10:32 | |
| *** matsuhashi has joined #openstack-trove | 10:32 | |
| *** krow has quit IRC | 10:35 | |
| *** matsuhashi has quit IRC | 10:37 | |
| *** pdmars has joined #openstack-trove | 12:04 | |
| *** pdmars has quit IRC | 12:05 | |
| *** radez_g0n3 is now known as radez | 12:23 | |
| *** radez is now known as radez_g0n3 | 12:23 | |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Datastore backup flexibility https://review.openstack.org/52905 | 12:27 |
| *** radez_g0n3 is now known as radez | 12:29 | |
| *** pdmars has joined #openstack-trove | 12:31 | |
| *** demorris has joined #openstack-trove | 12:59 | |
| *** robertmyers has joined #openstack-trove | 13:04 | |
| *** SushilKM has quit IRC | 13:07 | |
| *** nosnos has quit IRC | 13:09 | |
| *** nosnos has joined #openstack-trove | 13:10 | |
| *** nosnos has quit IRC | 13:14 | |
| *** haomaiwang has joined #openstack-trove | 13:42 | |
| *** grapex has joined #openstack-trove | 13:54 | |
| *** kevinconway has quit IRC | 13:57 | |
| *** shakayumi has joined #openstack-trove | 14:01 | |
| *** grapex has quit IRC | 14:02 | |
| *** grapex has joined #openstack-trove | 14:03 | |
| *** demorris has quit IRC | 14:04 | |
| *** demorris has joined #openstack-trove | 14:05 | |
| *** datsun180b has joined #openstack-trove | 14:07 | |
| *** jcru has joined #openstack-trove | 14:07 | |
| *** kevinconway has joined #openstack-trove | 14:07 | |
| *** haomaiwang has quit IRC | 14:14 | |
| *** haomaiwang has joined #openstack-trove | 14:14 | |
| *** SnowDust has joined #openstack-trove | 14:24 | |
| *** haomaiwa_ has joined #openstack-trove | 14:26 | |
| *** haomaiwa_ has quit IRC | 14:27 | |
| *** haomaiwa_ has joined #openstack-trove | 14:29 | |
| *** robertmyers has quit IRC | 14:30 | |
| *** haomaiwang has quit IRC | 14:30 | |
| *** robertmyers has joined #openstack-trove | 14:30 | |
| *** rnirmal has joined #openstack-trove | 14:37 | |
| *** jasonb365 has joined #openstack-trove | 14:43 | |
| *** Barker has joined #openstack-trove | 14:44 | |
| *** SushilKM has joined #openstack-trove | 14:50 | |
| *** shakayumi has quit IRC | 14:50 | |
| *** Barker has quit IRC | 14:52 | |
| *** shakayumi has joined #openstack-trove | 14:53 | |
| *** adrian_otto has joined #openstack-trove | 15:05 | |
| *** Barker has joined #openstack-trove | 15:05 | |
| *** SushilKM has quit IRC | 15:11 | |
| *** paul_lodronio has joined #openstack-trove | 15:14 | |
| *** yidclare has quit IRC | 15:20 | |
| *** SnowDust has quit IRC | 15:20 | |
| *** SnowDust has joined #openstack-trove | 15:21 | |
| *** radez is now known as radez_g0n3 | 15:21 | |
| *** dmakogon has quit IRC | 15:23 | |
| *** tanisdl has joined #openstack-trove | 15:30 | |
| *** dmakogon has joined #openstack-trove | 15:32 | |
| *** jmontemayor has joined #openstack-trove | 16:00 | |
| *** yidclare has joined #openstack-trove | 16:08 | |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Initial support for single instance Cassandra Database https://review.openstack.org/51884 | 16:09 |
| *** haomaiwa_ has quit IRC | 16:12 | |
| *** haomaiwang has joined #openstack-trove | 16:12 | |
| *** rnirmal_ has joined #openstack-trove | 16:15 | |
| *** haomaiwang has quit IRC | 16:16 | |
| *** SnowDust has quit IRC | 16:18 | |
| *** rnirmal_ has quit IRC | 16:20 | |
| *** rnirmal has quit IRC | 16:20 | |
| *** rnirmal has joined #openstack-trove | 16:20 | |
| *** SushilKM has joined #openstack-trove | 16:28 | |
| *** adrian_otto has quit IRC | 16:29 | |
| *** esp has joined #openstack-trove | 16:31 | |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Initial support for single instance Cassandra Database https://review.openstack.org/51884 | 16:34 |
| *** adrian_otto has joined #openstack-trove | 16:36 | |
| *** adrian_otto has quit IRC | 16:36 | |
| *** adrian_otto has joined #openstack-trove | 16:39 | |
| *** adrian_otto has quit IRC | 16:40 | |
| *** rnirmal_ has joined #openstack-trove | 16:47 | |
| *** rnirmal has quit IRC | 16:50 | |
| *** rnirmal_ is now known as rnirmal | 16:50 | |
| *** jasonb365 has quit IRC | 17:01 | |
| *** jasonb365 has joined #openstack-trove | 17:06 | |
| *** rnirmal has quit IRC | 17:06 | |
| *** rnirmal has joined #openstack-trove | 17:06 | |
| *** ashestakov has quit IRC | 17:09 | |
| *** ashestakov has joined #openstack-trove | 17:10 | |
| *** jasonb365 has quit IRC | 17:11 | |
| *** yogeshmehra has joined #openstack-trove | 17:13 | |
| *** SushilKM__ has joined #openstack-trove | 17:17 | |
| *** SushilKM has quit IRC | 17:17 | |
| hub_cap | kevinconway: around? | 17:27 |
| *** rnirmal has quit IRC | 17:30 | |
| *** rnirmal has joined #openstack-trove | 17:31 | |
| dmakogon | hub_cap, sup)))) | 17:38 |
| dmakogon | hub_cap, we missed you | 17:39 |
| hub_cap | hi! | 17:39 |
| hub_cap | ya i took a good small break | 17:39 |
| dmakogon | hub_cap, https://review.openstack.org/#/c/52905/ ||| https://review.openstack.org/#/c/50686/ | 17:39 |
