18:01:16 <hub_cap> #startmeeting trove 18:01:18 <openstack> Meeting started Wed Mar 26 18:01:16 2014 UTC and is due to finish in 60 minutes. The chair is hub_cap. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:01:19 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 18:01:21 <vipul> o/ 18:01:22 <openstack> The meeting name has been set to 'trove' 18:01:24 <grapex> o/ 18:01:25 <kevinconway> \o/ 18:01:35 <robertmyers> o/ 18:01:35 <juice> o/ 18:01:35 <SlickNik> o/ 18:01:48 <peterstac> o/ 18:01:52 <doug_shelley66> o/ 18:01:56 <cp16net> (ʘ෴̴͜ʘ) 18:02:02 <hub_cap> umm 18:02:03 <vipul> woah 18:02:03 <SnowDust> <o> 18:02:08 <cp16net> lol 18:02:11 <amcrn> o/ 18:02:14 <juice> cp16net: I know you have been waiting all week for this 18:02:15 <amrith> in some language that must look pretty 18:02:18 <amytron> o/ 18:02:28 <juice> cp16net: you probably got up early to rehearse: 18:02:37 <hub_cap> lol juice 18:02:40 <SnowDust> juice: LOL 18:02:40 <amytron> juice: how did you know?! 18:02:50 <amcrn> the early bird gets the worm juice, don't be jealous ;) 18:02:55 <hub_cap> #topic openstack summit 18:03:02 <grapex> juice: What's interesting is cp16net in real life has no mustache 18:03:04 <hub_cap> so, we only have 1 session proposed 18:03:07 <esmute> o/ 18:03:12 <vgnbkr> o/ 18:03:39 <kevinconway> hub_cap: we could probably fill several days of the conference with users 18:03:40 <hub_cap> lets rally and get some sessions pushed up 18:03:44 <hub_cap> LOOL kevinconway 18:03:46 <cp16net> hah 18:03:54 <vipul> what's the deadline? 18:04:07 <cp16net> let do eet 18:04:09 <annashen_> o/ 18:04:26 <hub_cap> vipul: good question, idont know 18:04:28 <SlickNik> I'm looking at pushing up info for a couple of sessions by the end of this week. 18:04:28 <hub_cap> les say today 18:04:29 <hub_cap> ;) 18:04:35 <vipul> :p 18:04:40 <hub_cap> ok lets try to push someting, anything, by EOW 18:04:49 <cp16net> word 18:04:58 <hub_cap> remember, anyone can propose 18:05:38 <hub_cap> and u can update it once u propose 18:05:46 <hub_cap> i wont start scheduling for about 2 more wks 18:05:54 <hub_cap> so i guess ill speak about the process a bit 18:05:58 <hub_cap> just so we are aware 18:06:11 <hub_cap> so anyone proposes to summit.o.o 18:06:24 <hub_cap> and i take those and i have the abilty to schedule them, regject them, mark as dups etc 18:06:41 <hub_cap> i rely on the core team to help 18:07:01 <hub_cap> so we can all get it iron outdddddddddddddd 18:07:08 <hub_cap> ^ ^ joke for grapex 18:07:21 <hub_cap> any questions on the process? 18:07:42 <doug_shelley66> i assume the actual time/room scheduling is done by the conf organizers? 18:07:59 <hub_cap> so the blocks of time are done by ttx 18:08:03 <doug_shelley66> so could be scattered anywhere across the 4 days 18:08:14 <hub_cap> and he consults the ptls to talk about the room blocks and how they coincide 18:08:17 <hub_cap> and if there are conflicts 18:08:23 <amytron> if a session is marked as a dup, do the two proposers need to pair up? 18:08:25 <doug_shelley66> ah ok 18:08:27 <hub_cap> but once we have a "block" we are in the same room 18:08:30 <vipul> do we get > 1 day now? 18:08:30 <hub_cap> amytron: if they wish :) 18:08:47 <hub_cap> vipul: nope but we get a full day iirc 18:08:51 <hub_cap> or 2x the ammt of sessions 18:08:52 <hub_cap> as for dups 18:09:00 <hub_cap> the one that is best written up usually "wins" 18:09:11 <hub_cap> but i will consult both authors and get them to work together 18:09:47 <hub_cap> these are design sessions, so both authors will likely bein the same room anyway 18:10:04 <hub_cap> so doug_shelley66 to reiterate (not sure if i did) we have 1 room the entire time 18:10:22 <vipul> yea and for folks new... these are super informal.. the presenter just starts the discussion 18:10:22 <doug_shelley66> hub_cap from tues to friday? or just 1 day? 18:10:28 <vipul> doesn't ahve to have everything prepared 18:10:30 <hub_cap> doug_shelley66: 1 day, back to back to back to back 18:10:45 <hub_cap> yup vipul, we end up having an etherpad / blueprint 18:10:48 <hub_cap> and then discuss it 18:10:52 <grapex> hub_cap: If all of ours end up being on the same day then I submit "Trove: Nap time" to be preceded by juice boxes. 