| *** kagan has quit IRC | 00:18 | |
| *** sacharya has joined #openstack-meeting-alt | 00:20 | |
| *** vipul is now known as vipul|away | 00:28 | |
| *** rmohan has quit IRC | 00:37 | |
| *** vipul|away is now known as vipul | 00:40 | |
| *** rmohan has joined #openstack-meeting-alt | 00:41 | |
| *** bdpayne has quit IRC | 00:41 | |
| *** esp has joined #openstack-meeting-alt | 01:23 | |
| *** rmohan has quit IRC | 01:33 | |
| *** rmohan has joined #openstack-meeting-alt | 01:34 | |
| *** demorris has joined #openstack-meeting-alt | 01:59 | |
| *** rmohan has quit IRC | 02:07 | |
| *** rmohan has joined #openstack-meeting-alt | 02:09 | |
| *** demorris has quit IRC | 02:58 | |
| *** vipul is now known as vipul|away | 03:24 | |
| *** bdpayne has joined #openstack-meeting-alt | 03:31 | |
| *** jcru has joined #openstack-meeting-alt | 04:01 | |
| *** vipul|away is now known as vipul | 04:20 | |
| *** sacharya has quit IRC | 04:45 | |
| *** bdpayne has quit IRC | 05:00 | |
| *** rmohan has quit IRC | 11:43 | |
| *** dhellmann-afk has quit IRC | 11:43 | |
| *** esmute has quit IRC | 11:43 | |
| *** juice has quit IRC | 11:43 | |
| *** annashen has quit IRC | 11:43 | |
| *** clarkb has quit IRC | 11:43 | |
| *** jcooley has quit IRC | 11:43 | |
| *** chmouel has quit IRC | 11:43 | |
| *** zykes- has quit IRC | 11:43 | |
| *** westmaas has quit IRC | 11:43 | |
| *** mordred has quit IRC | 11:43 | |
| *** ChanServ has quit IRC | 11:43 | |
| *** SpamapS has quit IRC | 11:43 | |
| *** cp16net|away has quit IRC | 11:43 | |
| *** lifeless has quit IRC | 11:43 | |
| *** vipul has quit IRC | 11:43 | |
| *** redthrux has quit IRC | 11:43 | |
| *** openstack has joined #openstack-meeting-alt | 15:21 | |
| *** ChanServ sets mode: +o openstack | 15:21 | |
| *** bdpayne has joined #openstack-meeting-alt | 15:27 | |
| *** djohnstone has quit IRC | 15:27 | |
| *** demorris has quit IRC | 15:35 | |
| *** djohnstone has joined #openstack-meeting-alt | 15:35 | |
| *** jcru has quit IRC | 15:53 | |
| *** jcru has joined #openstack-meeting-alt | 15:54 | |
| *** demorris has joined #openstack-meeting-alt | 15:54 | |
| *** imsplitbit has joined #openstack-meeting-alt | 16:03 | |
| *** rmohan has quit IRC | 16:23 | |
| *** rmohan has joined #openstack-meeting-alt | 16:24 | |
| *** esp1 has joined #openstack-meeting-alt | 16:58 | |
| *** esp1 has left #openstack-meeting-alt | 16:58 | |
| *** vipul is now known as vipul|away | 17:00 | |
| *** rnirmal_ has joined #openstack-meeting-alt | 17:02 | |
| *** rnirmal_ has quit IRC | 17:02 | |
| *** rnirmal_ has joined #openstack-meeting-alt | 17:03 | |
| *** rnirmal has quit IRC | 17:04 | |
| *** rnirmal_ is now known as rnirmal | 17:05 | |
| *** kagan has joined #openstack-meeting-alt | 17:18 | |
| *** dhellmann-afk has quit IRC | 17:18 | |
| *** jcru is now known as jcru|away | 17:25 | |
| *** jcru|away is now known as jcru | 17:30 | |
| *** kagan has quit IRC | 17:32 | |
| *** vipul|away is now known as vipul | 17:36 | |
| *** jcru is now known as jcru|away | 17:47 | |
| *** jcru|away is now known as jcru | 17:50 | |
| *** kagan has joined #openstack-meeting-alt | 17:58 | |
| *** esp1 has joined #openstack-meeting-alt | 18:05 | |
| *** esp1 has left #openstack-meeting-alt | 18:05 | |
| *** rmohan has quit IRC | 18:38 | |
| *** rmohan has joined #openstack-meeting-alt | 18:39 | |
| *** rnirmal has quit IRC | 18:57 | |
| *** demorris has quit IRC | 18:59 | |
| *** vipul is now known as vipul|away | 19:20 | |
| *** jcru is now known as jcru|away | 19:26 | |
| *** cp16net is now known as cp16net|away | 19:29 | |
| *** cp16net|away is now known as cp16net | 19:30 | |
| *** jcru|away is now known as jcru | 19:47 | |
| *** vipul|away is now known as vipul | 19:50 | |
| *** jcru is now known as jcru|away | 20:24 | |
| *** vipul is now known as vipul|away | 20:25 | |
| *** vipul|away is now known as vipul | 20:33 | |
| *** vipul is now known as vipul|away | 20:34 | |
| *** jcru|away is now known as jcru | 20:42 | |
| *** vipul|away is now known as vipul | 20:50 | |
| *** esp has joined #openstack-meeting-alt | 20:53 | |
| *** demorris has joined #openstack-meeting-alt | 20:55 | |
| *** datsun180b has joined #openstack-meeting-alt | 20:56 | |
