Friday, 2010-11-05

*** littleidea has joined #openstack00:00
*** EdwinGrubbs has quit IRC00:02
*** jakedahn has quit IRC00:02
*** rlucio has quit IRC00:05
*** mischer has quit IRC00:15
*** jc_smith has quit IRC00:16
*** jc_smith has joined #openstack00:16
*** hazmat has quit IRC00:18
*** littleidea_ has joined #openstack00:21
*** littleidea has quit IRC00:21
*** littleidea_ is now known as littleidea00:21
*** khussein has quit IRC00:23
*** littleidea has quit IRC00:24
*** littleidea has joined #openstack00:24
*** littleidea has quit IRC00:25
*** gustavomzw has joined #openstack00:27
*** Cybodog has joined #openstack00:34
*** Cybodog has quit IRC00:39
*** joschi has quit IRC00:39
*** pvo has left #openstack00:40
*** joschi has joined #openstack00:45
*** silassewell has quit IRC00:45
*** pvo has joined #openstack00:50
*** ChanServ sets mode: +v pvo00:50
*** littleidea has joined #openstack00:50
*** cole has joined #openstack00:57
*** cole has quit IRC00:59
*** gaveen has quit IRC01:12
*** dendrobates is now known as dendro-afk01:17
*** mdomsch has quit IRC01:21
*** dendro-afk is now known as dendrobates01:24
*** mdomsch has joined #openstack01:25
*** khussein has joined #openstack01:26
uvirtbotNew bug: #671236 in swift "LogProcessor fails first attempt to use it's InternalProxy" [Undecided,New]
*** littleidea has quit IRC01:26
*** khussein has left #openstack01:27
*** littleidea has joined #openstack01:35
notmynameheh. we found that bug earlier today, and there is already a merge proposal for it01:38
*** schisamo has quit IRC01:41
*** dendrobates is now known as dendro-afk01:48
*** pvo has quit IRC01:59
*** dagger has quit IRC02:03
*** pvo has joined #openstack02:05
*** ChanServ sets mode: +v pvo02:05
*** dagger has joined #openstack02:06
*** dagger has joined #openstack02:06
*** pvo has quit IRC02:07
*** pvo has joined #openstack02:23
*** ChanServ sets mode: +v pvo02:23
*** Cybodog has joined #openstack02:33
*** ArdRigh has quit IRC02:34
*** littleidea has quit IRC02:38
*** littleidea has joined #openstack02:48
*** pvo has quit IRC02:50
*** adeel has joined #openstack02:57
*** pvo has joined #openstack02:57
*** ChanServ sets mode: +v pvo02:57
*** Cybodog has quit IRC03:08
*** pvo has quit IRC03:12
*** joearnold has joined #openstack03:14
*** matiu_ has joined #openstack03:15
*** matiu_ has quit IRC03:25
*** joearnold has quit IRC03:34
*** pvo has joined #openstack03:46
*** ChanServ sets mode: +v pvo03:46
*** perestre1ka has joined #openstack04:08
*** perestrelka has quit IRC04:12
*** joearnold has joined #openstack04:20
*** itaifrenkel has joined #openstack04:30
itaifrenkelhello. Ive got a newbie Q.  Where can I find the test results report after running "python" ? (assuming I already closed the terminal)04:31
*** joearnold has quit IRC04:39
*** miclorb_ has quit IRC04:44
*** littleidea has quit IRC04:46
*** terje has quit IRC04:46
*** terje has joined #openstack04:47
*** littleidea has joined #openstack04:47
itaifrenkelhello. Ive got a newbie Q.  Where can I find the test results report after running "python" ? (assuming I already closed the terminal)04:48
*** ArdRigh has joined #openstack04:54
creihtitaifrenkel: I think most of the devs are probably asleep, but hopefully you will get your answer sooner or later :)04:54
*** DubLo7 has joined #openstack05:01
alekibangoitaifrenkel:   in   ./run_tests.err.log and in ./_trial_temp/05:02
*** joearnold has joined #openstack05:03
*** miclorb has joined #openstack05:03
itaifrenkelcreiht: tahnk you05:14
*** pvo has quit IRC05:23
*** joearnold has quit IRC05:24
*** stewart_ has joined #openstack05:30
*** stewart has quit IRC05:31
*** itaifrenkel has quit IRC05:32
*** stewart_ is now known as stewart05:35
*** ambo has left #openstack05:37
*** perestre1ka has quit IRC05:44
*** perestrelka has joined #openstack05:49
*** miclorb has quit IRC05:50
*** f4m8_ is now known as f4m805:52
*** zaitcev has quit IRC06:03
*** ArdRigh has quit IRC06:16
*** mdiver has joined #openstack06:18
*** ArdRigh has joined #openstack06:20
*** schisamo has joined #openstack06:25
*** miclorb has joined #openstack06:28
*** stewart has quit IRC06:35
*** stewart has joined #openstack06:48
*** DubLo7 has quit IRC06:55
*** romain_lenglet has joined #openstack06:59
*** arcane has quit IRC07:01
*** arcane has joined #openstack07:08
*** arthurc has joined #openstack07:09
*** miclorb has quit IRC07:18
*** metoikos has quit IRC07:18
*** miclorb has joined #openstack07:19
*** miclorb has quit IRC07:23
*** schisamo has quit IRC07:29
*** alekibango has quit IRC07:34
*** arcane has quit IRC07:36
*** arcane has joined #openstack07:37
*** allsystemsarego has joined #openstack07:41
*** dysinger has quit IRC07:54
*** alekibango has joined #openstack07:55
*** guigui1 has joined #openstack08:04
*** mdiver has quit IRC08:11
*** Dweezahr has quit IRC08:19
*** AlanBell has joined #openstack08:22
*** Dweezahr has joined #openstack08:34
*** adeel has quit IRC09:13
*** ibarrera has joined #openstack09:14
*** miclorb_ has joined #openstack09:24
*** nijaba has quit IRC10:04
*** nijaba has joined #openstack10:04
*** metoikos has joined #openstack10:26
*** ddumitriu has joined #openstack11:07
*** ctennis has quit IRC11:31
*** localhost has quit IRC11:33
*** ddumitriu has quit IRC11:33
*** localhost has joined #openstack11:34
*** ctennis has joined #openstack11:47
*** ctennis has joined #openstack11:47
*** Kdecherf has quit IRC11:51
*** Kdecherf has joined #openstack11:56
*** miclorb_ has