18:10:53 <juice> hub_cap: will the revolution be televised? 18:10:56 <hub_cap> and this time we wont have HELLA jetlag 18:11:09 <juice> or for those that are not there, they just don't participate 18:11:13 <hub_cap> grapex: did u consult juice ? 18:11:25 <hub_cap> juice: ummm i think we are getting better w/ televising 18:11:30 <juice> I know openstack provides video that next day or the day after 18:11:34 <hub_cap> but in the past its been pretty bad heh 18:11:43 <grapex> hub_cap: I actually meant juice which give us each a specially wrapped box with a present he picked out for us. 18:11:52 <hub_cap> grapex: ++ 18:11:58 <kevinconway> so do the proposers need to also be presenters? 18:12:01 <cp16net> how many session will we have? 18:12:07 <amytron> should we plan to stay for a full day on friday? are there usually a lot of sessions planned on the last day? 18:12:36 <hub_cap> amytron: the sessions go till EOD friday 18:12:37 <amrith> hub_cap: since some of us plan to come in late or leave early (not stay the whole week) it would help to know which day was the trove design sessions 18:12:55 <hub_cap> and there is _always_ a possiblity we will have sessions "moved" to friday because of conflicts 18:13:07 <hub_cap> amrith: ill ask ttx if what hes proposes is irond out 18:13:20 <amrith> can we assume that there are no design sessions on Monday and Tuesday? 18:13:30 <amrith> only W, T and F? 18:13:46 <hub_cap> but if say, 2 other projects are conflicting and they need us to switch 18:13:49 <hub_cap> i wont say no 18:13:55 <hub_cap> there was a bit of switching in HK but not a ton 18:14:06 <hub_cap> and i ALWAYS ALWAYS ALWAYS recommend staying the whole time 18:14:15 <vipul> amrith: i think sessions will be T-F 18:14:21 <amrith> ok, any heads up you can provide in this regard would be much appreciated. 18:14:22 <doug_shelley66> but you are saying that when you pick the topics you will only pick enough to fill 1 day? 18:14:23 <hub_cap> yes vipul T-F 18:14:34 <hub_cap> doug_shelley66: i will pick enough to fill our slots we are given 18:14:37 <esmute> T for tuesday or thursday? 18:14:38 <amytron> what time is "EOD" on friday? 18:14:43 <hub_cap> TUE-FRI 18:14:43 <vipul> lol 18:14:44 <amrith> t for thursday 18:14:47 <amrith> oh 18:14:47 <SlickNik> T=Tuesday 18:14:49 <hub_cap> monday is just keyonets 18:14:50 <amrith> t for tuesday 18:14:52 <vipul> good thing you clarified :) 18:14:53 <amrith> ok 18:14:57 <hub_cap> its on the splash page yall 18:14:57 <amrith> good thing you clarified 18:15:06 <vipul> https://www.openstack.org/summit/openstack-summit-atlanta-2014/ 18:15:08 <vipul> scroll down :) 18:15:10 <SlickNik> #link https://www.openstack.org/summit/openstack-summit-atlanta-2014/ 18:15:13 <hub_cap> thx vipul i was just getting that too 18:15:16 <hub_cap> thx SlickNik 18:15:39 <hub_cap> but srsly guys, i ALWAYS recommend staying, but a good bit of people do leave early 18:15:53 <hub_cap> so if u leave, and we have a friday session, then ill be there to hold down the fort 18:16:57 <cp16net> i just make the decisions for everyone 18:17:01 <cp16net> :-P 18:17:05 <cp16net> j/k 18:17:16 <esmute> you guys dont want to miss the party that we throw out anyways 18:17:37 <hub_cap> esmute: plz tell me yer party is not friday LOL 18:18:01 <juice> HP is cutting budget this year so we are trying to minimize attendees to the party 18:18:05 <esmute> throw down* Thanks juice for correcting 18:18:11 <esmute> lol 18:18:15 <hub_cap> juice: WEAK ;) 18:18:20 <juice> just kidding 18:18:35 <amrith> ah, I see. cost cutting, everyone brings their own juice. I'm claiming first dibs on juice ;) 18:18:36 * hub_cap goes to the press w/ juice 's comment 18:18:37 <juice> I'm sure we will have all you can eat chicken and waffles 18:18:49 <hub_cap> amrith: ummmmmmmmmm u can have all dibs on juice 18:18:51 <hub_cap> ;) 18:18:55 <cp16net> thats delish 18:18:56 <doug_shelley66> should i sell my HP stock? 