| dmakogon | hub_cap, here you go))) | 17:39 |
| hub_cap | dmakogon: i tried to run your cassandra code | 17:39 |
| hub_cap | it didnt work at all | 17:39 |
| dmakogon | hub_cap, because it need pre-defined image | 17:40 |
| dmakogon | cloud-init takes 12 min to acomplish | 17:40 |
| hub_cap | no the code itself | 17:40 |
| hub_cap | like teh guest code was getting errors having to do with the code | 17:40 |
| dmakogon | hub_cap, where it brokes ? | 17:40 |
| hub_cap | try to run it again | 17:41 |
| hub_cap | see if it is still running | 17:41 |
| dmakogon | hub_cap, try last patches | 17:41 |
| hub_cap | i tried this on thursday | 17:41 |
| hub_cap | so like, 4~ days ago | 17:41 |
| dmakogon | wow | 17:41 |
| dmakogon | it was old | 17:41 |
| dmakogon | now it works 100% | 17:41 |
| dmakogon | but it still need image | 17:42 |
| dmakogon | i'm working on creating manual for image | 17:42 |
| hub_cap | k | 17:42 |
| dmakogon | and i'll publish it on OpenStack wiki | 17:42 |
| hub_cap | k. so this wont work with vanilla image? | 17:43 |
| hub_cap | i thought you wanted them all to work w/ vanilla image? | 17:44 |
| dmakogon | hub_cap, basic vanilla image of ubuntu has strange sources | 17:44 |
| dmakogon | almost all packages are failed to be installed | 17:44 |
| dmakogon | that is why i need basic Ubuntu server 12.04 source list | 17:45 |
| dmakogon | apt-get update takes a lot of time | 17:45 |
| hub_cap | heh | 17:45 |
| hub_cap | yes it does :) | 17:45 |
| *** demorris has quit IRC | 17:45 | |
| ashestakov | dmakogon: can you describe problem with sources? | 17:45 |
| hub_cap | thats why we pre-install the packages at image build time | 17:45 |
| dmakogon | and after customizing repositories with cassandra repo it also needed to run apt-get update | 17:46 |
| hub_cap | \\\\\\\\\\\\\ | 17:46 |
| hub_cap | thats my son\] | 17:46 |
| hub_cap | he found my keyboard | 17:46 |
| dmakogon | ashestakov, in sources.list nova repositories missing some kind of deps | 17:47 |
| dmakogon | even when i do simple apt-get update through cloud-init it fails | 17:47 |
| ashestakov | dmakogon: then setup additional mirrors if you need | 17:47 |
| dmakogon | that is why i'm working on custom image | 17:47 |
| hub_cap | if a apt-get update through cloud-init fails you have other issues :) | 17:48 |
| dmakogon | ashestakov, than is what i'm really doing | 17:48 |
| hub_cap | that should never fail | 17:48 |
| ashestakov | i tried few recent images from ubuntu-images and everithing works | 17:48 |
| dmakogon | hub_cap, that is why i'm using default ubuntu server 12.04 sources.list | 17:48 |
| dmakogon | ashestakov, but it doen't work with mine cloud-init and also the same with mongo | 17:49 |
| ashestakov | i think you should find problems in your cloudinit script | 17:50 |
| dmakogon | hub_cap, ashestakov that is why i'm standing on custom images without ruining cloud-init functionality in trive | 17:50 |
| dmakogon | *trove | 17:50 |
| hub_cap | ive run apt-get update just fine in heat cloud-init, im not sure saying "dont run cloud init" is the right answer though | 17:51 |
| dmakogon | there is no problems in script, i've tested it in a various ubuntu server images | 17:51 |
| hub_cap | so the script doesnt work in cloud init? | 17:51 |
| hub_cap | but it does work when u just run it? | 17:51 |
| ashestakov | dmakogon: lets resolve it tomorrow | 17:51 |
| dmakogon | it doen't work with nova repos | 17:52 |
| dmakogon | but it work with simple ubuntu repos | 17:52 |
| dmakogon | *works | 17:52 |
| dmakogon | so, in any way it took to long to install all deps and setup trove on instance | 17:53 |
| dmakogon | as you can see i'm setting up trove with virtualenv | 17:54 |
| dmakogon | just emulating functionality of dep packages | 17:54 |
| dmakogon | i'm gonna leave now, will be reachable in 1h or above | 17:55 |
| *** dmakogon is now known as away_dude | 17:55 | |
| *** jasonb365 has joined #openstack-trove | 17:56 | |
| juice | hub_cap: grapex: vipul: slicknik: I think it would be wise to have a few more "audit" fields in our instances table. mostly "created_by", "updated_by", "deleted_by". currently, only created_by would be used in notifications, however it just makes good sense to keep track of who made these significant changes to the instances. if you agree, hub_cap, can you create a dummy b/p so that I can flesh it out. | 18:05 |
| hub_cap | juice: ++ | 18:06 |
| hub_cap | and maybe ask vipul or SlickNik about the BPs... i think they said tis better to ask for forgiveness ;) | 18:06 |
| juice | cool | 18:06 |
| hub_cap | but ill totally creat it | 18:06 |
| juice | oh is that so - yeah I don't understand our employers position on b/p #suits | 18:07 |
| hub_cap | juice: https://blueprints.launchpad.net/trove/+spec/shimmy-shimmy-coco-pop | 18:07 |
