Tuesday, 2011-05-03

*** dendro-afk is now known as dendrobates00:11
*** dendrobates is now known as dendro-afk00:30
*** GasbaKid has joined #openstack-meeting00:50
*** adjohn has joined #openstack-meeting00:50
*** ovidwu has joined #openstack-meeting01:07
*** ovidwu has quit IRC01:10
*** ovidwu has joined #openstack-meeting01:15
*** anotherjesse has quit IRC01:24
*** GasbaKid has quit IRC01:30
*** edconzel has joined #openstack-meeting01:41
*** edconzel has quit IRC01:49
*** glenc has joined #openstack-meeting01:54
*** glenc_ has quit IRC01:56
*** edconzel has joined #openstack-meeting01:56
*** anotherjesse has joined #openstack-meeting01:59
*** anotherjesse has quit IRC02:03
*** edconzel has quit IRC04:04
*** GasbaKid has joined #openstack-meeting07:57
*** GasbaKid has quit IRC08:41
*** GasbaKid has joined #openstack-meeting08:48
*** GasbaKid has quit IRC10:46
*** beebrox has joined #openstack-meeting10:48
*** adjohn has quit IRC11:15
*** beebrox has quit IRC11:34
*** adjohn has joined #openstack-meeting12:45
*** dabo has joined #openstack-meeting13:00
*** dendro-afk is now known as dendrobates13:18
*** dprince has joined #openstack-meeting13:22
*** johnpur has joined #openstack-meeting13:46
*** adjohn has quit IRC14:06
*** edconzel has joined #openstack-meeting14:16
*** agarwalla has joined #openstack-meeting14:27
*** agarwalla has left #openstack-meeting14:28
*** jkoelker has joined #openstack-meeting14:46
*** glenc_ has joined #openstack-meeting15:32
*** beebrox has joined #openstack-meeting15:35
*** glenc has quit IRC15:36
*** glenc_ is now known as glenc15:36
*** glenc_ has joined #openstack-meeting15:43
*** glenc has quit IRC15:44
*** glenc_ is now known as glenc15:45
*** dendrobates is now known as dendro-afk15:54
*** troytoma` is now known as troytoman-away16:18
*** mray has joined #openstack-meeting16:27
*** johnpur has quit IRC17:11
*** edconzel has quit IRC17:28
*** beebrox has quit IRC17:32
*** edconzel has joined #openstack-meeting17:34
*** edconzel has quit IRC17:39
*** edconzel has joined #openstack-meeting17:45
*** medberry is now known as med_out17:58
*** dendro-afk is now known as dendrobates17:59
*** med_out is now known as medberry19:29
*** troytoman-away is now known as troytoman19:39
*** Sumit has joined #openstack-meeting19:44
*** carlp_ has joined #openstack-meeting19:45
*** ttx has joined #openstack-meeting19:46
*** cbeck has joined #openstack-meeting19:48
*** danwent has joined #openstack-meeting19:51
*** Ryan_Lane has joined #openstack-meeting19:54
ttxer.. in one hour.20:01
sorenI got the whole add-two thing down pretty good. I just thought the meeting was at 2000 UTC :)20:04
*** markvoelker has joined #openstack-meeting20:09
*** dprince has quit IRC20:17
*** rnirmal has joined #openstack-meeting20:30
*** msivanes1 has joined #openstack-meeting20:37
*** eday has joined #openstack-meeting20:38
*** ramd has joined #openstack-meeting20:44
*** jlm^ has joined #openstack-meeting20:45
*** salv-orlando has joined #openstack-meeting20:49
*** ewindisch has joined #openstack-meeting20:51
*** dendrobates is now known as dendro-afk20:51
*** dendro-afk is now known as dendrobates20:52
*** spectorclan_ has joined #openstack-meeting20:55
*** KnightHacker has joined #openstack-meeting20:55
*** nhm has joined #openstack-meeting20:56
*** bblair48 has joined #openstack-meeting20:57
*** edconzel has quit IRC20:57
*** jdurgin has joined #openstack-meeting20:57
*** dubs has joined #openstack-meeting20:58
*** jk0 has joined #openstack-meeting20:58
*** somikbehera has joined #openstack-meeting20:58
*** User343 has joined #openstack-meeting20:59
*** comstud has joined #openstack-meeting20:59
*** NelsonN has joined #openstack-meeting20:59
*** agarwalla has joined #openstack-meeting21:00
* soren reports for duty21:00
*** vishy has joined #openstack-meeting21:00
*** anotherjesse has joined #openstack-meeting21:01
ttxnotmyname: around ?21:01
*** masumotok has joined #openstack-meeting21:01
notmynamettx: yo21:01
ttxyay, let's start.21:02
openstackMeeting started Tue May  3 21:02:13 2011 UTC.  The chair is ttx. Information about MeetBot at http://wiki.debian.org/MeetBot.21:02
openstackUseful Commands: #action #agreed #help #info #idea #link #topic.21:02
ttxWelcome everyone to our weekly OpenStack team meeting...21:02
ttxToday's agenda:21:02
ttx#link http://wiki.openstack.org/Meetings21:02
*** cynb_ has joined #openstack-meeting21:02
*** Zangetsue has joined #openstack-meeting21:02
ttx#topic Current release stage: Design21:03
*** openstack changes topic to "Current release stage: Design"21:03
ttxWe are just a few days after the design summit, and still digesting all the sessions...21:03
ttxThe PTLs are working on the Diablo plans.21:03
*** jaypipes has joined #openstack-meeting21:03
ttxThe idea is to have blueprints for the significant features we want to have in Diablo, in order to track them.21:03
ttxjaypipes: o/21:03
jaypipesttx: sorry for tardiness...21:04
*** dragondm has joined #openstack-meeting21:04
ttxVish already sent an email to Nova devs for missing assignees.21:04