18:19:18 <juice> doug_shelley66: it's skyrocketing 18:19:32 <doug_shelley66> ah because of the cost cutting :) i see 18:19:35 <hub_cap> dude im so having c&w there 18:19:53 <juice> what's the next topic :) 18:19:57 <hub_cap> lol ya 18:20:10 <vipul> lol juice doesn't want to get caught giving insider trading advice :) 18:20:15 <hub_cap> amrith: is up :) 18:20:21 <amrith> ok 18:20:23 <hub_cap> #topic datastore abstraction 18:20:31 <amrith> Two related topics that I want to discuss. 18:20:31 <amrith> One is the blueprint for the proposed change; not for icehouse, likely later. 18:20:31 <amrith> The second is the review for https://review.openstack.org/#/c/82080/ 18:20:31 <amrith> To the blue print. 18:20:40 <amrith> Unix'es have implemented abstractions like /etc/init.d or upstart to address the issue of a simple mechanism to handle service start and stop. It is an abstraction that allows the OS to easily start and stop services, and for the services to implement basic functionality in this regard. 18:20:42 <hub_cap> #link https://blueprints.launchpad.net/trove/+spec/trove-guest-agent-datastore-control 18:20:54 <amrith> yes, that one 18:20:58 <hub_cap> amrith: nothign is for icehouse anymore ;) 18:21:07 <amrith> absolutely 18:21:11 <amrith> says so in the bug and the bp 18:21:26 <amrith> Unfortunately, there are more than one standard. 18:21:26 <amrith> Therefore, a meta service like the guest agent needs to build an abstraction layer as well. 18:21:42 <amrith> The benefits are obvious. 18:21:42 <amrith> We should not have bugs like https://bugs.launchpad.net/trove/+bug/1295313 18:21:42 <amrith> The common abstraction would provide a mechanism in which each data store would describe the right way to start/stop/control itself. 18:21:42 <amrith> Defaults would be provided (use service start) or some such. 18:22:10 <amrith> so I guess my first question to core is this ... 18:22:18 <amrith> I've not done any of the implementation (yet) 18:22:24 <amrith> have some thoughts on the implementation though 18:22:43 <amrith> should this BP move to the BP review meeting which is on Monday or can we discuss now? 18:22:52 <amrith> reason I put it on now is because I have a part 2 to this conversation 18:23:01 <juice> amrith: I think you keep going 18:23:07 <yogesh> amrith: are we talking about an interface which all the guest agents should implement OR are u talking about a base guest agent implementation...? 18:23:30 <amrith> I am talking about an interface that would be part of each guest agent 18:23:45 <amrith> right now, there's some code in operating_system.py which attempts to abstract some of this knowledge 18:24:13 <vipul> so each datastore manager implement start() and stop()? and the logic for each datastore would be different? 18:24:31 <amrith> so each data store will tell how it wants start() and stop() to be done 18:24:37 <juice> amrith: a question off the top of my head is isn't the abstraction in the datastore/manager.py enough 18:24:39 <amrith> it could simply be to say "use service" in some fashion 18:24:50 <amrith> juice, I'll get to that in a second 18:24:52 <grapex> amrith: So this is just helped code to call the unix based stop / start scripts for an implementation if you want 18:24:53 <grapex> ? 