| juice | however if we start doing this ourselves then we miss out on the entertaining names you give them :) | 18:08 |
| hub_cap | TRU | 18:08 |
| *** SushilKM__ has quit IRC | 18:08 | |
| *** rnirmal has quit IRC | 18:12 | |
| *** rnirmal has joined #openstack-trove | 18:12 | |
| juice | hub_cap: on second thought you can delete shimmy-shimmy, esp reminded me that non-core members cannot edit existing bp | 18:13 |
| juice | I'll just create one and take the slap on the wrist | 18:13 |
| hub_cap | juice: !!!!!! | 18:13 |
| juice | it's ok hub_cap, I can handle it | 18:14 |
| * juice flinches | 18:14 | |
| * hub_cap socks juice with a natural axe | 18:14 | |
| * hub_cap socks juice with a yellow dolphin | 18:14 | |
| * hub_cap boxes juice with a large figurine | 18:14 | |
| * hub_cap bangs juice with a blotched lamp | 18:14 | |
| * hub_cap spills some coffee over juice's keyboard | 18:14 | |
| hub_cap | there | 18:14 |
| * esp gives juice a cookie | 18:15 | |
| juice | oddly enough i can imagine all of those except a natural axe…aft? | 18:16 |
| juice | wtf? | 18:16 |
| esp | ^ cookie filled with pig intestines | 18:16 |
| hub_cap | awesome esp | 18:16 |
| esp | lol | 18:16 |
| hub_cap | ya im not sure what a natural axe is | 18:17 |
| juice | esp - is that what that was? | 18:18 |
| juice | delicious! | 18:18 |
| esp | juice: vietnames cookies | 18:18 |
| esmute | vipul, SlickNik: Can you guys review this when you can? https://review.openstack.org/#/c/45708/ | 18:19 |
| vipul | Sure esmute | 18:21 |
| hub_cap | esmute: i -1'd you | 18:26 |
| hub_cap | plz add yourself in the | 18:26 |
| hub_cap | Co-Authored-By: name <name@example.com> | 18:26 |
| hub_cap | since it will show up as denis being the commiter | 18:26 |
| hub_cap | no i lie | 18:26 |
| hub_cap | itll show up as you as the commiter esmute | 18:27 |
| hub_cap | and you should put denis as the co-author | 18:27 |
| esmute | urgg processes processes | 18:28 |
| datsun180b | so is that how we're going to solve the copyright problem | 18:28 |
| hub_cap | esmute: ;) | 18:28 |
| hub_cap | datsun180b: HAHA no | 18:28 |
| datsun180b | oh good, i was bracing for an inbox full of -1s for everything in trove | 18:28 |
| hub_cap | LOL datsun180b | 18:28 |
| hub_cap | this is only some long time (but relatively unused) git commit msgs standards for OS | 18:29 |
| hub_cap | since the topic of > 1 commiter came up recently i thought id make sure we did it properly | 18:29 |
| datsun180b | i see the application of it | 18:29 |
| datsun180b | makes sense | 18:29 |
| *** Barker has quit IRC | 18:32 | |
| openstackgerrit | Steve Leon proposed a change to openstack/trove: Security groups workflow update https://review.openstack.org/45708 | 18:34 |
| *** cweid has joined #openstack-trove | 18:36 | |
| esmute | hub_cap: Take a look now. The -1 from you propagated to the new review ^ | 18:38 |
| *** Barker has joined #openstack-trove | 18:38 | |
| hub_cap | esmute: thats till the same review | 18:39 |
| esmute | hmmm | 18:39 |
| esmute | ok give me a min | 18:40 |
| *** rnirmal has quit IRC | 18:41 | |
| *** rnirmal has joined #openstack-trove | 18:41 | |
| openstackgerrit | Steve Leon proposed a change to openstack/trove: Security groups workflow update https://review.openstack.org/45708 | 18:43 |
| hub_cap | no more -1 esmute :) | 18:44 |
| esmute | could you review it though? It seems that the +2 from grapex went away | 18:44 |
| esmute | Thanks hub_cap! | 18:45 |
| hub_cap | it did? boo | 18:45 |
| *** jasonb365_ has joined #openstack-trove | 18:46 | |
| *** jasonb365 has quit IRC | 18:47 | |
| *** jasonb365_ is now known as jasonb365 | 18:47 | |
| esmute | yeah :( | 18:48 |
| *** rnirmal_ has joined #openstack-trove | 19:02 | |
| *** rnirmal_ has quit IRC | 19:03 | |
| *** rnirmal_ has joined #openstack-trove | 19:03 | |
| *** rnirmal has quit IRC | 19:03 | |
| *** rnirmal_ is now known as rnirmal | 19:03 | |
| *** away_guy has joined #openstack-trove | 19:03 | |
| *** away_guy is now known as dmakogon | 19:07 | |
| dmakogon | sup | 19:07 |
| dmakogon | i'm baaaaaack | 19:08 |
| *** adrian_otto has joined #openstack-trove | 19:12 | |
| dmakogon | hub_cap, here ? | 19:13 |
| hub_cap | hey ya im here | 19:13 |
| dmakogon | nice | 19:13 |
| hub_cap | will be breaking for lunch soon | 19:14 |
| dmakogon | hub_cap, what do you think about repo for images ? | 19:14 |
| *** adrian_otto has quit IRC | 19:14 | |
| hub_cap | we shouldnt need one | 19:14 |
| hub_cap | we can have a repo for dib elements | 19:15 |
| hub_cap | but there is no point in having images | 19:15 |
| dmakogon | oh | 19:15 |
| dmakogon | yes | 19:15 |
| dmakogon | i mean | 19:15 |
| dmakogon | recipes | 19:15 |
| dmakogon | specs | 19:16 |
| dmakogon | hub_cap, have you heard about OZ ? | 19:16 |
| hub_cap | do we need more than dib elements? | 19:16 |
| hub_cap | whats OZ dmakogon? | 19:16 |
| dmakogon | oz image builder | 19:16 |
| dmakogon | pretty nice | 19:16 |