| *** grapex has joined #openstack-meeting-alt | 20:56 | |
| *** SlickNik has joined #openstack-meeting-alt | 20:58 | |
| *** robertmyers has joined #openstack-meeting-alt | 20:58 | |
| *** saurabhs has joined #openstack-meeting-alt | 20:59 | |
| vipul | almost meeting time | 21:00 |
|---|---|---|
| SlickNik | eyah | 21:00 |
| SlickNik | yeah* | 21:00 |
| vipul | got folks here? | 21:00 |
| datsun180b | I speak only for myself | 21:00 |
| *** kmansel has joined #openstack-meeting-alt | 21:00 | |
| grapex | Yo | 21:00 |
| djohnstone | hiya | 21:01 |
| vipul | #startmeeting Reddwarf | 21:01 |
| openstack | Meeting started Tue Mar 12 21:01:16 2013 UTC. The chair is vipul. Information about MeetBot at http://wiki.debian.org/MeetBot. | 21:01 |
| openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 21:01 |
| *** openstack changes topic to " (Meeting topic: Reddwarf)" | 21:01 | |
| juice | present | 21:01 |
| openstack | The meeting name has been set to 'reddwarf' | 21:01 |
| vipul | agenda: | 21:01 |
| vipul | #link https://wiki.openstack.org/wiki/Meetings/RedDwarfMeeting | 21:01 |
| cp16net | \o | 21:01 |
| SlickNik | Here. | 21:01 |
| vipul | feel free to update as necessary | 21:01 |
| esmute | hello | 21:01 |
| vipul | #topic action items | 21:02 |
| *** openstack changes topic to "action items (Meeting topic: Reddwarf)" | 21:02 | |
| SlickNik | http://eavesdrop.openstack.org/meetings/reddwarf/2013/reddwarf.2013-03-05-21.59.html | 21:02 |
| vipul | ok first one.. didn't update the snapshots wiki yet | 21:02 |
| vipul | #action vipul ot update Snapshots wiki to handle deleted in swift use case | 21:02 |
| vipul | i swear i was going to do it today | 21:02 |
| vipul | other stuff came up | 21:02 |
| vipul | oh well | 21:02 |
| cp16net | ok sounds good | 21:03 |
| vipul | next.. SlickNik | 21:03 |
| SlickNik | Second one, I'm still working on Security Groups. | 21:03 |
| grapex | vipul: btw- robertmyers, imsplitbit and myself have a meeting tomorrow to look over that wiki article. | 21:03 |
| SlickNik | Cleaning up, extending the client — writing int tests. | 21:03 |
| vipul | grapex: awesome.. we've started some initial impl so appreciate any feedback | 21:03 |
| imsplitbit | grapex, I was talking about it with him | 21:03 |
| SlickNik | I should have a patch out for review by end of tomorrow. | 21:04 |
| grapex | vipul: Cool - I noticed there was a commit for the schema. That may be premature, it's hard for me to review schema changes w/o the associated code. | 21:04 |
| SlickNik | I'd appreciate it if I can get multiple eyeballs on it after that. :) | 21:04 |
| cp16net | #agree with grapex | 21:04 |
| datsun180b | #agree | 21:04 |
| vipul | SlickNik: cool | 21:04 |
| *** jdbarry1 has joined #openstack-meeting-alt | 21:04 | |
| grapex | Although if several people need ot sync up on work haivng it merged would make it easier. I think the typical approach is to create a branch that everyone works from, then one person does a gerrit review for the whole thing on completion | 21:04 |
| SlickNik | #action SlickNik still working on Security-Groups. Hope to have patch up for review by Wed/Thu. | 21:04 |
| vipul | grapex: regarding schema change.. i think we should strive for smaller patches.. makes things easier to reivew | 21:05 |
| vipul | but i see the point about not having enough context | 21:05 |
| vipul | so maybe we need to find a good balance.. so maybe schema change + models code? | 21:05 |
| vipul | and then API later? | 21:05 |
| grapex | vipul: Yes, if the models code had some kind of unit tests on it | 21:05 |
| grapex | Enough just to show how it's used | 21:05 |
| vipul | #agreed | 21:06 |
| SlickNik | Yeah, I like that. Schema + models has a bit more meat to it. | 21:06 |
| cp16net | sorry if i missed it was there docs around the schema changes? | 21:06 |
| imsplitbit | vipul I think the models code would def help | 21:06 |
| cp16net | so that we could see the path forawrd? | 21:06 |
| vipul | cp16net: https://wiki.openstack.org/wiki/Snapshot-design | 21:06 |
| vipul | :p | 21:06 |
| vipul | yea we tried to get that captured in the wiki design doc | 21:06 |
| cp16net | ok cool it looks like there is more there than when i last looked at it | 21:07 |
| cp16net | thanks | 21:07 |