18:25:30 <amrith> to vipul's quesiton, a data store that has a simple service command (eg mysql) would just use that but would explicitly state that. 18:25:38 <amrith> for a sevice (say cassandra) it would be the same thing 18:25:52 <amrith> for mongo on the other hand, the data store would provide the commands to execute for start/stop etc., 18:26:14 <amrith> to juice's question: the code in oeprating system.py is called adhoc by some data stores (I believe). 18:26:26 <amrith> and the code attempts to guess based on what it sees on the file system 18:26:34 <grapex> amrith: But what if a Trove operator wants to run MySQL on Windows? 18:26:36 <amrith> it attempts to pick which to do in some order 18:26:41 <grapex> J/k! 18:27:09 <amrith> but if we have a data store (for example) that requires multiple services, the code in oeprating system.py is not capable enough 18:27:28 <amrith> so in effect, we have a situation where each data store has to implement or adopt the code in operating_system.py 18:27:47 <amrith> instead, I'm proposing that in order to avoid bugs like the one amcrn found, that we make this a clear abstraction. 18:27:56 <amrith> to grapex question, yes 18:27:59 <amrith> it may not be scripts 18:28:02 <grapex> amrith: This sounds like it's just adding some helper code different implementations can choose to use. Doesn't seem too controversial. 18:28:03 <amrith> it may be commands that have to be run 18:28:11 <amrith> for example "mongodb --shutdown" 18:28:30 <amrith> grapex, i think the implementation would be along the lines you are proposing 18:28:34 <vipul> +1 i think each manager owning its own stop/start routines is good.. shouldnt' try to shoehorn everything into a single class 18:28:34 <amrith> it is largely helper code 18:28:54 <amrith> and it will allow each data store to pick its own start/stop without necessarily duplicating the code in many places 18:29:22 <amrith> and it would handle things like cleaning up PID files etc., 18:29:29 <amrith> which a simple KILL command doesn't do 18:29:50 <amrith> to the question of windows, sure it is supported, in v16 18:29:51 <amrith> j/k 18:29:52 <SlickNik> amrith: +1 I think this is a good idea. 18:30:09 <amrith> so, my second process question. 18:30:11 <amcrn> +1, on-board. 18:30:13 <amrith> what's the next stop? 18:30:22 <vipul> atlanta 18:30:27 <amrith> I flesh out the implementation details in the BP and come back to all-y'all at Atlanta? 18:30:48 <vipul> j/k I think this is a minor one 18:30:51 <amrith> ok, thx vipul 18:31:01 <amrith> now to the more controversial issue 18:31:02 <SlickNik> amrith: Clarification, does this tackle discovery of the start stop service? (systemd vs. upstart, for example) or is that v2? 18:31:07 <vipul> actually you should just fix it and do it before that 18:31:15 <vipul> if it's possible.. no need to wait for summit 18:31:21 <amrith> one second 18:31:31 <amrith> SlickNick, answer re: discovery 18:31:41 <amrith> I will try and keep that logic around 18:31:46 <amrith> but I'm dubious about it 18:32:14 <amrith> let me think more and come back to you with a cogent argument rather than some rambling 18:32:25 <amrith> OK? 18:32:47 <amrith> vipul: were you talking about the more controversial issue (https://review.openstack.org/#/c/82080/)? 18:33:00 <amrith> this is the other half of amcrn's https://review.openstack.org/#/c/81914/ 18:33:04 <SlickNik> amrith: Sounds good. 18:33:21 <amrith> SlickNick: thx 18:33:29 <vipul> amrith: i guess i don't know why this one is so controversial.. seems like we shoudl be using service stop for c* 18:33:39 <amrith> I don't know why it is that https://review.openstack.org/#/c/81914/ is +1 but https://review.openstack.org/#/c/82080/ is -1. I assume we don't have to rehash the email exchanges re: the MySQL bug that was part of the discussion of amcrn's fix. I was asked to "provide proof" which I did. But the person asking for it is now silent. 