| dmakogon | as i heard it used by Rax | 19:17 |
| *** ashestakov has quit IRC | 19:17 | |
| dmakogon | pip install oz | 19:17 |
| *** demorris has joined #openstack-trove | 19:17 | |
| dmakogon | https://github.com/rackerjoe/oz-image-build | 19:17 |
| dmakogon | it is easy to use | 19:18 |
| dmakogon | easy to write custom recipes | 19:18 |
| hub_cap | dib | 19:18 |
| hub_cap | thats what we use | 19:18 |
| dmakogon | what dib includes ? | 19:18 |
| hub_cap | https://github.com/openstack/diskimage-builder | 19:20 |
| dmakogon | maybe wiki page would be greate | 19:20 |
| dmakogon | ah | 19:21 |
| dmakogon | that one | 19:21 |
| hub_cap | thats the official image builder for openstack | 19:21 |
| *** pdmars_ has joined #openstack-trove | 19:22 | |
| dmakogon | we need to change build script for trove needs | 19:22 |
| dmakogon | it would prepate simple image for fast trove deployment on it | 19:23 |
| *** pdmars has quit IRC | 19:25 | |
| *** ashestakov has joined #openstack-trove | 19:25 | |
| dmakogon | manual could be stored in trove/docs | 19:25 |
| hub_cap | we already have dib elements for ubuntu/fedora for mysql | 19:26 |
| hub_cap | i dont see why we couldnt just add the ones for cassandra too | 19:27 |
| hub_cap | we have them for redis | 19:27 |
| hub_cap | grapex: around? | 19:28 |
| dmakogon | hub_cap, but for development reasons i want to work with my own trove repo | 19:29 |
| dmakogon | as i know kick-start clones trove for github into image | 19:29 |
| dmakogon | right? | 19:29 |
| dmakogon | and what about real OS integration, not only devstack/redstack | 19:30 |
| *** krow has joined #openstack-trove | 19:30 | |
| dmakogon | so, for now this problems are only mine, so i should deal with them | 19:31 |
| dmakogon | if i would not successed i will bring this topic again | 19:31 |
| openstackgerrit | Robert Myers proposed a change to openstack/trove: Add tests for hosts mgmt commands https://review.openstack.org/52713 | 19:32 |
| hub_cap | dmakogon: you should be able to use dib images anywhere | 19:33 |
| grapex | hub_cap: Greets | 19:33 |
| hub_cap | a image has nothing to do with devstack | 19:33 |
| hub_cap | grapex: ashestakov wanted to discuss the email topic | 19:33 |
| hub_cap | the type/version stuff | 19:33 |
| ashestakov | grapex: hi | 19:33 |
| grapex | ashestakov: Hello | 19:33 |
| hub_cap | so lets start w/ the nesting of datastore:{type, version} vs datastore_type/datstore_version right? | 19:34 |
| dmakogon | hub_cap, ok, i would, and i would add to review spec for building image for cassandra | 19:34 |
| hub_cap | dmakogon: ++ | 19:34 |
| hub_cap | i suggested this idea a LONG while ago to ashestakov and others, and i like the nesting | 19:34 |
| dmakogon | hub_cap, thanks | 19:34 |
| grapex | hub_cap: Nesting ftw! | 19:34 |
| hub_cap | ok to we are in agreement for that ashestakov | 19:36 |
| hub_cap | now the topic of default version vs active | 19:36 |
| hub_cap | can u elaborate grapex on how we need default version | 19:36 |
| grapex | hub_cap: I worry people may miss out if we have this conversation on here vs the ML | 19:36 |
| hub_cap | ill reply to the list | 19:37 |
| hub_cap | i just wanted to get some context | 19:37 |
| grapex | hub_cap: Sure. I think in general most users won't care which version they have and will just want whatever the service providers who sets up Trove suggests | 19:37 |
| grapex | Now, currently if a default is specified in the config it works, but IIRC (sorry if I get this wrong) thats only if there's only one service type in use. | 19:37 |
| grapex | The other thing is you can mark a version as "active" by setting the "active" boolean field of the datastore_version tables to True, and that makes it the default. | 19:38 |
| grapex | But that too doesn't work if multiple rows have "active" set to True, which is possible | 19:38 |
| grapex | So I suggested have "datastore_type" have a "default_version_id" field that could be set optionally if providers wished to provide users an optional version | 19:39 |
| hub_cap | i dont honestly forsee many groups using the active flag for > 1 fwiw | 19:39 |
| hub_cap | i think only amcrn wanted it | 19:39 |
| hub_cap | so the active version for the default datastore_type would be sufficient if u wanted to use the default datastore_type | 19:40 |
| grapex | hub_cap: True, but we should still make sure there's a way to achieve both goals. The fact there can be two active versions per type means its not a good way to establish a default. | 19:40 |
| hub_cap | i also dont forsee people using default datastore_type unless, cough cough, you have a old api | 19:40 |
| grapex | Or users that use an old API in production at two companies. :) | 19:40 |
| hub_cap | right | 19:41 |
| grapex | That I can see there's no harm in adding either feature though. | 19:41 |
| hub_cap | i can agree w/ that to a point :) | 19:41 |