quit IRC12:04
*** mdiver has joined #openstack12:15
*** ibarrera has quit IRC12:23
*** dizz|away has quit IRC12:28
*** gustavomzw has joined #openstack12:30
*** gustavomzw has quit IRC12:32
*** ibarrera has joined #openstack12:36
*** jdarcy_ has joined #openstack12:39
*** westmaas has joined #openstack12:49
*** mdiver has quit IRC12:55
*** PiotrSikora has quit IRC12:58
*** Kdecherf has quit IRC12:59
*** Kdecherf has joined #openstack12:59
*** schisamo has joined #openstack13:00
*** mdiver has joined #openstack13:01
*** PiotrSikora has joined #openstack13:01
*** pvo has joined #openstack13:01
*** ChanServ sets mode: +v pvo13:01
*** blpiatt has joined #openstack13:07
*** Cybodog has joined #openstack13:14
*** ibarrera has quit IRC13:20
*** jdarcy_ is now known as jdarcy13:22
*** taihen has quit IRC13:24
*** taihen has joined #openstack13:25
*** dendro-afk is now known as dendrobates13:25
dendrobatessoren ttx: If I create some 30min sessions after lunch13:30
*** ibarrera has joined #openstack13:33
sorendendrobates: yes?13:33
ttxdendrobates: yes?13:33
*** jkelly has joined #openstack13:39
*** littleidea has quit IRC13:42
*** littleidea has joined #openstack13:43
*** piken_going_home is now known as piken13:44
*** mischer has joined #openstack13:48
*** doude has joined #openstack13:49
*** Podilarius has joined #openstack13:51
*** dizz has joined #openstack13:51
*** dizz is now known as dizz|away13:55
dendrobatessoren ttx: If I create some 30min sessions after lunch, we should have enough slots for all the bps13:57
dendrobateswhat do you guys think13:57
dendrobateswe have some bps that are not large enough for an hour13:58
ttx"enough slots" meaning 3 parallel sessions at all time over 2 days right13:58
ttxor 2 parallel sessions at all times over 3 days ?13:58
ttxdendrobates: ^13:58
dendrobates3 parallel over 2 days with overflow onto the 3rd13:59
ttxdendrobates: my only gripe is that people will only be able to attend 33% of the sessions. Sounds like a scheduling nightmare to get everyone to be able to attend the sessions where they are "necessary"14:01
dendrobatessoren: opinion?14:02
*** smoser has joined #openstack14:02
ttxdendrobates: spreading over 3 days allows most people to attend 66% of the sessions. Maybe it's a bit too much.14:03
smoserhi. is there any discussion happening next week on bootloaders?  Ie, rather than loading a kernel (as is done in ec2), loading a bootloader either from partition information or as a 'kernel' ?14:03
smoserand dendrobates
ttxsmoser: Haven't seen any blueprint around that.14:04
dendrobatesvishy creiht: opinion?  which should we optimize for concurrency or time?14:04
smoseri didn't see one either, but failure to find something in a blueprints search didn't mean a lot to me14:04
*** pvo has quit IRC14:07
*** taihen has quit IRC14:07
*** taihen has joined #openstack14:08
*** hazmat has joined #openstack14:08
ttxdendrobates: My vote goes to spread over 3 days, and have soren/mtaylor specify which sessions they need to attend (so that we schedule them the first two days). Pack them at the beginning, leave some empty slots at the end.14:08
dendrobatesttx: ugh, that sounds like more scheduling work for me.14:09
ttxdendrobates: no pain, no gain :P14:10
dendrobatesttx: I want to get everyones opinion before we make a decision.  3 days does not sound too bad, day 4 o UDS is when my head starts to get fuzzy.14:10
ttxdendrobates: oh sure ! If most people are comfortable with 2 days with 3 parallel sessions + 2 hacking days, I don't mind14:11
ttxIt just sounds like people will miss a lot of sessions, then have two full hacking days...14:12
ttx(that's based on UDS experience, not Austin Design Summit experience, so feel free to fix me)14:12
creihtMy opinion is that there are too many to discuss :)14:14
ttxcreiht: too many subjects ? Or too many where you want to be present ?14:15
creihtboth :)14:15
ttxah :)14:15
creihtDo you really think you are going to implement that many blueprints in the next 6 months? :)14:16
dendrobatescreiht: no, but that is part of the process14:16
creihtI also doubt there are many blueprints that warrant a full 1 hour discussion14:16
creihtbut those are just my opinions :)14:17
dendrobatescreiht: we need to figure out what is important and reasonable14:17
creihtYou know what would be interesting, is to have a lightning talk session, where everyone has 5 mins to talk about their blueprint14:17
creihtthen vote14:17
creihtthe ones that get votes get discussed :)14:18
dendrobatescreiht: that is actually a good idea.  We should try that next time.14:18
* ttx admits to have trouble picking only 13 sessions to attend over a list of 54.14:19
creihtSo my vote then would be to keep 2 days and make a mjority of the sessions 30 mins instead of an hour14:19
creiht2 tracks14:19
dendrobatesnot another option! :)14:20
creihtthat's still my vote :)14:20
ttx30 min. is small -- especially if you keep 5 min. to switch between sessions. There are a few subjects that would definitely fit the bill, but the majority needs at least 40min.14:21
sorendendrobates: Yes?14:22
dendrobatesThat's why I was thinking about 30min session after lunch.14:22
sorendendrobates: Oh, /me reads scrollback14:22
creihtperhaps start with 30 mins, and if it requires more time, have an overflow area where the people really concerned with it can finish talk14:22
creihtthe nice thing about 30 mins is that it forces you to keep the discussion focused14:23