| vipul | np | 21:07 |
| vipul | ok we'll get another review that's got more meat | 21:07 |
| grapex | So the other thing regarding schema changes for a feature like snapshot- if throughout testing we discover there have been any mistakes, we'll we need to add additional migration files to tweak the schemas? Or is the plan to just change the one migration file created for that feature? | 21:07 |
| vipul | grapex: that's a good point.. | 21:08 |
| vipul | grapex: i would be OK with changing the same migration.. | 21:08 |
| cp16net | yeah i would rather it be a single change set at that point | 21:08 |
| vipul | the only thing is we shoudl coordinate to make sure it's not deployed | 21:08 |
| juice | I think it would be best to just wrap it on one | 21:08 |
| juice | or update the original | 21:08 |
| grapex | vipul: Me too, but internally at Rack I'd need to make sure I didn't put that code in the release until the feature was under wraps to keep ops from any headaches | 21:08 |
| vipul | maybe that can be communicated during the review.. | 21:08 |
| imsplitbit | one works as long as it isn't already deployed | 21:09 |
| vipul | grapex: yea, shoudl be fine until we get other people deploying it | 21:09 |
| grapex | Well, headaches is an exageration, it wouldn't be that bad really, just wouldn't want to run the migration before the feature is used. | 21:09 |
| cp16net | i know adding that once its deployed you cant change those migration scripts | 21:09 |
| imsplitbit | cause ti will be a pain to fix... yeah what grapex said | 21:09 |
| * cp16net been there.... | 21:09 | |
| * imsplitbit done that | 21:09 | |
| vipul | yea, so maybe the rule of thumb is change existing, and if anyone has concerns that can be xpressed in review | 21:09 |
| vipul | at which point we can split it out to a new migration | 21:09 |
| SlickNik | grapex: I'd don't think you guys would want a half feature (i.e. only models code) in release anyhow. But yes, that's something to be wary about. | 21:09 |
| grapex | vipul: Sounds good. | 21:10 |
| vipul | grapex: another question regarding the shcema | 21:10 |
| vipul | do you guys do constraints on your migrations? | 21:10 |
| vipul | or something that you handle out of sqlalchemy or not do at all | 21:10 |
| vipul | i think most of openstack there really aren't constraints expressed in the migrate scripts | 21:11 |
| grapex | vipul: We actually just use the sqlalchemy stuff directly and 80% have no issues | 21:11 |
| jcooley | constraints are a pain because existing data can set them off. | 21:11 |
| grapex | 20% we talk to Ops and do it by hand I think. | 21:11 |
| grapex | Oh, I see. | 21:11 |
| vipul | jcooley: agree, so maybe we don't do constraints in reddwarf either | 21:12 |
| grapex | I think once I tried to add non-null constraints and ran into problems. | 21:12 |
| vipul | ok action | 21:12 |
| grapex | Seems like a good idea honestly if we could get them to work. | 21:13 |
| jcooley | yep, worse is there are NULL values and someone applies a non-NULL constraint. then you crash. | 21:13 |
| cp16net | i dont remember either | 21:13 |
| jcooley | happens frequently when you add columns to a table. :) | 21:13 |
| vipul | for simple things like unique constraints, we might want to add those... | 21:13 |
| vipul | just no fkey, etc | 21:13 |
| SlickNik | I think FK constraints are goodness too for things that make sense (like ids) | 21:14 |
| vipul | as long as those constraints as defined when the table is first created not added later | 21:14 |
| esmute | i did add unique constraint in the quota table | 21:14 |
| SlickNik | Yes, agreed | 21:14 |
| SlickNik | You'd run into most problems when trying to add constraints to pre-existing data. | 21:14 |
| vipul | yep.. | 21:15 |
| vipul | ok next action was mine as well | 21:15 |
| vipul | #action vipul to submit snapshots API to database-api docs | 21:15 |
| esmute | yes | 21:15 |
| vipul | did not complete | 21:15 |
| vipul | next item: grapex | 21:15 |
| vipul | did you learn how to use action? | 21:15 |
| vipul | heh | 21:15 |
| grapex | No, I forgot. :( | 21:15 |
| SlickNik | that was hubcap's action, I think… :) | 21:15 |
| SlickNik | to "teach" grapex... | 21:16 |