| hub_cap | i dont want to make it less intuitive and have _more_ options than needed | 19:41 |
| hub_cap | the more we allow for super confusing flexibility the more we have to 1) doc, 2) have bugs, 3) have nonstandard uses for things in the public codebase | 19:42 |
| grapex | To me, not having the ability to define default versions with a row on the datastore_type table is very unintuitive. | 19:42 |
| hub_cap | well its really default_type for your use case right? | 19:43 |
| hub_cap | and it just so happens its possible to have > 1 version avail for said default_type | 19:43 |
| grapex | Ok | 19:43 |
| hub_cap | so to me i dont see the need for adding more to that other than how you configure it | 19:44 |
| hub_cap | i think itd be smart to have a way, maybe via ordering, to define what happens if you have a default type w/ > 1 active service | 19:44 |
| grapex | So you're saying put a "priority" or something like that on each version? | 19:45 |
| grapex | I don't know- I don't understand why adding a default_version_id to the type would be confusing to people. Maybe this is just a matter of taste. | 19:46 |
| ashestakov | different way - pick fist active version (not depends on priority or etc), if user dont take care of exactly version | 19:46 |
| grapex | ashestakov: How would the "first" active version be decided? | 19:47 |
| grapex | Sorting by version string? | 19:47 |
| ashestakov | grapex: just get first row of SELECT result | 19:47 |
| hub_cap | ya but that is not the best way to do it | 19:47 |
| grapex | ashestakov: Would there be an ORDER BY clause? | 19:47 |
| hub_cap | cuz different things can affect the select result | 19:48 |
| hub_cap | every time ive seen a resultset.get(0) i cringe | 19:48 |
| hub_cap | ok so i think that i see grapex's point | 19:48 |
| hub_cap | if this is just adding a default_version_id to the datastore_type table then it might be easiest | 19:48 |
| ashestakov | maybe + ordering, or random, if user really dont take care | 19:48 |
| hub_cap | but ashestakov it also means you have a few different ways to check for versions | 19:49 |
| hub_cap | let me gist | 19:49 |
| ashestakov | if default_version_id is inactive? | 19:50 |
| hub_cap | grapex: ashestakov https://gist.github.com/hub-cap/7089806 | 19:51 |
| hub_cap | i have a question about what takes precedence | 19:51 |
| *** krow has quit IRC | 19:51 | |
| grapex | ashestakov: I'd argue that if the default version isn't set, it should just fail if the user doesn't specify a version rather than look at the active version. | 19:52 |
| grapex | But I'm not that serious about it. As long as we test all cases we should be fine | 19:52 |
| grapex | It would just be a hassle if an operator intended to set a default, didn't, and thought things were working as they intended until they added a second active version and their users began to see errors | 19:53 |
| grapex | Seems like only honoring a default_version field would honor the principle of least surprise | 19:53 |
| ashestakov | i think users must know what they do and specify version | 19:54 |
| hub_cap | i dont like having to specify a version | 19:58 |
| hub_cap | chances are 99% of the time u wil have 1 active version | 19:58 |
| hub_cap | and its the version your os is installing | 19:58 |
| hub_cap | grapex: dont think of mysql 5.1 and 5.5 as versions of mysql datastore type | 19:59 |
| ashestakov | i mean if more the one active | 19:59 |
| hub_cap | they will have to be different types because u cant upgrade from 5.1 to 5.5 | 19:59 |
| ashestakov | hub_cap: we really cant upgrade 5.1 to 55? | 20:00 |
| hub_cap | there are tons of incompatible changes between 51 and 55 ashestakov | 20:00 |
| hub_cap | i dont want us to build that in by default | 20:00 |
| demorris | hub_cap: not true per 5.1 to 5.5 | 20:01 |
| ashestakov | hub_cap: i know, i didnt research it deep | 20:01 |
| demorris | it is more involved, but you could handle 5.1 to 5.5 migrations, and other DBMS's might make it easier | 20:01 |
| demorris | major versions upgrades should not be ruled out | 20:02 |
| hub_cap | demorris: they are up to the deployers | 20:02 |
| grapex | hub_cap: Interesting. So then the nickname would be something like "mysql5.1" if that idea even gets consensus | 20:02 |
| hub_cap | if you want to see rackspace do the work to upgrade it, then id advise you to have them build it into the upgrade process | 20:02 |
| grapex | Sorry guys I've got a meeting to go to | 20:03 |
| demorris | hub_cap: correct, it would be up to the person running it on if it was supported or not | 20:03 |
| hub_cap | have fun w/ it grapex!! demorris gives you fun | 20:03 |
| grapex | I'll try to keep up with this on the ML | 20:03 |
| demorris | hub_cap: yes, MySQL is full of land mines for point release upgrades | 20:04 |
| redthrux | ML pls | 20:04 |
| *** adrian_otto has joined #openstack-trove | 20:04 | |