creihtA lot of the discussions last year seemed to drag on14:23
smosersoren, i wouldn't mind discussing non-linux-kernel loading at the summit, if there is time available and a solution hasn't been already agreed on.14:24
creihtJust some observations14:25
* ttx looks at ways to accomodate 9 slots instead of 6.14:25
*** littleidea has quit IRC14:25
dendrobatescreiht: it should be different this time.  Hopefully more focused.14:25
sorensmoser: What's currently being worked on is whole-disk-images where we'd just boot whatever is on there. Put a boot loader in the image, and you get to boot.14:25
sorendendrobates: I don't know really. I don't like the 30 min + overflow to somewhere else idea, though. It makes scheduling even more impossible. We'd have to make sure people don't have two sessions in a row that they're "essential" for or something.14:27
*** littleidea has joined #openstack14:27
*** blakeyeager has joined #openstack14:27
smosersoren, that sounds reasonable... its not backwards compatible, but maybe that doesn't matter (its easy enough to wrap a partition image in a partition table).14:28
dendrobatessoren creiht: do you have any issues with extending to Thursday?14:28
sorendendrobates: Well, sure :)14:28
dendrobatesI mean apart from that14:29
sorendendrobates: But I guess it's better than three session at a time over two days.14:29
smoserthe EC2 cluster compute instantce types use xen HVM, so I'm guessing they have some accomodation for this in the ec2 api.14:29
ttxsoren: you will be able to attend the same number of sessions anyway14:29
sorenttx: I'll be gone by Thursday, remember?14:29
ttxsoren: yes.14:29
*** westmaas has quit IRC14:29
smoserits possible that they just key whole disk versus partial disk off of a register argument indicating "hvm".14:29
dendrobatesif we extend to 3 days and mix in some 30 min sessions that should make things more manageable14:29
sorenttx: Oh, I see what you mean.14:30
creihtdendrobates: It doesn't matter to me much since we have only like 4 or so blueprints to discuss :)14:30
* soren was reading too fast14:30
ttxsoren: but adding more sessions on the trwo first days will not make you able to attend more.14:30
ttxsoren: heh :)14:30
creihtdendrobates: And I volunteer the swift discussions for the 30 min slots :)14:30
ttxsoren: you just need to make sure the ones you care about are scheduled over the first two days.14:30
sorenttx: I care about all of them!14:30
ttxsoren: then prepare your triplicates clones, you'll need them :)14:31
dendrobatesannegentle: your content stuff seems to be scheduled in the business tracks, is that where you want to be?14:31
ttxso we have...14:31
dendrobatessoren: then maybe you should attend the entire summit.14:31
ttxOption 1: three parallel 55min sessions over two days, a few 30 min slots.14:32
ttxOption 2: two parallel 55min sessions over three days14:32
ttxOption 3: two parallel 35-40min sessions over two days14:32
dendrobatesoption -1:  Screw the summit, lets drink beer14:32
creihtsummiting is hard, lets go shopping14:33
ttxoption -2: screw planning, let's unconference it14:33
creihtoption -2++14:33
dendrobatesttx: they main problem with the unconferences is that they are usually useless.14:33
ttxdendrobates: agreed. Difficult to get outcomes14:34
ttxdendrobates: my vote is on 2, 3, 114:34
dendrobatesI like 2 and 3 combined.  4 days with some 30 min sessions.14:35
dendrobateser 3 days14:35
ttx2+3... that would make... option 5.14:35
*** cole has joined #openstack14:35
*** arcane has quit IRC14:35
* ttx starts to regret having chosen decimal14:35
ttx3 days x 30min. sessions... that would make like 1.6 sessions at the same time.14:36
ttxdendrobates: then my vote is on 2, 3, 5, 1.14:37
*** gustavomzw has joined #openstack14:42
*** arcane has joined #openstack14:42
ttxdendrobates: option 1 -> 13 time slots. option 2 -> 20 time slots. option 3 -> 19 time slots. option 5 -> 29 time slots.14:42
ttxfor... 54 accepted blueprints.14:43
*** dragondm has joined #openstack14:43
sorendendrobates: *grubmle*14:45
sorendendrobates: Whose idea is it anyway that these things start on Tuesdays?14:45
sorenWhat is wrong with Mondays? Nothing!14:45
* soren likes Mondays14:45
ttxsoren: for those attending two days... optoin 1 -> 13 time slots. option 2 -> 13 time slots. option 3 -> 19 time slots. option 5: 19 time slots.14:46
*** kirkland has quit IRC14:46
coleI dislike traveling on a sunday to start a conference on monday, i'd rather use a work day to travel than my weekend =)14:46
*** f4m8 is now known as f4m8_14:46
* ttx pauses14:48
*** gustavomzw has quit IRC14:48
sorencole: Live closer.14:49
sorencole: Or deal.14:49
*** gondoi has joined #openstack14:49
sorencole: But good point.14:50
annegentledendrobates: I'm giving a talk about content in the business track, yes. Also running the doc sprint Thurs. and Fri.14:53
*** johnpur has joined #openstack14:53
*** ChanServ sets mode: +v johnpur14:53
annegentledendrobates: we should have a good graphic for the logo on the registration site today, Todd and I are workin' on it.14:54
*** guigui1 has quit IRC14:55
colei should have given a talk on US Joint Task Force Special Publication 800-53 and FIPS 140-2 compliance for cloud infrastructure14:56