ttxBy next week we should have a more precise plan.21:04
ttx#topic Default release schedule21:04
*** openstack changes topic to "Default release schedule"21:04
ttxI talked to Vish on Friday and we came up with a default release schedule21:04
ttx#link http://wiki.openstack.org/DiabloReleaseSchedule21:05
*** zns has joined #openstack-meeting21:05
*** sandywalsh has joined #openstack-meeting21:05
ttxOne milestone every 4 weeks, with some extra time for the first, to account for BP work and design summit digestion21:05
ttxLast milestone is skipped since we have a coordinated release instead21:05
ttxWe might add an extra gamma-like milestone towards the end...21:05
ttxUnless they have a good reason not to, projects should try to stick to that default schedule21:06
ttxQuestions ? Remarks ?21:06
NelsonNWhen will be possibl;e to start submitting code?21:06
carlp_Sounds good, looks like we should get some rapid iterative development with this.21:06
ttxNelsonN: it's already possible.21:07
ttxSince April 14.21:07
NelsonNok great21:07
sorenttx: I don't see any freezes or anything on that release schedule.21:07
sorenttx: Do we not do freezes? Just random snapshots of trunk on those days?21:07
ttxsoren: depends on the project. Whenever I have more precision from PTLs I'll add it21:08
vishysoen: the plan we discussed was for the first milestone to do a QA branch a couple of days before21:08
ttxBut we probably won't freeze. Just branch out a QA thing.21:08
vishyuntil automated testing is complete enough that we feel comfortable just tagging the latest stable build21:09
vishysoren: ^^21:09
*** koji-iida has joined #openstack-meeting21:09
ttxsoren: the schedule will get more precise over time. I wanted to set the milestones dates early.21:09
ttxOk, let's move on then21:10
ttx#topic Bits from the PTLs: Nova21:10
*** openstack changes topic to "Bits from the PTLs: Nova"21:10
ttxvishy: anything to announce ?21:10
ttxothers: any question for the Nova PTL ?21:10
ttxvishy: I suspect you're waiting on answers and assignees on your recent ML post ?21:11
jaypipesvishy: excellent post to ML about decisions. thanks for doing that.21:11
vishyttx: no announcements21:11
vishyttx: more emails and blueprints coming soon :)21:11
vishyttx: i didn't request specific assignees in that post21:12
ttx#topic Bits from the PTLs: Glance21:12
*** openstack changes topic to "Bits from the PTLs: Glance"21:12
jaypipesWe're assigning blueprints currently to D1 and D2 milestones. https://launchpad.net/glance/+milestone/diablo-121:12
ttxvishy: ok, only BPs so far.21:12
jaypipesExpect to see some emails about Glance BPs and milestones in a few hours.21:12
ttxjaypipes, vishy: are we standardizing milesones names ? diablo-1 ? diablo1-20110602 ?21:13
jaypipesAnd a ML post about new images API stuff coming in D1.21:13
jaypipesttx: doesn't matter to me...21:13
pvodX seems to make sense.21:13
jaypipesttx: I think it's easier to say D1 than D1-20110602 ;)21:13
ttxjaypipes: I agree, but Vish wanted to add te date in it...21:14
glencit also doesn't look as silly if you slip it by a few days21:14
pvothe milestones already have a date field that is required.21:14
jaypipes++ to pvo and glenc21:14
ttxvishy: ok with that ?21:14
vishyas long as the date exists somewhere easy to find21:14
vishyi don't think it needs to be in the name21:15
jaypipesvishy: it's attached to the milestone...21:15
ttxor on the release schedule.21:15
vishythat is fine with me21:15
ttxok, thx21:15
ttxanything else on Glance ?21:15
jaypipesttx: eh, not right now... I have some stuff to talk about re: Keystone, though...21:15
jaypipesKnightHacker: that you, Khaled?21:16
jaypipeswhois confirms...21:16
KnightHackerjaypipes: Yes.21:16
anotherjessejaypipes: have you had a chance to play with tests for keystone?21:16
jaypipesanotherjesse: yes.21:16
jaypipesanotherjesse, ttx, KnightHacker: can I propose a topic for later on for Keystone?21:16
ttx#topic Bits from the PTLs: Swift21:17
*** openstack changes topic to "Bits from the PTLs: Swift"21:17
ttxnotmyname: something on your mind ?21:17
notmynameI don't have anything yet. just been working a little on the github account21:17
notmynamenothing for swift yet21:17
ttxAny question for notmyname ?21:17
ttxok then...21:18
ttx#topic Feedback from the design summit21:18
*** openstack changes topic to "Feedback from the design summit"21:18
ttxnotmyname: fire21:18
notmynamewell, since you were talking milestones with the other projects, our milestones will be (I think) version numbers21:19
notmynamelike 1.3.X21:19
notmynamejust FYI21:19
ttxhmm... and you turn 1.3.8 into 1.4.0 ? sounds weird to me21:19
vishynotmyname: are you doing 4 week milestones as well?  Or is it a "when it is ready" point release?21:20
notmynameI think it will be more of a "when it's ready", but we will target something roughly monthly21:20
notmynamettx: I'm not sure that diablo == 1.4.021:20
ttxnotmyname: oh, I see.21:20
notmynameI'd like to discuss more with the other devs on that, but we aren't making fundamental changes to the architecture at this point21:21
notmynameat least, not until pluggable rings are imported :-)21:21