18:33:53 <amrith> CORE: What is your guidance on this? 18:34:10 <amrith> could I get a code review either up or down for https://review.openstack.org/#/c/82080/ 18:34:18 <hub_cap> well amrith if it was denis_makogon its cuz he had to leave for the meeting 18:34:24 <kevinconway> i'd prefer a dramatic reading of the email exchanges 18:34:32 <amrith> I can't grok why service start works in ubuntu but service stop doesn't ;) 18:34:36 <grapex> Who here is using Casandra or running it today? 18:34:38 <hub_cap> kill should not be used 18:34:44 <vipul> if you say it works.. i'd +2 it :D 18:34:50 <amrith> kevinconway: action item, I'll do the dramatic reading (in costume in Atlanta) 18:34:58 <konetzed> amrith: why arnt these config values and not hardcoded 18:34:59 <konetzed> ? 18:34:59 <SlickNik> amrith: I'd +2 the review based on Viswa's reply. We should _not_ be using kill. 18:35:14 <amrith> I pasted into chat a login session with service start and stop 18:35:23 <konetzed> amrith: so anyone can use what ever they want for start and stop w/o having to change the code 18:35:29 <amrith> ViswaV and I had a very detailed discussion of this as well 18:35:40 <vipul> amcrn: what do you think? you guys run cassandra? 18:36:00 <amcrn> yeah, let me test the stop server command, and ensure it works 18:36:02 <amrith> konetzed: the idea is that in the blueprint they would be to make this do'able 18:36:19 <amcrn> because it looks like everyone agrees that kill is overkill if service-stop works 18:36:30 <hub_cap> konetzed: i think thats a good question 18:36:35 <grapex> amcrn: That's my view- it's overkill, *if* something else works. 18:36:56 <amrith> give me a second and I'll find the URL to the paste 18:37:11 <grapex> I'll admit it everyone... I'm not above writing a call to "kill" into the guest code, but only if there's a great reason. Probably don't want to kill a service we'll be handing to end users either. 18:37:23 <grapex> Better for an admin to log in and figure out why it won't die. 18:37:48 <kevinconway> we can use Python's os.kill() 18:37:55 <kevinconway> skip the subshell 18:38:16 <grapex> kevinconway: But first we'll call grep to get the process ID, so we know we have the right one. 18:38:17 * hub_cap marks kevinconway as a perma-troll 18:38:18 <vipul> yea than we wouldn't feel so guilty 18:38:19 <amrith> http://paste.openstack.org/show/74160/ 18:38:42 <amrith> so, there's more to killing the process than the kill with the right pid 18:38:46 <hub_cap> so amrith i think koko is saying make the start/stop/restart configurable from the config.file 18:38:48 <amrith> there's also the issue of cleaning up a pid file 18:39:00 <hub_cap> as opposed to putting it in a .py file 18:39:03 <amrith> tht would be a good option for the bp 18:39:16 <hub_cap> yea i think thats valid, so if kevinconway wants to use pkill w/ his bash agent he can 18:39:18 <amrith> do you want to do that now, given that amcrn's fix is in already? 18:39:32 <esmute> lol 18:39:36 <amrith> amcrn: pl take a look at the paste output I put above 18:39:37 <SnowDust> koko :D ! 18:39:40 <amrith> it has the service start/stop 18:39:43 <kevinconway> pkill -9 -f * to be exact 18:40:00 <amrith> hub_cap: you proposed init 0 yesterday. That worked. 18:40:11 <hub_cap> HAHAHAHA 18:40:31 <amrith> so what does core propose? 18:40:39 <amrith> I have a +1 and a -1 18:40:42 <juice> kevinconway: the ONLY problem with that is that the guest agent won't be around to report back status 18:40:53 <hub_cap> i think we should do this 18:40:57 <kevinconway> juice: it's scheduled on a cron job to restart 18:40:59 <kevinconway> self-healing 18:41:00 <hub_cap> but i also think we should go further and make it configurable 18:41:03 <hub_cap> lol kevinconway 18:41:07 <grapex> hub_cap: Agreed 18:41:09 <amrith> if a data store dies in the forest and there is no guest agent to see, does it send a notification? 