*** mdomsch has quit IRC14:59
*** infernix has quit IRC15:04
*** migrantgeek has joined #openstack15:05
*** migrantgeek has left #openstack15:06
*** littleidea has quit IRC15:09
*** littleidea has joined #openstack15:09
*** littleidea has quit IRC15:14
*** littleidea has joined #openstack15:14
*** befreax has joined #openstack15:15
*** infernix has joined #openstack15:21
dendrobatesI'm still leaning towards option 5, I'm going to start rearranging the schedule in 90 mins.  So now is the time to make any arguments15:24
ttxOption 5 is fine by me. As long as slots are at least 40min long I think we are ok.15:25
sorendendrobates: Sorry, what is option 5? I see options 1-3?15:25
ttxsoren: <ttx> 2+3... that would make... option 5.15:26
*** ppetraki has joined #openstack15:26
ttx<dendrobates> I like 2 and 3 combined.  3 days with some 30 min sessions.15:26
sorenSure, that's ok with me. I may have a preference, but not enough to care to argue about it.15:26
ttxdendrobates: 5 with a mix of 55 and 30 min sessions ? Or with all 40-45 min sessions15:26
dendrobates5 with a mix of 55 and 30 min sessions15:27
ttxdendrobates: that should work. Wouldn't work at UDS because too many people play with scehduling, but here it should be manageable15:27
annegentlecole: Like the installation prompts, let me know if you want any writing help. I'm soooo glad to see it scripted15:28
dendrobatesannegentle: I saw your screencast.  Awesome!15:28
ttxdendrobates: (to keep the 30min. sessions in the 30min slots)15:28
annegentledendrobates: I uh um uh er thanks :)15:30
*** dysinger has joined #openstack15:31
sorenannegentle: Where can I see this screencast?15:31
annegentlesoren: here ya go
creihtannegentle: that's awesome15:40
*** mdiver has quit IRC15:41
* soren pauses for dinner and such15:42
pikenGod, I hate writing specs. just coding is so much nicer. lol15:42
*** cole has quit IRC15:42
pikenwho cares if you forget what your coding half way through.15:43
*** pvo has joined #openstack15:44
*** pvo has joined #openstack15:44
*** ChanServ sets mode: +v pvo15:44
*** kirkland has joined #openstack15:47
*** ctennis has quit IRC15:49
*** metcalfc has joined #openstack15:50
*** metcalfc has joined #openstack15:50
*** mdiver has joined #openstack15:51
ttxtermie: if you're attending the summit, you should register at
*** pvo has quit IRC15:59
*** metoikos has quit IRC16:01
*** pvo has joined #openstack16:02
*** ChanServ sets mode: +v pvo16:02
*** abecc has joined #openstack16:04
termiettx: i don'tthink i am :/16:04
termiettx: my location doesn;t make it so easy16:05
termiei want to go to the tokyo meetup though16:05
termie... since i am in tokyo at the moment16:05
termiebut that won't stop me from dropping tons of blueprints on y'all16:05
termieTEXAS BABy16:05
ttxtermie: ah ok.. you're marked as drafter of one of the specs to be discussed in San Antonio16:05
* ttx checks16:05
termiettx: hopeso16:05
termiettx: but i think i am making some others lead the discussion16:06
termiettx: i have mr. mckenty signed up to be my avatar16:06
ttxtermie: ok, could you set them as drafter, even if that's temporary ?16:06
ttxtermie: we (ab)use the drafter field to have an idea of who will lead the discussion at the summit16:07
*** mdiver has quit IRC16:07
*** mischer has quit IRC16:08
*** mdiver has joined #openstack16:08
*** abecc has joined #openstack16:09
*** ibarrera has quit IRC16:10
edayttx, termie: I can help represent that one too :)16:12
edayre: scheduling (reading back), I'm for making the least amount of concurrency. Since most of the blueprints are mostly on nova, it's hard to split them since you want the same folks in those talks. It's not like UDS where you have lots of very different projects, this is all pretty tightly coupled :)16:16
dendrobateseday: even uds has concurrency issues.  I agree though.  We will try to minimize concurrency.16:18
*** schisamo has quit IRC16:19
dendrobatessoren: have you been in the summit sw today?  Trying to edit it now results in a traceback16:22
edaydendrobates: I'm sure, but this is probably even more difficult. 2 concurrent sessions will already be pretty tough, 3 would be nearly impossible to have the right devs there16:22
dendrobatessoren Daviey:  any ideas.16:27
*** littleidea_ has joined #openstack16:28
*** littleidea has quit IRC16:28
*** littleidea_ is now known as littleidea16:28
*** ppetraki has quit IRC16:30
edaysoren: can you look at ? you had a 'needs fixing', want to be sure you approve now16:31
*** daleolds has joined #openstack16:32
_0x44dendrobates: For the S3 blueprint should I have put it in a different definition status? It doesn't really need any explicit definition beyond "s3" needs to be supported to the same extent we support swift/filesystems/pants16:34
dendrobatesI think it's fine.  Who is going to lead the discussion at t eh summit?16:35
_0x44dendrobates: I don't know that it needs discussion, I've already written it. I just haven't committed to a branch because it's waiting on resolution of the backend-imports blueprint16:37
*** msinhore has joined #openstack16:37
*** silassewell has joined #openstack16:37
eday could probably be combined with distributed-data-store, we can talk nosql there (or it may be moot given outcome)16:37
dendrobatesok, good to know16:37
_0x44dendrobates: But since it's a feature I didn't want to use a bug ticket to track it.16:37
dendrobatesttx can you remove nosql-datastore-adapter16:37
_0x44dendrobates: Thanks :)16:38
ttxdendrobates: declined.16:39
*** pvo has quit IRC16:39
*** silassewell has quit IRC16:39
*** schisamo has joined #openstack16:40
johnpur_0x44: you should talk to creiht about s3-swift support. it would be interesting to see if your code will hook up to his implementation.16:42