jaypipesnotmyname: so you're saying that instead of milestones named D1, you'll have milestones named, say, "1.3.6"?21:21
notmynamejaypipes: yes21:21
jaypipesnotmyname: k, understood...21:21
ttxnotmyname: ok, keep us posted on further progress.21:22
notmynamewill do21:22
ttxreally moving to next topic now21:22
ttx#topic Feedback from the design summit21:22
*** openstack changes topic to "Feedback from the design summit"21:22
ttxFor those who were present but missed the postmortem session at the end...21:22
ttxAnything on your mind ? What went well, what went wrong ?21:22
znsIs there a summary somewhere of the postmortem?21:23
jlm^Shouldn't have multiple differing copies of agenda.21:23
pvomobile scheduling tool was nice.21:23
anotherjessettx: it might be nice to have a clearer separation of the summit and conference part21:23
ttxanotherjesse: We might place the design summit *before* the conference next time21:23
anotherjesseit can be confusing to folks to have them overlap21:23
spectorclan_anotherjesse: we plan to market as independent events next time21:23
ttxanotherjesse: I'm fully with you. The business people like to have it overlap though, so make sure you give then that feedback21:23
dabosometimes seating and/or people made it hard to hear the discussion21:23
ttxdabo: I'd blame rooms, that were too large for our need21:24
bblair48Hard to call them design discussions with 150 people in the room.21:24
notmyname+1 to mobile scheduling21:24
anotherjesseI think the unconference sessions were nice - would have liked to have a day just for those topics21:24
znsMore whiteboards21:24
anotherjesse(that folks want to talk about - not overlapping with summit)21:24
dabo+1 on whiteboards21:24
jaypipeszns: ++21:24
spectorclan_Got it on the whiteboards21:24
* jaypipes would like smaller rooms21:24
anotherjessemaybe we overlap the unconference and conference -- not the summit & conference :)21:24
* ttx would like projectors and screens up until the end21:25
spectorclan_ttx: that was a scheduling issue on our end, we thought 3:30 would be it. my mistake21:25
jaypipesif the design summit were first, then last DS day mixed with first conference day, then decisions made in the DS could be explained to business folks better..21:25
spectorclan_jaypipes: that is what I proposed today21:25
ttxjaypipes: yes, that's the rationale.21:25
spectorclan_design summit (M-W) Conf (W-R)21:26
jlm^Organizing which sessions -- can this happen on the open list?  There were "by invite" sessions.21:26
ewindischhaving a mobile site was nice, but finding it was relatively difficult.21:26
jaypipesjlm^: which ones were " by invite"?21:26
anotherjessethe pre-conference naas ones?21:26
dabono overlapping teenybopper events next time!21:26
znsMobile sucked on BlackBerry. (I know BlackBerry sucks, but gotta support em… for a couple more years at least).21:26
ttxdabo: +121:26
jaypipesdabo: lol21:26
jlm^The networking ones before Thu.21:26
*** jamesurquhart has joined #openstack-meeting21:26
spectorclan_zns:we are aware of the blackberry issue21:27
znsOr provide iPhones ;-)21:27
spectorclan_did everyone take the online survey?21:27
znsNo. Link to survey?21:27
jaypipesjlm^: no design summit discussion should have been "by invite"21:27
jlm^I missed (didn't hear about) the Mon one, someone forwarded me a msg about the Tue one.21:27
ttxjlm^: I think those weren't by invite, they were only "organized at the very last moment"21:28
ttxjlm^: but we'll make sure to avoid that next time...21:28
spectorclan_I am not aware of the NAAS stuff on Monday - do we know who planned it?21:28
troytomanjlm^: sorry. we just pulled together the authors of the various naas blueprints. not meant to be invite only21:28
jlm^The organizing msgs didn't go to an open list, they went to a list of invitees.21:28
ttxspectorclan: Erik Carlin, I think21:29
spectorclan_ok. I will make sure all things come thru me so I can keep track21:29
spectorclan_and make sure the items are public21:29
ttxany other quick feedback before we switch to Keystone ?21:29
jlm^Also, they weren't on the schedules.21:29
ttxjlm^: they weren't part of the design summit.21:29
ttxjlm^: until Thursday...21:29
spectorclan_BTW, looks like Boston in first week of October - starting my search this week for an event location; open to all ideas21:30
znsEverything in October should be in Munich...21:30
vishyspectorclan_: +1 for summit in Hawaii21:30
jk0what happened to Europe?21:30
anotherjessevishy: +1 for hawaii21:30
salv-orlandoI thought the design summit was coming to Europe :-)21:30
ttxjk0: +121:30
spectorclan_vishy: Hawaii is not a state so no can do21:30
dendrobatesboston is horrible.  We should not make all the Japanese devs travel here again.21:30
spectorclan_jk0: Europe or Asia is target for 2012 Fall event21:31
dendrobatesspectorclan: why?21:31
spectorclan_Not enough time to plan an event in Europe this year21:31
spectorclan_Event team at Rackspace wants more time to put it together21:31
spectorclan_I am looking to setup mini summits in Japan this year for developers and possibly europe as well21:32