18:41:40 <amrith> so hub_cap: do you want to do that for both start and stop, now? 18:41:42 <grapex> amrith: Nope. 18:41:47 <amrith> I'm asking the tactical question for now 18:41:48 <hub_cap> amrith: im pretty sure i saw a commercial about that recently 18:41:59 <SlickNik> We could make it configurable as part of the abstraction bp amrith suggested previously. 18:42:02 <grapex> hub_cap: Was it for life insurance? 18:42:09 <hub_cap> so amrith i think its ok to do that in multiple parts, but i dont think itll be a ton of work to do them together 18:42:16 <hub_cap> grapex: it was for toaster pastries 18:42:23 <grapex> hub_cap: Odd 18:42:26 <hub_cap> hahahahaha 18:43:07 <hub_cap> vipul: SlickNik grapex what do yall think? 18:43:37 <grapex> I am down for this blueprint. I do think in the future we might want to make things more configurable but this seems like it will work for everyone atm. 18:43:40 <SlickNik> I'm good with doing it incrementally. 18:43:46 <vipul> i'm ok with the current patch, and as part of the start() stop() abstraction mentioned earlier.. we can make it somehwat configurable 18:44:04 <konetzed> i think it should end up configurable sooner rather then later but i think its ok getting in amrith changes in now, just my 3 cents 18:44:56 <esmute> what is being configurable? Whether to use systemd/upstart/killall etc? 18:45:07 <cp16net> looking at this more i think all of the {datastore}/system.py is just a big configuration 18:45:24 <cp16net> shouldnt this be in configurations? 18:45:29 <hub_cap> ++ 18:45:45 <SnowDust> cp16net : +1 18:45:57 <amrith> OK folks, I'll be prepared to talk more about this in Atlanta (the blueprint, that is). if you approve https://review.openstack.org/#/c/82080/ it is good to go. 18:46:33 <SnowDust> if cfg.py defines datastore options why not start/stop/ for it as commands too ? 18:46:43 <amrith> cp16net: yes I think it should be configurations but preferably in some way such that you don't have duplicated code that ends up being a bug fix requiring identical changes in five places. 18:47:09 <robertmyers> SnowDust: yes, we should do that 18:47:11 <amrith> SnowDust: that is a good place to put it. 18:48:51 <amrith> Other thoughts? hub_cap, amcrn, grapex, SlickNick 18:49:02 <hub_cap> honestly 18:49:11 <hub_cap> this file is a big config file just like its been said already 18:49:16 <hub_cap> we should, porbably in a diff review 18:49:23 <hub_cap> move it all to config based w/ sane defaults 18:49:30 <grapex> hub_cap: And once the freeze thaws lets move the guest agent around a bit 18:50:06 <grapex> I think at that time it'll be pretty simple to see how the configs work and whoever fixes it might get an easier consensus. 18:50:06 <hub_cap> we could probably do CONF.ubuntu.blah and CONF.redhat.blah (im sure we can come up w/ better names but u get the point) 18:50:11 <hub_cap> ++ 18:50:13 <SnowDust> amrith, my bp that abstracts cfg.py for datastore options https://blueprints.launchpad.net/trove/+spec/refactoring-datastore-options-in-cfg is already approved, i can extend that to include what i was talking 18:50:15 <amcrn> sorry, back, was on phone call 18:50:43 <amrith> SnowDust: let's talk about that approach 18:50:58 <amrith> will u be in Atlanta? 18:51:03 <hub_cap> way to go amcrn we wouldnve never known 18:51:13 <hub_cap> amrith: this should be solvable befoere atl 18:51:14 <grapex> amrith: Maybe we don't need to wait until Atlanta 18:51:17 <hub_cap> LOL 18:51:45 <amrith> hard to put a face to a name on IRC 18:51:54 <hub_cap> AHHHHH 18:51:56 <SnowDust> amrith i got my atendee pass free .. sponsor tesora tickets :) as well .. i will be there :D 18:51:56 <amrith> austin was good 18:52:00 <hub_cap> nm then amrith 18:52:02 <amrith> also for brisket 18:52:06 <amrith> and I'm a vegetarian 18:52:14 <hub_cap> YESHHHHH 18:52:21 <hub_cap> ok so one last thing 18:52:24 <hub_cap> i assume we are done w/ this 18:52:30 <hub_cap> amrith: do u feel good about the outcome? 