| grapex | As for that decorator, I did some thinking and closed the blueprint. | 21:16 |
| SlickNik | slacker... | 21:16 |
| grapex | I was thinking, adding a decorator like that defeats the purpose of gating on tests. | 21:16 |
| vipul | decorator to only run tests against certain versions? | 21:16 |
| grapex | vipul: yes | 21:16 |
| vipul | i hope we solved the client issue | 21:16 |
| vipul | by making it pull directly from github | 21:17 |
| grapex | It's frustrating, but if we disable tests on brand new features we could check in tests that fail to work and not realize it until we tick up the client version later. | 21:17 |
| esp | grapex: maybe the change we make to reddwarf's pip-requires will be enough to keep the client in sync | 21:17 |
| grapex | It seems a better solution is to keep them in Gerrit purgatory until everything is ready to go. | 21:17 |
| esp | I've been trying to check in client changes first.. | 21:17 |
| datsun180b | Right | 21:17 |
| grapex | esp vipul: I think the git change vipul merged will really help us. | 21:17 |
| vipul | esp: yea if we stick to that pattern, seems like we'll be ok | 21:17 |
| grapex | it sucks for the person submitting the feature but will ultimately benefit the wider team | 21:18 |
| datsun180b | #agree | 21:18 |
| vipul | ok last item: SlickNik | 21:18 |
| SlickNik | I'm not sure I'm a fan of the decorator myself, so as much as I don't like gerrit purgatory, I think that might be the better option here. | 21:18 |
| vipul | trimming redstack | 21:18 |
| SlickNik | I looked into refactoring redstack, and started on it... | 21:19 |
| grapex | SlickNik: Nothing more fun than refactoring bash eh? | 21:19 |
| SlickNik | Grapex, let's just say It's a little hairier than I thought :) | 21:19 |
| datsun180b | Better than refactoring awk | 21:19 |
| vipul | it woudl be good to know some of the 'other' functions in redstack are used by anyone | 21:19 |
| grapex | datsub180b: Or batch files. | 21:19 |
| vipul | like start-deps, etc | 21:20 |
| SlickNik | I'll probably have done after this sec-groups stuff (hopefully by end of this week) | 21:20 |
| grapex | vipul: We end up using some functions for internal things, mostly billing testing. | 21:20 |
| grapex | And personall I call stop and start-deps frequently when making changes to Nova. | 21:21 |
| SlickNik | The issue is that there is a lot of code that is "slightly" different… So I'm trying to find most of so I can de-dupe and reduce the footprint.. | 21:21 |
| vipul | grapex: we should trim if we can anywhere.. maybe we need them though | 21:21 |
| vipul | SlickNik: do you have a bug/bp for that? | 21:21 |
| SlickNik | Yes, I have a bug…one sec... | 21:21 |
| vipul | maybe we can get grapex to comment on that bug with stuff they use or don't | 21:21 |
| SlickNik | #link https://bugs.launchpad.net/reddwarf-integration/+bug/1154170 | 21:21 |
| SlickNik | #action SlickNik working on trimming down redstack to use local.sh https://bugs.launchpad.net/reddwarf-integration/+bug/1154170 | 21:22 |
| vipul | awesome | 21:22 |
| vipul | thanks! | 21:22 |
| vipul | ok that's it for action items... | 21:23 |
| SlickNik | np. | 21:23 |
| vipul | #topic limits updates | 21:23 |
| *** openstack changes topic to "limits updates (Meeting topic: Reddwarf)" | 21:23 | |
| vipul | most of the limits stuff is merged | 21:23 |
| esp | so I'm in the middle of some fixin' | 21:23 |
| esp | should be done between today and tomorrow. | 21:23 |
| vipul | sweet | 21:24 |
| esp | I had to refactor view code in the API and Client to get it working with XML | 21:24 |
| grapex | esp: Nice. | 21:24 |
| vipul | good timing on the xml tests grapex | 21:24 |
| esp | yeah should be less crusty.. | 21:24 |
| vipul | probably wouldn't have been caught til much later | 21:24 |
| vipul | esp: lol | 21:24 |
| grapex | vipul: Thanks Vipul, we've had some issues lately at Rack so I figured I'd flip them on. | 21:25 |
| esp | xml is soo 90's | 21:25 |
| vipul | do other openstack projects support xml? | 21:25 |
| grapex | lol, yeah, there are many "Enterprise Professionals" in suits who will be happy ours works. :) | 21:25 |
| vipul | just curious, not sure what the stance is there | 21:25 |
| grapex | vipul: In theory. ;) | 21:25 |