notmynameall I ask is that the date and place be firm as soon as possible for those of us who will try to plan travel around it. airfare and hotels will be cheaper the sooner they are booked21:32
dendrobatesI call BS on that.  six months is more than enough time.21:32
spectorclan_notmyname: the plan is to finalize the location and dates this month.21:32
pvospectorclan_: pssst. Hawaii is a state.21:32
spectorclan_dendrobates: they are part of the decision process for helping to plan and finance the event21:32
alekibangottx: ... a bit late comment to release schedules:  i would love shorter iterations, like in agile programming -- backed by lots of automated tests21:32
ttxok, we'll move the rest of this discussion to open discussion21:33
comstudjust curious: which entity is responsible for making the decision?21:33
comstud(of where, when, etc)21:33
dendrobatesdo we need someone other than Rackspace to put it on?21:33
ttxalekibango: whenever we have automated tests up to snuff maybe21:33
anotherjessespectorclan_: hawaii became a state recently - 195921:33
*** bblair48 has quit IRC21:33
spectorclan_I just saw on Fox they are not a state???21:33
dabojust ask the donald21:33
comstudI was going to comment on the Hawaii thing, too, but I assume this was a joke21:33
ttxDon't ask Fox anything.21:33
alekibangottx: i think i can help with this if someone will put few computers and configs in21:33
spectorclan_Event location discussion - we can have another talk later on this and the thinking of the current plan21:33
termiethere are conference facilities in hawaii, this is not an unusual activity for them21:34
ttxalekibango: it's being worked on. It's the key to everything21:34
alekibangoi agree21:34
spectorclan_Hawaii is VERY EXPENSIVE21:34
alekibangobugs are really tough21:34
alekibangoeven for old hackers21:34
ttx-- STOP -- please complain about location in 10min21:34
vishyspectorclan_: true, but it is cheaper for the japanese...21:34
vishyttx; ok21:34
ttx#topic Keystone21:34
*** openstack changes topic to "Keystone"21:34
ttxwant to squeeze the keystone topic requested before we fire location ideas21:34
anotherjesseJay - want to give an update on testing?21:35
ttxjaypipes: your topic21:35
jaypipeszns, KnightHacker: OK, so I was wondering if you guys would be opposed to replacing bottle with how Nova/Glance/Swift do WSGI app loading?21:35
KnightHackerI think that makes sense.21:35
znsNot at all. Intending to. In fact, there is a nobottle branch. Just haven't got to it yet.21:35
KnightHackerWe only used bottle because it was a little faster for us to prototype things21:35
jaypipesok, no worries... will the nobottle branch be merged any time soon?21:36
*** Ryan_Lane has quit IRC21:36
* jaypipes was trying to write integration tests with Glance and found it odd that identity.py used a different load process than echo/echo/echo.py ;)21:36
KnightHackerThis branch was created a while back and it'll probably need to be re-created.21:36
KnightHackerI mean, we can make that a priority I guess.21:36
anotherjesseKnightHacker: should we just convert it over with a new branch?21:37
KnightHackerI don't think that it is a big deal.21:37
znsNo date. I don't know the state of the branch. Might be too far back. We are meeting internally at Rackspace tomorrow to plan what to code next. We can bring nobottle up.21:37
jaypipesKnightHacker, zns: would be cool to make that a priority. just because that would make starting/stopping servers in integration tests much easier/more standardized...21:37
jaypipesKnightHacker, zns: ok, sounds good. can you invite me to that meeting?21:37
ttx#action KnightHacker zns to get the nobottle branch merged asap21:38
ttxjaypipes: anything else ?21:38
jaypipesawesome. ok, that's all I wanted to bring up right now. thx.21:38
ttx#topic Open discussion21:38
*** openstack changes topic to "Open discussion"21:38
ttxFWIW I'm taking a few days vacation starting tomorrow, and I'm at a conference next week. Should read email though.21:38
ttxOK, location again21:39
jaypipesanything but Santa Clara is fine by me :)21:39
ttxI think we should choose a place that has some relatively-local openstack developers21:39
jk0+1 for somewhere other than Boston21:39
dendrobatesin 20 minutes we will be having a meeting to discuss the Network project21:39
comstudjay, lol21:39
ttxEurope or Japan makes sense... Hawaii or Boston less so21:39
salv-orlandodendrobates: on this channel?21:39
dendrobatesbasically logistics stuff21:39
dendrobatessalv-orlando: yes21:39
termiehawaii was a compromise on japan21:39
ttxthough I won't oppose Hawaii. Or Reunion Island.21:39
sandywalshNova Scotia ;)21:39
sandywalsh(hell no)21:40
vishyI have another discussion topic to bring up when the location argument is done21:40
dendrobatesWe need to get the LP stuff setup and organize teams21:40
comstud+1 :)21:40
spectorclan_vishy: go ahead with your discussion - I will send an email on Deve mailing list21:40
glencIceland - very inexpensive these days21:40
*** msivanes1 has quit IRC21:40
vishyso forums.openstack.org21:40