| *** mmcdaris has joined #openstack-trove | 20:04 | |
| hub_cap | redthrux: im only discussing the feature of the api and how to define it, not the actual process | 20:05 |
| hub_cap | ill let demorris work that out w/ you redthrux and grapex ;) | 20:05 |
| *** tanisdl has quit IRC | 20:06 | |
| ashestakov | so, versions... | 20:06 |
| grapex | Ok, it turns out I don't have to go to that meeting. | 20:07 |
| *** demorris_ has joined #openstack-trove | 20:07 | |
| grapex | vipul SlickNik: https://review.openstack.org/#/c/52665/1 <-- This fixes an issue we've observed with pagination- turns out it wasn't as flexible as it should've been. | 20:08 |
| hub_cap | ok im replying to the ML ashestakov grapex | 20:08 |
| *** demorris has quit IRC | 20:08 | |
| *** demorris_ is now known as demorris | 20:08 | |
| SlickNik | looking, grapex | 20:09 |
| ashestakov | hub_cap: grapex can we finish decision today? | 20:10 |
| *** vipul is now known as vipul-away | 20:10 | |
| *** vipul-away is now known as vipul | 20:10 | |
| grapex | ashestakov: Depends on if everyone is happy with the result. :) This is a huge addition to the API that will really help a number of things along, but we need to make sure we get it as close to what we want right off the bat or it'll be frustrating to change later. | 20:11 |
| hub_cap | sent grapex | 20:13 |
| hub_cap | yes grapex this is a linchpin of the API | 20:13 |
| hub_cap | we need to get it right the first time | 20:13 |
| kevinconway | are we still talking about versioned images? | 20:15 |
| grapex | kevinconway: Yes. It's the new users. :) | 20:16 |
| *** Ikhudoshyn__ has joined #openstack-trove | 20:20 | |
| Ikhudoshyn__ | hub_cap: ping | 20:21 |
| hub_cap | Ikhudoshyn__ ikhudoshyn: pong | 20:25 |
| ashestakov | grapex: can you gist #3 from you ML message? | 20:25 |
| *** vipul is now known as vipul-away | 20:25 | |
| *** tanisdl has joined #openstack-trove | 20:26 | |
| dmakogon | hub_cap, how could we tests our new datastores ? | 20:28 |
| *** Ikhudoshyn__ has quit IRC | 20:28 | |
| dmakogon | hub_cap, i mean mongo and cassandra ? | 20:28 |
| *** Ikhudoshyn__ has joined #openstack-trove | 20:29 | |
| dmakogon | hub_cap, how soon could our client could be updated ? | 20:29 |
| dmakogon | hub_cap, we are able to do only unit testing =/ | 20:30 |
| Ikhudoshyn__ | hub_cap: as for the mongo stuff, i didnt quite get what you with Tim have agreed on. | 20:30 |
| grapex | ashestakov: https://gist.github.com/anonymous/7090441 | 20:30 |
| hub_cap | dmakogon: client be update? | 20:30 |
| Ikhudoshyn__ | hi, Den.. | 20:30 |
| dmakogon | sup | 20:30 |
| dmakogon | hub_cap, i mean to support multi-datastores-types | 20:31 |
| Ikhudoshyn__ | the same question)) | 20:31 |
| hub_cap | Ikhudoshyn__: hes saying you need to use the existing test framework to test this until we get tempest testing working | 20:31 |
| hub_cap | dmakogon: same woudl go for you | 20:31 |
| ashestakov | grapex: like datastore.type.id, datastore.type.name ? | 20:31 |
| hub_cap | ashestakov: i think thats what he was talking about | 20:33 |
| ashestakov | grapex: and appripriate field should be filled depends on input type (uuid or string)? | 20:33 |
| hub_cap | uuid is string ;) | 20:33 |
| Ikhudoshyn__ | hub_cap: so we have to update -integration. but we cant have both in one review. what should come first? | 20:33 |
| *** jmontemayor has quit IRC | 20:33 | |
| ashestakov | "MySqL" is string too, will need to match regex then | 20:34 |
| hub_cap | Ikhudoshyn__: we will merge the code once we see your updates to -integration | 20:34 |
| hub_cap | ashestakov: no, just make sure its the name of the service_type or the id of the service_type | 20:34 |
| hub_cap | id think just pass datastore.type | 20:34 |
| *** Barker has quit IRC | 20:34 | |
| Ikhudoshyn__ | hub_cap: got it. that's what i looked for | 20:35 |
| hub_cap | and we can check if its a name or uuid by doing the select first on name of the avail datastores, then doing a select on the id if its not there | 20:35 |
| ashestakov | hub_cap: name should be limited to 36 chars too? | 20:36 |
| *** jmontemayor has joined #openstack-trove | 20:36 | |
| hub_cap | in the db u mean ashestakov? | 20:36 |
| ashestakov | hub_cap: in input | 20:37 |
| *** Barker has joined #openstack-trove | 20:37 | |
| hub_cap | ah u mean for json schema validation? | 20:37 |
| hub_cap | id make it the same as the field in the db, or leave it off the db field | 20:38 |
| hub_cap | wait | 20:38 |
| hub_cap | leave it off the validation field | 20:38 |
| hub_cap | cuz it will fail if it cant find a type | 20:38 |
| ashestakov | hub_cap: is it safe to pass not validated strings to sqlalchemy? | 20:40 |
| dmakogon | grapex, hub_cap, https://review.openstack.org/#/c/52905/ could you take at this ? | 20:43 |
| hub_cap | ashestakov: i dont think unexpected things will happen if you pass a 38 character string to 36 char limited str field in sqlalchemy | 20:44 |