Davieydendrobates: that is not an error i've seen :/16:43
Davieydendrobates: Frustratingly, it's very hard to debug without data that makes it go bang. :/16:44
dendrobatesDaviey: D you recommend flushing the db and starting over16:46
*** jkakar has joined #openstack16:46
Davieydendrobates: or giving me a db dump ;)16:46
*** gustavomzw has joined #openstack16:47
dendrobatesDaviey: you have access to the server.  summit.openstack.org16:47
dendrobatesyour ssh keys are on it16:47
Davieydendrobates: Oh... right... what is my userid?16:47
dendrobatesDaviey: and everything is under /work16:47
Davieydendrobates: how very LSB of you :P16:48
*** joearnold has joined #openstack16:49
*** gustavomzw has quit IRC16:49
dendrobatesDaviey: daviey16:49
jaypipesdendrobates, eday: yeah, good call on declining nosql datastore...16:49
Davieyjaypipes: But will it web scale?16:49
pikenWhat no  MongoDB is webscale you know.16:49
creiht/dev/null is webscale :)16:50
jaypipesDaviey: the decision was to scrap it and use the distributed-data-store blueprint instead, that's all.16:50
jaypipescreiht: yes, that is much more scalable.16:50
dendrobates/dev/null is Web scale 2.0: Cloud scale16:50
pikenno, /dev/null is OS2 scale ;)16:51
Davieydendrobates: We've been talking about writing a persistence storage backend for /dev/null16:51
pikenWhat gets me though. Is when ever I use that POC, I can never get the right data back from a query... Maybe I should have used erlang......16:51
*** kirkland has quit IRC16:52
ttxDaviey: done already @
edayDaviey: just map /dev/null for writes, /dev/random for reads. Can't be much worse than the quality of information out there already :)16:53
edaypiken: you talking trash about erlang?!?16:53
*** kirkland has joined #openstack16:53
pikenHey, it is good at what it was made for. Network switching for VOIP, Cellular, and TCP/IP networks.16:54
pikenBut if I see one more oh, lets use erlang so I can use a trendy word and technology that is shit outside of its scope, I think I will puke. lol16:54
_0x44johnpur: This is for glance16:55
_0x44johnpur: It's a backend for the image repository16:55
pikenAnd yes, I have written a ton of erlang for highspeed application data routing on top of clustered apache at HostWay when I worked there. lol16:55
Davieyttx: lol, not seen that!16:56
ttxDaviey: "All complaints and feature requests will be immediately stored using our S4-backed user request database."16:56
_0x44johnpur: We want the image proxy/registry to be able to use any number of backends so we no longer have to worry about hosting them anywhere. People can just register an image with parallax and have it get into nova with no different interaction than any other image.16:56
*** mischer has joined #openstack16:57
edaypiken: I've written clustered object stores and am writing a SQL database in it now... it works fine for other applications too, IMO :)16:57
_0x44johnpur: So swift s3 support won't really help16:59
ttxok, eow here... See you on Monday evening, I guess17:01
pikeneday: oh god, I think I just had an aneurysm. I would never wnat to try to go that far with it. Data switching is where I draw the line. lol17:01
dendrobatesttx: bye17:01
edaypiken: there are right and wrong ways to go about it, heh17:02
edaypiken: check out some of the libs Basho/Riak have17:02
*** abecc_ has joined #openstack17:04
*** abecc has quit IRC17:04
*** abecc_ is now known as abecc17:04
pikenI have seen Riak. I am not a fan. lol17:04
*** jbryce has joined #openstack17:05
Davieydendrobates: looking now17:06
edaypiken: I take it you don't like couchdb either? :)17:08
pikenno, not really. lol. But I like it better then Riak17:08
dendrobatesyou know what I hate, mysql, and anyone whoever worked on it ;)  Important data belongs on post-it notes, stuck under keyboards.17:10
*** khussein has joined #openstack17:10
dendrobatesthat's web scale17:10
_0x44dendrobates: I think there's a post-it notes under keyboards engine for MySQL Cluster Grade Enterprise.17:11
Davieydendrobates: austin-microsoft-hyper-v-support is the spec causing bangage17:12
*** neogenix has joined #openstack17:12
dendrobatesDaviey: I'll fix it and re-import17:12
Davieydendrobates: I have the page open, shall i just del it?17:13
dendrobateswhat is the actual problem with the bp?17:13
Davieydendrobates: TBH, not sure.. haven't investigated that far :)17:15
dendrobatesDaviey: the bp looks fine, is the problem with the meeting in the summitdb?17:15
dendrobatesDaviey: how did you determine that it was that bp causing the problem>17:16
Davieydendrobates: i caught the error at run time, and print's :)17:16
Davieydendrobates: Interesting... re-importing it has caused it to go bang again17:16
DavieyI wonder what has caused that :/17:16
dendrobatesDaviey the wiki url has spaces in it.17:17
dendrobatesI'll fix it.17:17
Davieywhy would someone do that!? :)17:17
dendrobatesDaviey: MS people wrote it and they tend to use spaces in filenamed17:17
Davieyah, sounds reasonable17:18
dendrobatesI have verified that it is the wiki url17:18
neogenixHave the technical track agendas been published?17:19
dendrobatesThe url is not handled correctly in the code.  Its really a bug17:19
dendrobatesneogenix: not yet, in a few hours17:19
Davieydendrobates: ack17:19
neogenixdendrobates: Great, I've a few people asking around here, so I'll tell them I'll send it along to them as soon as it's out and about :D17:20