| grapex | They all do, I'm just kidding | 21:25 |
| grapex | remember that talk we went to about improving support though? | 21:25 |
| grapex | Apparently there are some issues | 21:26 |
| grapex | btw | 21:26 |
| grapex | I should probably talk about this a bit | 21:26 |
| vipul | yea it seems like everyone wants to drop support | 21:26 |
| cp16net | oh yeah... haha | 21:26 |
| vipul | but obviously that's a huge change so they can't | 21:26 |
| cp16net | xml support in nova no one could agree it sounded like | 21:26 |
| cp16net | yeah | 21:26 |
| grapex | Well the other projects have XML and JSON views that are very different | 21:26 |
| imsplitbit | esp totally, I feel retro when I use it | 21:26 |
| grapex | where as most of our views "just work" in XML with a small bit of serialization configuration | 21:26 |
| grapex | I feel like we've lucked out | 21:27 |
| esp | imsplit: yep. listening to Toto as I code it. | 21:27 |
| esp | oh wait, that's more 80's | 21:27 |
| imsplitbit | yeah | 21:27 |
| vipul | grapex: Yea, it's nice not having to maintain two separate views | 21:27 |
| imsplitbit | but funny nonetheless | 21:27 |
| imsplitbit | I'm all for use JSON or go home | 21:27 |
| grapex | by the way, I'm testing this out internally before I subject you guys to it, but I think it may be possible to use a tool like XmlLint or Json schema in conjunction with the client. I have code that I'm running now that extends the client to save the request / response text into files | 21:27 |
| *** yidclare has joined #openstack-meeting-alt | 21:27 | |
| vipul | but i think esp found some assumptions that were made.. like the depth of the tree | 21:28 |
| grapex | my next goal is to run xmllint and json schema against it to make sure the old XSD is accurate... whether it is is a bit of a mystery now. :) | 21:28 |
| esp | grapex: probably a good idea. | 21:28 |
| grapex | vipul: True. For something like limits that was taken from Nova, it may require work to make the views correct in both. | 21:29 |
| vipul | grapex: so would that be permanent in rdclient? or just a one time thing to test accuracy against xsd | 21:29 |
| *** imsplitbit has quit IRC | 21:29 | |
| grapex | if I get this working I'm not sure where we'd run it... I might just write up publicly how to do it so when we have real mode in CI on Jenkins somewhere we can kick it off. | 21:29 |
| esmute | but the 80's was such a great decade. ET, Nintendo, Thundercats and no Justin Beaver | 21:29 |
| vipul | LOL | 21:29 |
| grapex | It would use rdclient but the extra code is in the tests. Its just one class, pretty slim. | 21:29 |
| esp | esmute: true, there were some good moments in there. | 21:29 |
| esp | esmute: I think you might have had a better childhood than me. | 21:30 |
| vipul | grapex: Yea, it may be good to have it part of CI but may not be necessary to have validation in the client? | 21:30 |
| grapex | esmute: The really sad thing is that if they made a Thundercats movie today Jerry Bruckheimer would direct and Justin beaver would be the main star. | 21:30 |
| vipul | who's justin beaver | 21:30 |
| vipul | lol | 21:30 |
| grapex | vipul: Yeah, don't worry, I won't burden the client code / repo with that code. | 21:30 |
| esmute | oh no god no | 21:31 |
| vipul | evil twin of justin beiber? | 21:31 |
| grapex | vipul: Anyway, I'll keep you guys up to date on how it goes. | 21:31 |
| esp | lol | 21:31 |
| vipul | grapex: ok cool | 21:31 |
| SlickNik | You don't know Justin Beaver? Dam…. | 21:31 |
| esmute | haha. Where is the auto-spell-correcting bot here? | 21:32 |
| cp16net | grapex action? | 21:32 |
| grapex | esmute: Ha! I spelled it that way too. And yet, I feel no shame in getting it wrong... :) | 21:32 |
| vipul | good call cp16net | 21:32 |
| SlickNik | Grapex, that sounds good… | 21:32 |
| SlickNik | (both the xml and no shame parts) | 21:33 |
| vipul | oh sorry, hub_cap didn't 'teach' you about action | 21:33 |
| * grapex Look into validating the API with XmlLint and Json schema. | 21:33 | |
| vipul | #action grapex to look into xml validation | 21:33 |
| vipul | done | 21:33 |
| grapex | vipul: Thanks. :) | 21:33 |
| vipul | #topic Security Groups update | 21:34 |