18:52:34 <hub_cap> service is valid 18:52:36 <amrith> I'm assuming that reviewers will do their +1's offline 18:52:37 <amrith> I'm good 18:52:43 <amrith> thx all 18:52:48 <amcrn> so reading through the chat log: we're going to merge the service start/stop (assuming stop actually works), but in the long-term amrith and others will work out the long-term strategy? 18:52:48 <hub_cap> but we also need to eventually make tehse configurable 18:52:54 <amrith> SnowDust: sorry have to dissapoint on the tickets 18:53:04 <hub_cap> amcrn: gosh why dont u say waht i said but smarter 18:53:19 <SnowDust> amrith: .. will feel happy to discuss on the approach anyway 18:53:24 <amcrn> :) 18:53:35 <hub_cap> #topic icehouse rc1 18:53:45 <hub_cap> soooooo 18:53:51 <hub_cap> we are one bug out from being RC1 complete 18:53:57 <hub_cap> we will merge that today 18:54:01 <cp16net> yay 18:54:07 <hub_cap> and ttx will cut RC1 tomorrow 18:54:07 <SnowDust> congrats !!! 18:54:13 <cp16net> you feel the excitment? 18:54:21 <cp16net> (sp) 18:54:23 <vipul> (party) 18:54:36 <SnowDust> cores, <3 18:54:36 <grapex> hub_cap: I knew if we tried hard enough we'd eventually be able to merge all the bugs to RC1. 18:54:37 <SlickNik> hell yea! 18:54:41 <hub_cap> HAHAH 18:54:48 <hub_cap> once RC1 is cut 18:54:49 <grapex> So, when do we ship the CDs? 18:54:55 <amcrn> grapex: just click +2 next to every review, it's the gates job to protect us, right? 18:54:57 <hub_cap> grapex: lets let ubuntu do that ;) 18:55:00 <amcrn> ;) 18:55:04 <hub_cap> amcrn: ++ 18:55:06 <SlickNik> lol @ amcrn 18:55:09 <vipul> grapex: so you can put it in your microwave :p 18:55:11 <hub_cap> so once RC1 is cut, juno officially opens 18:55:12 <grapex> I've been looking forward to putting it up next to my copy of Norton Desktop Commander 98. 18:55:24 <grapex> hub_cap: Awesome. 18:55:32 <hub_cap> hey m$ opensourced "word for windows" 18:55:34 <cp16net> grapex: i need 500 hours free 18:55:36 <amrith> sorry to be the party pooper ... amcrn and grapex (maybe others) have to +1 my review ;) 18:55:48 <amrith> I'll get right on to the guest agents on Windows blueprint after that 18:55:53 <vipul> i heard the same for MS DOS... 18:56:00 <hub_cap> yuppp 18:56:04 <hub_cap> so i just wanted to say 18:56:08 <vipul> only 30 years later 18:56:10 <grapex> vipul: Free Dos as a service.... 18:56:29 <hub_cap> core will be looking at all the reviews, so plz dust them off, take them out of abandon mode 18:56:32 <hub_cap> and rebase them 18:56:43 <vipul> i bet people are dieing to have that 18:56:43 <hub_cap> we will likely do a run thru of tox before looking at code 18:56:46 <hub_cap> recheck no bug ;) 18:57:08 <SnowDust> vipul, :p 18:57:17 <hub_cap> so if your code fails tox, we will not look at it yet ;) 18:58:17 <hub_cap> ok so if no one has Qs im going to end this early 18:58:21 <kevinconway> o/ gpg keys ==> sign them 18:58:38 <vipul> oh what about the vote on Cluster API? 18:58:53 <kevinconway> i need to wire bitcoins to a wealthy long lost relative in another country 18:58:59 <kevinconway> i can't do that without a well signed key 18:59:20 <vipul> kevinconway: they prefer real money i think 18:59:24 <amcrn> vipul: imsplitbit doug_shelley66 and myself mentioned in the chat earlier this morning that we need to sync up next week 18:59:36 <imsplitbit> yeah 18:59:40 <imsplitbit> I'm in 18:59:42 <vipul> amcrn: Ok, sounds good.. 18:59:47 <hub_cap> lol kevinconway 18:59:59 <hub_cap> cool on that note 19:00:01 <hub_cap> #endmeeting