| grapex | Ok, now there's that meeting. Talk to you all later! | 20:45 |
| *** adrian_otto has quit IRC | 20:45 | |
| *** Ikhudoshyn__ has quit IRC | 20:45 | |
| *** adrian_otto has joined #openstack-trove | 20:45 | |
| ashestakov | hub_cap: if we searching by name? | 20:46 |
| hub_cap | ashestakov: right | 20:46 |
| *** adrian_otto1 has joined #openstack-trove | 20:46 | |
| ashestakov | but doesent matter, can add validation later | 20:46 |
| hub_cap | ashestakov: yup | 20:46 |
| *** adrian_otto has quit IRC | 20:46 | |
| *** Ikhudoshyn__ has joined #openstack-trove | 20:46 | |
| hub_cap | mysqllllllllllllllllllllllllllllllllllllllllllllllllllllllllllll != mysql would be fine if you sent that to sqlalchemy (i would _assume_) | 20:47 |
| *** Ikhudoshyn__ has quit IRC | 20:47 | |
| *** Barker has quit IRC | 20:47 | |
| *** mmcdaris has quit IRC | 20:47 | |
| ashestakov | can we summarize all points? | 20:48 |
| hub_cap | ive summarized all of them i think in the email, pls reply with whatever else to summarize | 20:49 |
| *** mmcdaris has joined #openstack-trove | 20:49 | |
| *** Barker has joined #openstack-trove | 20:49 | |
| *** vipul-away is now known as vipul | 20:50 | |
| ashestakov | hub_cap: can you comment last message from grapex? | 20:52 |
| *** jrodom has joined #openstack-trove | 21:03 | |
| *** yidclare has quit IRC | 21:03 | |
| *** pdmars_ has quit IRC | 21:05 | |
| *** yidclare has joined #openstack-trove | 21:06 | |
| hub_cap | juice: 2013-10-21 20:32:04.234 | AssertionError: 2013-10-21T20:31:59Z != 2013-10-21T20:31:58Z | 21:08 |
| hub_cap | did u not get around to fixing it? | 21:08 |
| hub_cap | robertmyers: hows the horizon re-integration coming? | 21:09 |
| *** adrian_otto1 has quit IRC | 21:11 | |
| ashestakov | hub_cap: grapex, about point #2, we can use type from version with more precedence then default type, (only if type not specified in request) | 21:12 |
| hub_cap | i woudl assume the opposite, that the default_type is the higest precedence | 21:14 |
| ashestakov | hub_cap: now it implemented this way | 21:15 |
| *** adrian_otto has joined #openstack-trove | 21:15 | |
| ashestakov | hub_cap: and if default type != type from version = error | 21:15 |
| hub_cap | oh i read things wrong ashestakov | 21:15 |
| hub_cap | i think if you specify a typeID and not a versionID u should be able to get the versionID right? | 21:15 |
| hub_cap | and if you specifiy a versionID u can get the typeID for that version | 21:16 |
| *** jcru has quit IRC | 21:16 | |
| ashestakov | hub_cap: if specified type_id and not version - pick active, if multiple active - error or default | 21:17 |
| ashestakov | if specified version without type - get type from version (now is opposite - using default type) | 21:17 |
| hub_cap | ashestakov: yes to the 2nd thing | 21:19 |
| hub_cap | the first thing i think about it a bit different | 21:20 |
| *** yidclare has quit IRC | 21:20 | |
| hub_cap | if specified type and not version - look for default, if no default look for active, if > 1 active error | 21:20 |
| ashestakov | hm, what if default version is inactive? | 21:21 |
| hub_cap | hmm, error i think | 21:22 |
| robertmyers | hub_cap: the review is done and waiting for the openstack gate to change requirements | 21:22 |
| hub_cap | but grapex needs to know this is a issue w/ his approach :) | 21:23 |
| hub_cap | robertmyers: hooooray, can u link it? | 21:23 |
| robertmyers | https://review.openstack.org/#/c/51234/ | 21:23 |
| hub_cap | was it easier w/ the new troveclient robertmyers? | 21:23 |
| robertmyers | a little bit | 21:23 |
| ashestakov | hub_cap: it can be complex with default, more cases for error | 21:23 |
| hub_cap | yes ashestakov | 21:24 |
| hub_cap | jenkins doesnt like the version requirement eh robertmyers | 21:24 |
| robertmyers | yeah the openstack rejects its | 21:24 |
| robertmyers | but it works | 21:25 |
| hub_cap | do u have a requirments request as well? | 21:25 |
| robertmyers | no | 21:25 |
| hub_cap | well uh oh | 21:25 |
| hub_cap | id think we should push this first | 21:25 |
| hub_cap | https://review.openstack.org/#/c/52598/ | 21:25 |
| hub_cap | and then we can upgrade the requirements file | 21:26 |
| robertmyers | hub_cap: that one still references <1 | 21:27 |
| robertmyers | ? | 21:27 |
| hub_cap | sec robertmyers | 21:29 |
| *** cweid has quit IRC | 21:38 | |
| *** vipul is now known as vipul-away | 21:39 | |
| juice | has anyone extended trove.db.sqlalchemy in the sense of adding additional models? | 21:39 |
| hub_cap | robertmyers: https://review.openstack.org/#/c/52137/ | 21:39 |
| juice | it appears as though it is currently wired to be hard coded as to which models are included and since it is a module function, no clear way to "extend" the behavior | 21:40 |
| *** vipul-away is now known as vipul | 21:40 | |