edayhm, our rss feed is broken. every time there is a new post a bunch of older articles get duplicated17:20
* neogenix is sitting next to khussein, talking OS 17:20
dendrobateslunch, then schedule17:20
*** doude has quit IRC17:20
*** jakedahn has joined #openstack17:21
creihtdendrobates: lol17:21
khussein* is distracted by neogenix17:22
Davieydendrobates: fixed17:23
*** Pentheus has quit IRC17:24
*** ctennis has joined #openstack17:24
*** ctennis has joined #openstack17:24
*** littleidea has quit IRC17:25
vishychuckle, just got to the part where the video completely avoids trying to pronounce my name17:28
*** dizz|away has quit IRC17:29
vishy"an excellent script produced of our community members"17:29
zulgah...3 year old wont nap before speech therapy = cranky child17:29
neogenixvishy: lol, give them the phonetic spelling :D17:30
vishynah it is much more entertaining this way17:30
*** jc_smith has joined #openstack17:30
*** westmaas has joined #openstack17:31
*** mdiver has left #openstack17:32
*** joearnol_ has joined #openstack17:32
creihtvishy: lol17:32
*** joearnold has quit IRC17:34
*** westmaas1 has joined #openstack17:34
*** westmaas has quit IRC17:36
*** jkelly has quit IRC17:36
*** ppetraki has joined #openstack17:42
*** ppetraki has quit IRC17:51
*** littleidea has joined #openstack17:53
*** westmaas1 has quit IRC17:57
devcamcarvishy: which video?17:58
devcamcarand howdy all17:58
edaydevcamcar: howdy18:00
*** blakeyeager_ has joined #openstack18:03
*** khussein has left #openstack18:04
*** khussein has joined #openstack18:04
*** blakeyeager has quit IRC18:05
*** littleidea has quit IRC18:08
*** blakeyeager_ has quit IRC18:08
*** befreax has quit IRC18:09
*** xfrogman5 has joined #openstack18:10
annegentlevishy: hee hee, sorry about that, you community member you. :)18:12
dendrobateshas anyone seen gundlach around?18:14
*** daleolds has quit IRC18:15
*** joearnold has joined #openstack18:16
vishyannegentle: no problem, we had a bunch of the dev team watching the video, and someone said: "I wonder how she is going to pronounce your name?", so we all erupted into laughter at some point.18:17
*** Cybodog has quit IRC18:17
vishys/some point/that point18:17
*** xfrogman5 has quit IRC18:17
*** joearnol_ has quit IRC18:18
*** abecc has quit IRC18:18
*** silassewell has joined #openstack18:20
*** joearnold has quit IRC18:20
annegentlevishy: that's hilarious! I was gonna say "vishy" but realized not everyone knows your IRC handle :)18:20
*** jc_smith has quit IRC18:21
*** ppetraki has joined #openstack18:22
vishyha, most people call me vish, actually.  Unfortunately it was taken on freenode18:25
*** littleidea has joined #openstack18:27
*** xfrogman5 has joined #openstack18:31
*** daleolds has joined #openstack18:33
*** cole has joined #openstack18:34
*** littleidea has quit IRC18:34
*** jeremyb has joined #openstack18:39
*** jc_smith has joined #openstack18:40
*** gustavomzw has joined #openstack18:48
*** devcamcar has left #openstack18:48
*** littleidea has joined #openstack18:51
dendrobatesschedule strawman for the first 2 days:18:51
*** zaitcev has joined #openstack18:53
*** jkelly has joined #openstack18:53
edaydendrobates: I thought tues. morn were general sessions with everyone? (before biz/dev split)18:54
dendrobateseday: just the first hour.18:54
dendrobateseday: there is no way we can get through everything if we give up half of one day.18:55
jkellyIf someone with no real "hands-on" experience with openstack wanted to "dive in", do you think the dev sessions are going to require too much preexisting knowledge?18:57
vishywhat happened to region/datamodel?18:57
dendrobatesthis is not complete, there will be sessions on Thursday too.18:58
dendrobatesvishy: what is the URL for the bp?18:59
colejkelly: tons of good input at first summit when lots of people had no "hands on" experience :)18:59
jkellyThat's reassuring, thanks :)18:59
neogenixdendrobates: final?18:59
dendrobatesneogenix: not yet19:00
vishydata one
neogenixdendrobates: k. Wondering if my bp will make it :|19:00
dendrobatesneogenix: it is never really final, we can and will add and remove sessions during the summit19:00
neogenixdendrobates: ok.19:00
vishyand region one is here:
dendrobatesvishy: damnit, none of the ones, marked review got imported.19:02
dendrobatesjust what I needed more sessions19:02
neogenixdendrobates: what are the criteria for a bp making it for discussion?19:02
vishyThose two are the important ones19:02
dendrobatesvishy: I'll get them in19:03
*** Cybodog has joined #openstack19:03
*** DubLo7 has joined #openstack19:03
*** sophiap has joined #openstack19:05
uvirtbotNew bug: #671580 in swift "ratelimit breaks without memcache" [Medium,New]
*** sparkycollier has joined #openstack19:10
*** ppetraki has quit IRC19:12
*** gustavomzw has quit IRC19:14
dendrobatesaghhhhhh! we just don't have enough time at the summit to discuss all the blueprints19:22
*** jc_smith_ has joined #openstack19:22
*** joearnold has joined #openstack19:23
vishysome of the blueprints are tiny and probably don't need discussion19:25
vishyyou can pull out mine about project_vpn19:25
vishythe code already exists19:25
vishydoesn't really need a blueprint per se19:25
dendrobateswe definitely don't need to discuss things that are finished, a blueprint is still helpful though19:26
*** jc_smith has quit IRC19:26
*** jc_smith_ is now known as jc_smith19:26
*** dizz has joined #openstack19:27
*** mischer has quit IRC19:27
creihtdendrobates: the talk about large file support can probably be combined with devcamcar's about client-side chunking19:29