sorenAh, that thing.21:41
vishyseems a little odd to me21:41
*** midodan has joined #openstack-meeting21:41
ttxvishy: don't tell me21:41
termiei'd like it gone21:41
dendrobatesthat it is already up?21:41
vishythe overwhelming majority on the list seemed to be a stackexchange style site21:41
vishyyet we have forums instead21:41
spectorclan_jordan rinke put it up and asked me to blog and promote21:41
ttxvishy: JordanRinke got fed up by the discussion less than 24 hours after starting it and "just made it"21:41
antonymwould it make sense to split the lists and irc channels by projects?  they seem to be getting pretty busy21:41
alekibangoi think more documentation and complete config examples should be written on sane nova network architectures...  like few examples here:  http://stackops.org/display/documentation/Standard+Multinode+deployment       Having more and well documented examples with explanation is what is really missing in nova manuals.  But it needs help of those people who really understand nova well.21:41
notmynameantonym: +121:42
dendrobatesforums are a lot of work if you don't want them to suck21:42
vishyttx: I got that, but if it is under openstack.org it seems "official" and i don't think there was consensus on forums being the right idea, seems like it should at least go through the PPB...21:42
dendrobatesantonym: I suggest we just move off the noisy projects21:42
jlm^antonym: Agreed.  The unified ML is too overloaded.21:43
dendrobatesonce you get busy enough you get your own list21:43
jamesurquhartdendrobates: +121:43
ttxvishy: fully agree21:43
dendrobateswe don't want to completely fragment21:43
antonymwell #openstack is a big question and answer room in addition to development, if we seperated dev for nova and swift, it might make a little more sense21:43
vishymaybe we just put it on the PPB agenda as part of our figure out the various things that the ppb should be in charge of21:43
pvovishy: I agree that it would give the impression that its fully supported and staffed.21:44
sorenvishy: +1. I think that's really more of an issue that anything else.21:44
alekibango+1 for separation of irc21:44
sorenvishy: Forums being what we want or not, the fact that it just magically appeared in the openstack.org namespace is problement.21:44
*** Ryan_Lane has joined #openstack-meeting21:44
sorenproblematic. "problement"?!?21:45
vishysoren: agreed, this is just what we discussed at the ppb last week.  I'll put it on the agenda21:45
sorenbutter fingers.21:45
sorenI don't speak French.21:45
znsYou just did!21:45
sorenNot on purpose, at least.21:45
*** koji-iida has quit IRC21:46
ttxok, anything else ?21:46
antonymi have #openstack-swift and #openstack-nova reg'd from when i set the channels up a while back if we did want to seperate those out21:46
dendrobatesantonym: how about just #openstack-devel21:47
jlm^Now that we have more projects, it makes sense.21:47
antonymis there a ton of nova and swift talk that overlaps?21:47
jaypipesnot really...21:47
*** NelsonN has left #openstack-meeting21:47
*** koji-iida has joined #openstack-meeting21:47
ttxI haven't witnessed that much overlap21:47
dendrobatesI like seeing all the swift stuff, though21:48
dendrobatesbut perhaps it's just me.21:48
* jaypipes would prefer a single channel and mailing list still... I still don't see it being a huge issue.21:48
antonymyou can sit in all the channels for that :)21:48
jlm^If you like seeing Swift, sign up for it.  :)21:48
ttxantonym: could you prepare a plan, we'll had a topic for the next week meeting and discuss it21:49
ttxI don't want to rush anything in that area21:49
*** spectorclan_ has quit IRC21:49
ttxor maybe raise a thread to take options21:49
ttxBack to preferred topic: What happened of Brussels ?21:49
notmynamefrom the swift perspective, it's hard to keep up with questions that are asked if they aren't specifically addressed to a swift dev. the other stuff (generally nova stuff) tends to drown it out21:49
antonymttx: sure, i can make a mailin list topic if that's cool21:49
ttxantonym: yes, with a basic proposal, should spawn a bit of discussion21:50
antonymsounds good21:50
ttx#action antonym to raise a thread about IRC separation on the ML21:50
ttxOK, any other topic ?21:51
ttxI guess not.21:52
ewindischI like the idea of Brussels. New York for the US, or for Asia - Japan, Seoul, or Singapore.21:52
*** openstack changes topic to "Openstack Meetings: http://wiki.openstack.org/Meetings | Minutes: http://eavesdrop.openstack.org/meetings/openstack-meeting/2011/"21:52
openstackMeeting ended Tue May  3 21:52:33 2011 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)21:52
openstackMinutes:        http://eavesdrop.openstack.org/meetings/openstack-meeting/2011/openstack-meeting.2011-05-03-21.02.html21:52
openstackMinutes (text): http://eavesdrop.openstack.org/meetings/openstack-meeting/2011/openstack-meeting.2011-05-03-21.02.txt21:52
openstackLog:            http://eavesdrop.openstack.org/meetings/openstack-meeting/2011/openstack-meeting.2011-05-03-21.02.log.html21:52
ttxFeel free to continue to vote for Hawaii, Europe or other.21:52