| *** openstack changes topic to "Security Groups update (Meeting topic: Reddwarf)" | 21:34 | |
| SlickNik | Already covered this earlier. | 21:34 |
| vipul | k cool.. | 21:34 |
| SlickNik | Expect a patch up for review soon. | 21:34 |
| vipul | nice | 21:34 |
| SlickNik | (by Thursday more concretely) | 21:35 |
| vipul | #action Snapshots Update | 21:35 |
| vipul | whoops | 21:35 |
| vipul | #topic Snapshots Update | 21:35 |
| *** openstack changes topic to "Snapshots Update (Meeting topic: Reddwarf)" | 21:35 | |
| vipul | #link https://wiki.openstack.org/wiki/Snapshot-design | 21:35 |
| vipul | grapex, imsplitbit please review ^^ | 21:35 |
| vipul | appreciate feedback on that.. | 21:35 |
| juice | swift client is about wrapped up | 21:36 |
| juice | went with the good ol' fake | 21:36 |
| grapex | vipul: We're looking today, we'll have a meeting tomorrow over it. | 21:36 |
| juice | realizing that being new to python and making some big changes doesn't make good chemistry | 21:36 |
| vipul | To recap some of the discussion i had with imsplit bit, we're supporting point in time backups and recovery to 'that' point in time. Not supporting point-in-time recovery (restore to 5 mins ago) | 21:37 |
| vipul | recovery is to a new instance only | 21:37 |
| grapex | vipul: This is for snapshots? | 21:37 |
| vipul | yes | 21:37 |
| juice | so the user just can't pick the "snapshot" they want and say restore? | 21:37 |
| vipul | juice: exactly | 21:37 |
| vipul | juice: they get exactly what the snapshot contained | 21:38 |
| grapex | Ok, cool | 21:38 |
| cp16net | in a new instance | 21:38 |
| grapex | So btw | 21:38 |
| grapex | in the end, this is going to be a call to the guest that will upload a file to swift, right? | 21:38 |
| juice | how much storage do they get in swift and what happens when a snapshot will exceed those contents? | 21:38 |
| juice | grapex: yes | 21:38 |
| SlickNik | yes grapex, that's the idea... | 21:38 |
| vipul | grapex: Yea, so we'll zip+upload | 21:38 |
| vipul | juice: we'll need to look into chunking | 21:39 |
| vipul | juice: i believe swiftclient already does this for you | 21:39 |
| juice | chunking the post | 21:39 |
| vipul | if > 5 gigs | 21:39 |
| juice | for performance? | 21:39 |
| juice | ah | 21:39 |
| cp16net | juice: thats a good point because you may have to chop up the zip | 21:39 |
| vipul | which i think is the max file size | 21:39 |
| cp16net | object file size | 21:39 |
| grapex | if you guys add a fake method to the guest in fake mode, you should be able to write even the integration tests soon. In fake mode I think all the API will do is say "sure!" when you ask it to upload a a snapshot | 21:39 |
| vipul | yea think that's the max limit in swift? could be wrong | 21:39 |
| juice | yes 5gb is the max size (which isn't the same for folks like was) | 21:39 |
| grapex | vipul: sorry if this is in the wiki, but is the idea to use the action stuff hub_cap was looking at? | 21:40 |
| juice | at least it is stated as such and I have tried on occasion to exceed it when storing season one of walking dead | 21:40 |
| vipul | grapex: No, currently we don't have the framework for that in place | 21:40 |
| vipul | grapex: we could come back aorund and add the hooks into what hub_cap submit? | 21:41 |
| SlickNik | I wonder if we can do something cool with incremental block delta storage. (Might have to come in later). | 21:41 |
| grapex | vipul: I think so | 21:41 |
| grapex | The action stuff he showed me, which is in Nova now, looks like it might solve a lot of the issues we've had with asynchronous actions so far | 21:41 |
| cp16net | oh yeah that would be awesome | 21:42 |
| grapex | vipul: Too bad he's been moving, as soon as he get's back I'll let him know to sync up with you guys. | 21:42 |
| vipul | grapex: Ok, yea i need to spend more time looking into the nova feature, i knnow he tried to explain it last meeting i think | 21:42 |
| cp16net | #action hub_cap to sync up with vipul on the nova (instance_actions/task_actions) | 21:43 |
| vipul | grapex: I think there is also some open questions around myISAM.. which we will support.. but it's just a binary copy of the files (wht xtrabackup) | 21:43 |
| vipul | also means your db is locked | 21:44 |