pikenThe start of my spec is out at and attached to my blueprint at
dendrobatescreiht: thanks, will do19:29
pikenI have started POC implementation as I write the rest of the spec.19:29
creihtalso for anyone who is interested, the first cut of the swift multi-node docs are up at:
pikenvishy: I have had some luck with a VirtFSManager using your VolumeManager as a base class. As soon as the legal docs are done for corp submissiosn as a corp partner next week. I will get a merge ready for you on your branch.19:31
*** sparkycollier has quit IRC19:32
pikenAlso, I want to put it up for discussion. I have some tests I did using the python-apt module, and the python-yum module. Would it be prefered for the unified installer to use those directly instead of using the command line apt-get and yum??19:33
*** sparkycollier has joined #openstack19:34
edaydendrobates: we could go into friday too, I think most of the devs will be there19:37
edaydendrobates: rabbitmq-ha and bexar-message-queue-celery could probably be combined to "message queue discussion"19:38
*** JimCurry has joined #openstack19:43
sorenErr.. How did I get subscribed to
sorenand more importantly: Why is it being used for newsletters. srsly. That's not announcements.19:45
*** khussein has quit IRC19:45
*** neogenix has quit IRC19:45
edaydendrobates:    file-injection-at-creation and openstack-api-image-modification could probably be dropped for summit discussion, those are pretty self explanatory19:46
johnpurdendrobates: if i search on sprints==ods-b do i get all the bps accepted for discussion? are we still discussing next week both bexar and bexar+1?19:48
dendrobatesjohnpur: yes to both, but some topics have been combined19:49
dendrobatesall of pvo19:49
dendrobatesjohnpur: all of pvo's bps have been condensed to 1 2hour session19:49
johnpurdendrobates: thx. on the sprint tag what does an x versus a check mean?19:51
*** sophiap has quit IRC19:52
dendrobatesjohnpur: x means declined, check means approved19:52
dendrobatesjohnpur: approved does not mean scheduled though19:53
*** joearnold has quit IRC19:53
johnpurlol, sounds dumb, but what does declined mean?19:53
sorenAnyone can propose stuff for any sprint.19:54
colejohn: opposite of accepted :P19:54
sorenThe "owners" of the sprint can accept or decline topics for it.19:54
colei love how openstack summit feels identical to UDS =)19:55
sorenPure coincidence, I'm sure.19:56
johnpurdendrobates: 13 of pvo's bps are declined. should i be concerned?19:56
johnpurcole: i knew there would be a joker in the crowd!19:56
sorenjohnpur: If they hadn't been declined, they'd be listed in the summit system as needing scheduling.19:57
sorenjohnpur: ..and since they've all been turned into a single session..19:57
* creiht adds a "Needs more Ubuntu" blueprint19:57
*** kw1 has joined #openstack19:57
*** kw1 has left #openstack19:58
pikenHow often are the summits going to be. Wasn't the Austin summit like 2 months ago?19:59
johnpursoren: thx.19:59
notmynamepiken: every quarter, I think20:00
notmynameor every 6 months20:00
sorenpiken: It was mid-July.20:01
*** blakeyeager has joined #openstack20:01
pikenah, i thought it was later then that.20:02
*** smoser has quit IRC20:02
pikenWell, I will have to see if I can make it to the next one after this. Once LexisNexis is set as a corp partner/contributer, at least me and a member of 2 from our ops team might be able to show up for some discussions and design.20:03
johnpurmy day to be question man... doesn't show the declined bps. can i get a combined view that shows all the ods-b bps?20:04
edayjohnpur: they should all be marked bexar, but are not, so would be the thing to look at. instead just use:
dendrobateseday: they are not yet20:08
*** littleidea has quit IRC20:08
*** littleidea has joined #openstack20:09
johnpureday|dendrobates: thx.20:09
dendrobatesjohnpur: some of the ones declined have nothing to do with openstack and were incorrectly proposed20:11
johnpurdendrobates: no kidding... interesting reading tho...20:11
dendrobatesjohnpur: most of the RS stuff falls under service-provider-readiness20:12
johnpurdendrobates: i have been asked to summarize the non-rs bps for RAX. figured a good first step would be to ennumerate them...20:13
*** sparkycollier has quit IRC20:13
*** littleidea has quit IRC20:14
*** littleidea has joined #openstack20:15
*** littleidea has quit IRC20:17
vishypiken: awesome.  I guess there were too many differences in implementation to abstract at the driver level?20:18
*** JimCurry has quit IRC20:22
vishysoren: I'm porting my cloudpipe changes.  It involves a number of changes to, including crls and such20:22
*** dizz is now known as dizz|away20:22
vishysoren: we may have to sync up on the move over to m2crypto20:22
sorenvishy: ngh..20:23
sorenvishy: Ok. I'm about 60% done, I'd say.20:23
vishysoren: the main differences are not using intermediate ca's and keeping track of and revoking user certs20:24
*** mdomsch has joined #openstack20:28
*** mdomsch is now known as mdomsch_BOS20:28
*** almaisan-away is now known as al-maisan20:28
sorenvishy: You don't need intermediate CA's anymore?20:30
vishysoren: no they don't really make sense20:30
sorenvishy: Awesome. Then I'm way more than 60% done :)20:30
vishysoren: we just have a different root ca per project20:30
sorenMeh, details, details.20:30