*** koji-iida has quit IRC21:53
dendrobates8 minutes until the Network organization meeting21:53
*** nhm has left #openstack-meeting21:53
danwentcan't wait :)21:53
*** comstud has left #openstack-meeting21:53
eday+1 for boston, I need to be there anyway in the fall :)21:53
dendrobatesdanwent:  it should be pretty fast.21:53
jk0so I take it Amsterdam is no longer an option? :)21:53
ttxjk0: heh21:54
jk0you know you were all thinking about it21:54
danwentdendrobates: np.  glad to see our timing worked out perfectly21:54
ewindischjk0, day trip from Brussels, no?21:54
jk0fair enough21:55
*** adjohn has joined #openstack-meeting21:55
*** AlexNeef has joined #openstack-meeting21:55
*** msivanes has joined #openstack-meeting21:56
*** jk0 has left #openstack-meeting21:58
*** KnightHacker has left #openstack-meeting21:59
*** anotherjesse has quit IRC21:59
*** koji-iida has joined #openstack-meeting21:59
*** zns has left #openstack-meeting21:59
*** romain_lenglet has joined #openstack-meeting21:59
*** anotherjesse has joined #openstack-meeting21:59
danwenthello networking folks :)22:00
carlp_hey Dan!22:00
salv-orlandohello dan22:00
Zangetsuehi all22:00
markvoelkerAloha (oh wait, did we not agree on Hawaii?  Darn....)22:00
dendrobatesI just have a few question to run by everyone, so it should be quick22:00
danwentprobably best if everyone gives a shout-out, as there are many non-network folks lurking still.22:00
somikbeheraHellp All22:00
danwentI'm all for hawaii.22:00
danwentnetwork summit ;P22:00
jlm^Yo yo network dudes.22:00
carlp_Wasn't the IETF summit in Hawaii where IPv6 was developed?22:01
*** ryu25 has joined #openstack-meeting22:01
carlp_oh so many years ago22:01
danwentcarlp_ : that would explain a lot22:01
salv-orlandoor the ALOHA protocol?22:01
dendrobatesis there any reason to use meetbot?22:01
danwentrick, your call.22:01
*** ryu25 has quit IRC22:02
adjohndendrobates: for logging?22:02
*** anotherjesse has quit IRC22:02
dendrobatesadjohn: yes22:02
dendrobatesI'll go ahead22:02
openstackMeeting started Tue May  3 22:02:41 2011 UTC.  The chair is dendrobates. Information about MeetBot at http://wiki.debian.org/MeetBot.22:02
somikbeheradendrobates: it would be good to have meeting minutes22:02
openstackUseful Commands: #action #agreed #help #info #idea #link #topic.22:02
*** ryu_ishimoto has joined #openstack-meeting22:02
dendrobates#topic LP project placement22:03
*** openstack changes topic to "LP project placement"22:03
dendrobatesOk, as most of you know I created the Network-service project on Launchpad22:03
dendrobatesMy idea was to make it a project group that all of us could own and create subprojects below it22:04
dendrobatesI thin there is value to us staying as a group during the early stages22:04
dendrobatesbut there is also the option of joining futurestack22:04
dendrobatesall the incubation projects22:04
jlm^What's futurestack?22:05
ramdIs network-service incubation project?22:05
danwentare they mutually exclusive?22:05
dendrobateser, I mean pre incubation projects22:05
troytomanI was asked if we could keep the IPAM work in Nova, any objection for that component?22:05
dendrobatesramd: no it isn't22:05
somikbeheraCan Network-service be in futurestack?22:05
dendrobatessomikbehera: no, LP cannot do nested groups22:05
AlexNeefI think that network will likely have a lot of touch points with other projects22:05
AlexNeefwould future stack give our work better exposure to other groups?22:06
carlp_AlexNeef: agreed22:06
ramdQuantuma nd Donabe is a go for Diablo, right?22:06
markvoelkerjlm^: https://launchpad.net/futurestack22:06
dendrobatesramd: they are not official openstack projects yet22:06
salv-orlandoI think melange as well is lined up for diablo22:06
carlp_Melange is a lined up for Diablo as well22:06
jlm^Melange == IPAM, right?22:07
adjohntroytoman: wouldn't IPAM be useful as a standalone project? Or are there reasons for that being in nova?22:07
carlp_jlm^: yes22:07
troytomanjlm^: yes22:07
dendrobateswe need to be careful to follow the process correctly22:07
dendrobatesthere has been some push back from other projects not doing it correctly and I want to avoid that22:07
troytomanadjohn: it would still be a separate service(s) but anotherjesse wanted to see it worked on within the nova project22:08
romain_lengletso, what's the timeframe for the nova-related stuff: ipam and network refactoring?22:08
romain_lengletstill diablo?22:08
dendrobatestroytoman: will it be a separate branch?22:08
danwentdendrobates: if we were in futurestack, is the idea that we would have separate projects within futurestack for quantum and donabe that would appear at the same level as something like atlas?22:08
troytomandendrobates: working out that detail (repo, branch) etc.22:08
dendrobatesdanwent: yes22:08
dendrobatesromain_lenglet: since Diablo was extended to 6 months,  I would hope so.22:09
troytomanromain_lenglet: yes. we probably need to land the refactoring in an early milestone thought22:09
danwentI probably have a slight preference for using futurestack if it makes it easier for other people to see what we are doing.22:09