| juice | should I ask this question on the mail list? | 21:41 |
| juice | btw which conversations should be handled on the list vs. irc? | 21:41 |
| hub_cap | there really isint a standard | 21:42 |
| juice | seems like design decisions should be handled on the list | 21:42 |
| juice | and conversation about beer in the irc but the rest seems like the gray in-between :) | 21:43 |
| *** adrian_otto has quit IRC | 21:43 | |
| *** ashestakov has quit IRC | 21:45 | |
| hub_cap | i dont agree | 21:45 |
| hub_cap | i want beer convos on list | 21:46 |
| hub_cap | period | 21:46 |
| juice | damn just when I thought I had the answer | 21:47 |
| juice | so it looks like I can call sqlalchemy.orm.mapper at any point so I think that nullifies my question | 21:47 |
| *** adrian_otto has joined #openstack-trove | 21:49 | |
| *** dmakogon has quit IRC | 21:50 | |
| openstackgerrit | A change was merged to openstack/trove: Security groups workflow update https://review.openstack.org/45708 | 21:52 |
| openstackgerrit | Ed Cranford proposed a change to openstack/trove: Splits paging config to max and default https://review.openstack.org/52665 | 21:53 |
| *** yidclare has joined #openstack-trove | 21:53 | |
| *** adrian_otto has quit IRC | 22:00 | |
| *** datsun180b has quit IRC | 22:02 | |
| *** grapex has quit IRC | 22:02 | |
| *** yogeshmehra has quit IRC | 22:02 | |
| *** grapex has joined #openstack-trove | 22:02 | |
| *** grapex has quit IRC | 22:07 | |
| *** robertmyers has quit IRC | 22:12 | |
| *** Barker has quit IRC | 22:14 | |
| *** demorris has quit IRC | 22:18 | |
| *** jasonb365 has quit IRC | 22:36 | |
| *** tanisdl has quit IRC | 22:41 | |
| hub_cap | has everyone proposed things for the summit!?!? | 22:43 |
| hub_cap | ive approved 3 of them so far, we have 2 left to choose slots for | 22:43 |
| vipul | now i need to figure out what to talk about :p | 22:48 |
| openstackgerrit | Kui Shi proposed a change to openstack/python-troveclient: Sync up exceptions.py from oslo https://review.openstack.org/51953 | 22:52 |
| *** tanisdl has joined #openstack-trove | 22:54 | |
| hub_cap | vipul: u have a session that ive scheduled | 22:55 |
| hub_cap | no downtime upgrades | 22:55 |
| hub_cap | ideally vipul id like to get all 4 core members to run their own | 22:58 |
| hub_cap | so we spread the work out a bit so to speak | 22:58 |
| hub_cap | vipul: https://review.openstack.org/#/c/51953/ <-- no longer mucks w/ openstack common classes | 23:00 |
| *** rnirmal has quit IRC | 23:01 | |
| *** jodom has joined #openstack-trove | 23:03 | |
| *** jodom has quit IRC | 23:03 | |
| *** jmontemayor has quit IRC | 23:04 | |
| *** vipul is now known as vipul-away | 23:06 | |
| *** jrodom has quit IRC | 23:07 | |
| *** adrian_otto has joined #openstack-trove | 23:09 | |
| *** vipul-away is now known as vipul | 23:20 | |
| vipul | hub_cap: woohoo | 23:20 |
| hub_cap | ya vipul is good | 23:21 |
| vipul | are you doing one on cluster api? | 23:21 |
| hub_cap | well i was thinking if we only had 1 more slot do we want to talk about the v2 api or clustering? | 23:22 |
| hub_cap | id assume clustering honestly... | 23:22 |
| hub_cap | but i wantd some input on it | 23:22 |
| *** paul_lodronio has left #openstack-trove | 23:22 | |
| vipul | we really need to get that finalized | 23:22 |
| openstackgerrit | A change was merged to openstack/python-troveclient: Sync up exceptions.py from oslo https://review.openstack.org/51953 | 23:22 |
| vipul | i know some folks eager to get started on it | 23:22 |
| hub_cap | yes its pretty final | 23:22 |
| vipul | the latest version imsplitbit proposed? | 23:22 |
| hub_cap | id say we are 90+% done... lets chat w/ imsplitbit tomorrow on whats left | 23:22 |
| vipul | ok -- if it's a done deal.. then v2 api makes sense | 23:23 |
| hub_cap | yup | 23:23 |
| hub_cap | just gotta see what the state of the cluster union is tomorrow | 23:24 |
| hub_cap | and make sure that grapex is bought in on 1 session that he can run | 23:24 |
| vipul | https://wiki.openstack.org/wiki/Trove-Replication-And-Clustering-API <-- was this the final one? | 23:24 |
| * hub_cap shrugs | 23:25 | |
| vipul | imsplitbit: ^^ please update if wrong | 23:25 |
| hub_cap | imsplitbit: wont see this | 23:26 |
| hub_cap | he has a bouncer and never checks it | 23:26 |
| hub_cap | imsplitbit: is worthless wrt irc | 23:26 |
| hub_cap | ;) | 23:27 |
| hub_cap | i can say that cuz he wont see it | 23:27 |
| hub_cap | which will make it true | 23:27 |
| hub_cap | mwahaha | 23:27 |
| *** yogeshmehra has joined #openstack-trove | 23:35 | |
| *** jrodom has joined #openstack-trove | 23:42 | |
| openstackgerrit | Dan Nguyen proposed a change to openstack/python-troveclient: Add --bypass-url option to trove client https://review.openstack.org/52461 | 23:49 |
| *** vipul is now known as vipul-away | 23:52 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!