*** al-maisan is now known as almaisan-away20:31
vishyintermediate cas weren't really doing anything useful, and in fact i couldn't really figure out how to configure openvpn to do what we needed using them20:32
vishybtw, using m2crypto for generating key pairs instead of ssh-keygen was about 3x slower20:33
vishydon't know how it is with certs20:33
*** westmaas has joined #openstack20:33
*** dizz|away has quit IRC20:34
*** rlucio has joined #openstack20:36
sorenvishy: The raw keygeneration stuff or all of the fiddling about with the keys, turning them into this format, then the other, etc.?20:36
sorenvishy: ...because I simplified that a /lot/.20:37
vishythe commented out code20:37
dendrobatesvishy: 3x how slow?  .05 , .1 sec, 5 sec20:37
vishydendrobates: I don't remember, it was a long time ago, but i think it was something like 1 second down to 300ms20:37
sorenI can try doing some benchmarks when I'm done.20:38
dendrobatessoren didn't debian find a way to speed up key generation a few years ago?  :)20:38
vishycoolness, as long as m2crypto is using sane c libraries, it should be able to do it at a reasonable speed.20:39
*** silassewell has quit IRC20:39
*** dizz has joined #openstack20:51
*** mdomsch_BOS has quit IRC21:02
*** joearnold has joined #openstack21:02
*** jc_smith has quit IRC21:05
dendrobatesannegentle: how log do you need your doc sprint to be?  Is it a full day?21:05
*** dizz is now known as dizz|away21:06
*** ranjib has joined #openstack21:07
*** mischer has joined #openstack21:07
*** pvo has joined #openstack21:12
*** pvo has joined #openstack21:12
*** ChanServ sets mode: +v pvo21:12
*** Cybodog has quit IRC21:13
*** gustavomzw has joined #openstack21:13
annegentledendrobates: Thursday and Friday morning drop-in if possible. But if we don't have the room, then Thursday.21:13
annegentledendrobates: Logo image fixed on the summit site21:14
dendrobatesWe have an entire extra room for you, so no worries21:14
*** westmaas has quit IRC21:14
annegentledendrobates: excellent, thanks.21:14
*** ranjib has quit IRC21:15
annegentledendrobates: I think the logo's ok even with the red-on-red, we futzed with it quite a bit, though I can do a background image if you need. I think it's good-to-go tho.21:15
dendrobatesI think it looks fine.  Thanks21:16
*** gustavomzw has quit IRC21:17
*** littleidea has joined #openstack21:18
*** xfrogman5 has quit IRC21:22
*** miclorb_ has joined #openstack21:23
redbomtaylor: what do we need to do to get an updated eventlet in the swift-core ppa?21:30
*** allsystemsarego has quit IRC21:33
mtaylorredbo: we just need to upload a new soure packages - do we need to make one?21:33
redboI'd like to, there was a bug fixed in 0.9.13 that we're hitting.21:35
*** littleidea has quit IRC21:37
sorenvishy: m2crypto is a very thin wrapper around OpenSSL.21:37
sorendendrobates: *chuckle*21:37
*** cole has quit IRC21:38
*** miclorb_ has quit IRC21:38
uvirtbotNew bug: #671644 in swift "In swift.proxy.Controller, logic in account_info and container_info is problematic" [Undecided,New]
*** ArdRigh_ has joined #openstack21:43
*** ArdRigh has quit IRC21:43
*** msinhore has quit IRC21:44
*** ArdRigh_ is now known as ArdRigh21:44
mtaylorredbo: gotcha. uploading now21:46
*** miclorb has joined #openstack21:48
mtaylorredbo: ok - have uploaded new packages for lucid, maverick and natty21:50
mtaylorredbo: once those hit, i'll update hudson and also the debian package for swift to require >= 0.9.1321:50
creihtmtaylor: thanks!21:52
*** jakedahn has left #openstack21:53
creihtnow if only we could get tarmac/hudson to swift/1.1.... :)21:53
*** jakedahn has joined #openstack21:53
*** jakedahn has joined #openstack21:53
*** jdarcy has quit IRC21:56
*** dfg_ has quit IRC21:56
*** Ziad has joined #openstack21:56
*** jbryce has quit IRC22:09
*** pvo has quit IRC22:10
*** khussein has joined #openstack22:10
*** romohr has joined #openstack22:14
*** jkelly has quit IRC22:17
khusseindendrobates: Ziad just told me that we need to target the authn and multi-tenant blueprints. I was wondering what's the name of the target?22:24
dendrobateskhussein: there is no need to target anything yet22:25
khusseindendrobates: OK Cool. So, he just wanted to make sure that you know about these two blueprints and that they are in the current agenda.22:26
dendrobatesthey are22:26
khusseinAwesome, thanks :)22:26
*** gondoi has quit IRC22:28
zaitcevI cannot attach anything to blueprint?22:28
zaitcevDidn't we have an associated pastebin for such cases?22:29
*** miclorb has quit IRC22:30
*** pvo has joined #openstack22:33
*** ChanServ sets mode: +v pvo22:33
*** DubLo7 has quit IRC22:33
dendrobatesthere is a place to link to a wiki22:33
dendrobatesor etherpad22:33
khusseinzaitcev: You can create your specs on or, then link to it in Launchpad.22:34
zaitcevOh, thanks. I used my homepage for now, but there's an issue of lifetime.22:35
* soren calls it a week22:40
*** pvo has quit IRC22:42
*** romohr has quit IRC22:43
*** Podilarius has left #openstack22:43
*** Ziad has quit IRC22:50
redbomtaylor: cool, thanks.22:51
*** blakeyeager_ has joined #openstack22:53
*** blakeyeager has quit IRC22:55
*** dendrobates is now known as dendro-afk22:57
*** blakeyeager_ has quit IRC22:57
*** dragondm has quit IRC22:59
*** dendro-afk is now known as dendrobates23:25
*** joearnold has quit IRC23:42
*** blpiatt has quit IRC23:43
*** DubLo7 has joined #openstack23:46
*** johnpur has quit IRC23:46
*** daleolds has quit IRC23:55

Generated by 2.14.0 by Marius Gedminas - find it at!