uvirtbottroytoman: Error: "though" is not a valid command.22:09
dendrobatesdanwent: I am not sure that it does22:09
dendrobatesand we can share some resources between us if we use our own project group22:10
salv-orlandoif melange stays is nova, I think there is no much sense in having Network-Service as a project group22:10
danwentdendrobates: I don't feel strongly here.22:10
salv-orlandoas melange will not be in it22:10
adjohnIt would also be a better fit for donabe, as it will include non-networking stuff eventually22:11
danwentI see, does future stack mean we wouldn't have our own bug system, etc?22:11
AlexNeefI would imagine melange will have a lot of interaction with quantum right?22:11
dendrobatessalv-orlando: that is a good point, but I don;t want to cede IPAM without discussing it more22:11
adjohndanwent: each repo has it's own system22:11
AlexNeefas would donabe - we really want to keep the source close22:11
salv-orlandodendrobates: agreed22:11
dendrobatesquantum and donabe will be tightly coupled22:11
danwentadjohn: ok, thanks.22:11
dendrobatesanyway, we are not locked in to any decision22:12
markvoelkerdanwent dendrobates: Agreed....Quantum/Donabe/etc are more inter-related than, say, Atlas and RedDwarf.  Might make sense to have them in a separate "container" rather than added in to FutureStack.  But I also don't feel strongly.22:12
carlp_To be honest, even though Melange won't be in it I like the idea of having a rallying point for all network stuff.  This should make it easier for others to contribute ideas without everyone working in the dark.22:12
salv-orlandocarlp_: +122:12
ramdHaving a network-service help as we add more network related services22:12
AlexNeefcarlp_: +122:12
dendrobatesI expect the networking services to be added all the time22:12
troytomanI think there is value in keeping a network focused project22:12
markvoelkerdendrobates: +122:12
troytomanthere is a lot to sort out there and a focused group is important. look what we were able to do in 1 week22:13
danwentsounds like consensus for a separate project?22:13
dendrobatesok, it seems we have a consensus, anyone strongly disagree22:13
AlexNeefcan you sumarize the consensus22:13
somikbeheraI agree on a focussed group. +122:13
dendrobatesThat we will have a project group named Network services where we can put network services22:13
danwentproposed consensus is that we will have a separate lp 'network-service' group, as rick already created, rather than use futurestack.22:14
AlexNeefgreat +122:14
dendrobatesI created it as a project, the LP admins will have to change it to a group22:14
danwentgot it.22:14
AlexNeefsorry if this is basic, but can you explain difference between project and group22:14
danwentI have much to learn about launchpad, i see :)22:14
*** mray has quit IRC22:15
danwentAlexNeef : +122:15
adjohnAlexNeef: groups can have sub projects22:15
dendrobates#action change network services to a group and create projects under it22:15
dendrobates#topic admin team22:15
*** openstack changes topic to "admin team"22:15
adjohnopenstack is a group for example, nova is a project22:15
dendrobateswe need to create a basic team to own the resources22:15
dendrobatesI suggest we try to have one person from each group involved on the team22:16
danwent"group" ?22:16
somikbeheraAre all others ready to seed a team today or we should wait till next week.22:16
romain_lengletdendrobates: are you suggesting one team for the whole network service project?22:17
dendrobatesthis is not for development, just for ownership of the LP resources22:17
ramdI volunteer BTW what is the admin role22:17
romain_lengletI volunteer too22:17
dendrobatesromain_lenglet: we can have many teams22:17
dendrobatesthis team is mostly about Launchpad maintainance22:18
danwentcan we talk more about what resources need to be owned?22:18
somikbeherawhat are the responsibilities of this group?22:18
dendrobatesthe admins can edit the text of the project group, add new projects, and maybe a few other things22:18
dendrobatesWe will also need to create drivers22:19
danwentis that a launchpad term?22:19
dendrobatesdrivers are responsible for approving blueprints and setting milestones and releases22:19
somikbehera you mean project drivers not the "drivers" ;)22:19
salv-orlandodo we need a "network-service-core" team as well?22:20
AlexNeefi see in futurestack it's the same group22:20
dendrobatesit can be if we want22:20
AlexNeefand the group can be as large as we want or there is a limitation?22:20
dendrobateswe can also decide if we want separate or combined dev teams22:20
dendrobatesit can be as large as we want22:21
danwentWhat I want to avoid is putting together some kind of structure just based on people volunteering vs. people contributing code :)22:21
dendrobateswe can also just let openstack-admins own our project group22:21
danwentI think we have a rough outline of the blueprints we want to do from the summit and hopefully we can work together to refine the details.22:21
dendrobatesdanwent: I agree22:22