| vipul | during that copy | 21:44 |
| vipul | which is something we don't have to worry about when you do xtrabckup against innodb | 21:44 |
| vipul | just stuff to think about when you guys review the spec | 21:44 |
| *** yidclare has quit IRC | 21:45 | |
| grapex | vipul: Ok, we'll look into it and get back to you sometime tomorrow or Thursday. | 21:45 |
| vipul | ok cool | 21:45 |
| grapex | I'll print out that wiki article and read it with some fine cognac and a slice of cheese. | 21:45 |
| vipul | #action annashen to amend patch to include models + tests w/migrate script | 21:45 |
| vipul | grapex: you might need some tequila just for good measure | 21:46 |
| vipul | ok anything with snapshots? | 21:46 |
| vipul | if not... | 21:46 |
| SlickNik | nope, sounds good. | 21:46 |
| grapex | vipul: Some shots to go along with snapshots? | 21:46 |
| vipul | heh | 21:46 |
| esp | grapex: I think I would read more wiki articles if I did it your way :) | 21:46 |
| SlickNik | heh | 21:47 |
| vipul | #topic Releasing python-reddwarfclient | 21:47 |
| *** openstack changes topic to "Releasing python-reddwarfclient (Meeting topic: Reddwarf)" | 21:47 | |
| vipul | so i put this in there just cuz there was some confusion around how to release | 21:47 |
| vipul | and the new process | 21:47 |
| vipul | SlickNik made a patch to openstack-ci to support auto-releasing of python-reddwarfclient on a git tag | 21:47 |
| vipul | #link https://wiki.openstack.org/wiki/Release-python-reddwarfclient#Releasing_python-reddwarfclient | 21:48 |
| datsun180b | I remember seeing that | 21:48 |
| grapex | Btw, thanks SlickNik! | 21:48 |
| vipul | I tried to capture what i know of the process ^ | 21:48 |
| grapex | vipul: So as to frequency: I feel like the client almost never becomes backwards incompatable | 21:48 |
| datsun180b | Seems to be the case | 21:49 |
| vipul | gist of it is.. anyone in reddwarf-core can push a tag to gerrit 'git tag 1.1.1', 'git push --tags gerrit' | 21:49 |
| grapex | How would you all feel to a release following the completion of each new feature? | 21:49 |
| SlickNik | There's two things that come into effect when you tag a changeset in git and push it to gerrit... | 21:49 |
| datsun180b | how do we put the fire of someone accidentally tagging a busted release with a numeric version | 21:49 |
| datsun180b | i assume I'm going to do that twice | 21:49 |
| grapex | datsun180b: We write their name on the board or add a line to it. At five strikes we call their parents. | 21:50 |
| datsun180b | i guess gerrit will catch it first? | 21:50 |
| esp | hmm..so can we post tag? | 21:50 |
| datsun180b | grapex: what if batman does it | 21:50 |
| esp | lol | 21:50 |
| vipul | datsun180b: i think you may need to push twice to replace the 'bad' version | 21:50 |
| SlickNik | 1. If the tag matches an alpanumeric regex of the form d*.d*.d*.<word>d*, it is a pre-release and will make a tarball and upload it to tarballs.openstack.org | 21:51 |
| esp | gotcha | 21:51 |
| cp16net | oh thats awesome | 21:51 |
| SlickNik | (where <word> is alpha/beta/a/b/c/d/e/g | 21:51 |
| clarkb | tags should be treated as immutable fwiw | 21:51 |
| clarkb | you cannot garuntee that a tag that is moved will update sanely everywhere | 21:52 |
| datsun180b | that's very important | 21:52 |
| SlickNik | 2. If the tag matches a release version such as 0.1.2, it's treated as a release and will tarball and also upload to pypi... | 21:52 |
| grapex | clarkb: Will it also push the docs to PyPi? | 21:52 |
| SlickNik | Right now it doesn't push the docs, grapex. | 21:53 |
| clarkb | no docs. there are docs jobs that do other things | 21:53 |
| grapex | SlickNik: Ok. | 21:53 |
| clarkb | you can have a job to do that in theory though | 21:53 |
| vipul | clarkb: what's the relationship between the version='xx' in setup.py and git tag 'version' | 21:54 |
| clarkb | rtfd and docs.o.o are currently supported | 21:54 |
| vipul | is there one that takes precedence? | 21:54 |
| clarkb | vipul you shouldnt use a hard coded on anymore | 21:54 |
| clarkb | using oslo you get dynamic versions based on git tags. mordred knows all about it because he wrote it | 21:55 |
| vipul | Ah yes.. ok | 21:55 |