dendrobateswe do have some things to setup, that have to have owners and maintainers22:22
dendrobatesI want it to be owned by the group and not a single individual22:22
ramddendrobates: agree22:23
dendrobateswe can always add and subtract from the group later22:23
dendrobatesI probably have the most LP knowledge because  isetup a lot of the openstack groups and process.22:23
danwentOk.  I prefer to keep things as "flat" as possible for now.  I think as people contribute real time and energy, work with each other and gain respect, etc. leadership will naturally emerge.22:24
salv-orlandoand in your knowledge we trust :-)22:24
danwentI'm fine with rick running things for LP for now.... he's a good guy :)22:24
somikbehera+1 ;)22:24
ramdsalv-orlando: +122:24
dendrobatesI would in the least like to create an admin team and add myself and Dan.22:25
danwentOr not me...  I don't want this to look like anyone is "grabbing" anything.22:25
dendrobatesJust in case I get hit by a bus22:25
danwentRick at least has a history in the community.22:25
AlexNeefI think you should add dan and others that are leading the sub-projects22:26
troytomandanwent: more like volunteering you for more work ;)22:26
somikbeherawe do need Fault tolerance22:26
danwentI'm fine either way.22:26
dendrobatesit is both fault tolerance and volunteering. :)22:26
salv-orlandosomikbehera: for that we would need an admin in each continent to have 24/7 presence!22:26
danwentok.  i need to run to another meeting soon.  other topics?22:27
dendrobates#topic next meeting22:27
danwentjust want to give people a heads up that I have started adding blueprints along the lines of what we outlined on friday.22:27
*** openstack changes topic to "next meeting"22:27
dendrobatesi think we should come to the next meeting prepared to discuss dev process and assignments22:27
danwentmy goal is to have drafts up in advance of the next meeting so people can voice concerns and talk about what they want to work on.22:27
carlp_sounds good22:28
danwentdefinitely want to keep the momentum going22:28
dendrobateshas everyone seen the wiki page I created?22:28
troytomanI have also started creating blueprints for melange and updating the the wiki with links22:28
ramdAdding few more BP palceholders as we speak22:28
dendrobateswe can start adding info to the wiki22:29
dendrobatesand there is a header with a link to a meeting agenda that we can all modify22:29
dendrobatesI will start all the LP magic.22:30
dendrobatesPlease use the wiki as much as possible22:30
somikbeherathanks dendrobates - the wiki is a good start to get everything ready for next week.22:30
salv-orlandodan, which are the names for the blueprints you have registered? I can find the melange ones, but not the quantum blueprints22:30
dendrobatesWe can let danwent get to his meeting now.22:30
dendrobatesThanks for coming everyone, I am very excited to get to work.22:30
danwentdo you guys see blueprints here: https://blueprints.launchpad.net/network-service22:31
danwentperhaps its a permissions thing?  Or i am putting them in the wrong place?22:31
carlp_danwent: yep22:31
dendrobatesI see them22:31
danwentis an example name22:31
danwentnot done yet.... was just creating some during the main OS meeting :)22:31
ramdI see them22:32
dendrobatesthat is the link to add a blueprint to network-service22:32
danwentok.  thanks guys.  need to run.  talk to you next week.22:32
*** openstack changes topic to "Openstack Meetings: http://wiki.openstack.org/Meetings | Minutes: http://eavesdrop.openstack.org/meetings/openstack-meeting/2011/"22:32
openstackMeeting ended Tue May  3 22:32:29 2011 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)22:32
openstackMinutes:        http://eavesdrop.openstack.org/meetings/openstack-meeting/2011/openstack-meeting.2011-05-03-22.02.html22:32
openstackMinutes (text): http://eavesdrop.openstack.org/meetings/openstack-meeting/2011/openstack-meeting.2011-05-03-22.02.txt22:32
openstackLog:            http://eavesdrop.openstack.org/meetings/openstack-meeting/2011/openstack-meeting.2011-05-03-22.02.log.html22:32
dendrobatesthanks again everyone22:32
salv-orlandothanks and have a nice day/evening/night22:32
*** koji-iida has quit IRC22:33
*** ramd has left #openstack-meeting22:33
*** msivanes has left #openstack-meeting22:33
*** markvoelker has left #openstack-meeting22:33
*** dubs has left #openstack-meeting22:33
*** jlm^ has left #openstack-meeting22:33
*** romain_lenglet has left #openstack-meeting22:34
*** midodan has left #openstack-meeting22:34
*** masumotok has quit IRC22:34
*** jdurgin has left #openstack-meeting22:34
*** User343 has quit IRC22:35
*** danwent has left #openstack-meeting22:35
*** ryu_ishimoto has left #openstack-meeting22:37
*** troytoman is now known as troytoman-away22:40
*** Zangetsue has left #openstack-meeting22:40
*** AlexNeef has quit IRC22:42
*** rnirmal has quit IRC22:45
*** adjohn has quit IRC22:53
*** dabo has left #openstack-meeting23:11
*** jkoelker has quit IRC23:23
*** medberry is now known as med_out23:36

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