| vipul | so we should probably setup.py from olso in too.. | 21:55 |
| SlickNik | grapex/clarkb, there might be some figuring out to do with credentials for the pypi upload, so we might need to flush that out when we're ready to release to pypi... | 21:55 |
| vipul | #action vipul to look into bringing in setup.py from oslo into python-reddwarfclient so version remains dynamic | 21:55 |
| grapex | SlickNik: Ok. | 21:56 |
| grapex | clarkb SlickNik vipul: Thanks for all your work on this. | 21:56 |
| vipul | yup thanks guys | 21:56 |
| SlickNik | np, anytime! | 21:57 |
| cp16net | +1 | 21:57 |
| SlickNik | thanks clarkb! | 21:57 |
| vipul | ok anything else about versioning/releasing client? | 21:57 |
| vipul | btw, reddwarf doens't use either, it pulls tarball from github directly | 21:58 |
| SlickNik | I'll update that wiki page with the actual release/pre-release regexp locations for reference… | 21:58 |
| vipul | yes, please do SlickNik, thanks | 21:58 |
| vipul | #topic Open Discussion | 21:58 |
| *** openstack changes topic to "Open Discussion (Meeting topic: Reddwarf)" | 21:58 | |
| vipul | anything else we need to cover | 21:59 |
| SlickNik | #action update the release wiki page with regexps for release/pre-release | 21:59 |
| SlickNik | #action ^^^ SlickNik | 21:59 |
| SlickNik | nope, nothing else from my end. | 21:59 |
| vipul | wow look at that we might make time | 21:59 |
| grapex | vipul: We are on fire today! | 22:00 |
| vipul | w00t | 22:00 |
| SlickNik | hellz yea! | 22:00 |
| cp16net | have a great day ev1 | 22:00 |
| vipul | #endmeeting | 22:00 |
| *** openstack changes topic to "OpenStack meetings (alternate) || Development in #openstack-dev || Help in #openstack" | 22:00 | |
| openstack | Meeting ended Tue Mar 12 22:00:31 2013 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 22:00 |
| openstack | Minutes: http://eavesdrop.openstack.org/meetings/reddwarf/2013/reddwarf.2013-03-12-21.01.html | 22:00 |
| openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/reddwarf/2013/reddwarf.2013-03-12-21.01.txt | 22:00 |
| openstack | Log: http://eavesdrop.openstack.org/meetings/reddwarf/2013/reddwarf.2013-03-12-21.01.log.html | 22:00 |
| vipul | 1 hour exactly | 22:00 |
| cp16net | peace out | 22:00 |
| vipul | later | 22:00 |
| esp | :) | 22:00 |
| SlickNik | nice. later guys.. thanks! | 22:00 |
| juice | where is kaganos? | 22:01 |
| juice | kagan? | 22:01 |
| juice | :) | 22:01 |
| kagan | i'm here | 22:01 |
| vipul | kagan's just happy he's not talking about percona anymore | 22:01 |
| juice | no questions :) | 22:01 |
| kagan | ok then ... | 22:01 |
| kagan | anyone actually created a percona image so far ? | 22:01 |
| juice | that's a wrap | 22:01 |
| esp | kagan: on my list. I tried last week but my env was hosed. | 22:02 |
| kagan | i guess that means "no" … ;) | 22:02 |
| *** esp has left #openstack-meeting-alt | 22:03 | |
| *** demorris has quit IRC | 22:03 | |
| grapex | Awesome. Talk to you later guys! | 22:04 |
| *** saurabhs has left #openstack-meeting-alt | 22:04 | |
| *** sacharya has quit IRC | 22:07 | |
| *** jcru has quit IRC | 22:08 | |
| juice | see ya grapex | 22:08 |
| *** datsun180b has quit IRC | 22:13 | |
| *** cp16net is now known as cp16net|away | 22:17 | |
| *** demorris has joined #openstack-meeting-alt | 22:19 | |
| *** robertmyers has quit IRC | 22:23 | |
| *** djohnstone has quit IRC | 22:26 | |
| *** kmansel has quit IRC | 22:34 | |
| *** jdbarry1 has quit IRC | 22:44 | |
| *** demorris has quit IRC | 22:50 | |
| *** cloudchimp has quit IRC | 22:57 | |
| mordred | vipul, grapex: I'm just now changing all of that - ping me before you do work on setup.py | 23:41 |
| grapex | mordred: Thanks for the heads up! I'll pass it along. | 23:52 |
| SlickNik | will do, mordred... | 23:54 |
| mordred | grapex, SlickNik: I mean, you should DEFINITELY get on the oslo version bandwagon - I just made is 100% better yesterday and it's not released/landed yet | 23:54 |
| mordred | you're gonna think it's SO SEXY | 23:54 |
| SlickNik | w00t, let me check it those curves on gerrit… | 23:55 |
| *** vipul is now known as vipul|away | 23:56 | |
| SlickNik | mordred: is this it? https://review.openstack.org/#/c/24033/ | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!