[00:11] hey, mjc appears up yay [00:11] yup! [00:11] I'd be interested to know what the issue was [00:23] rick_h_: gl on the winter tire search - it's damn near impossible to compare tires because no one has experience with both :) so 'ratings' are essentially worthless [00:23] so flip a coin ;) [00:25] hatch: yea, when I got tires on the subaru it seemed I found some really good info. Now I can't seem to find anything [00:25] buy the same ones for the bee dub [00:25] then again I got a couple different ones and I was always happy with them. [00:25] vee* [00:25] well, they're not around any more [00:25] oh boo [00:26] I think I have the x-ice on my truck [00:26] and the subie tires were all season with decent winter tread. I'm just getting straight winter since I need to get 17" wheels anyway (19' [00:26] s on there now) [00:26] so first time doing the full swap of wheel and all for seasons [00:26] I think I'm going to go with something with some more tread next time though for deeper snow [00:26] yea, the blizzak has a little more tread [00:27] I'm leaning that route atm [00:28] yeah - the ice performance of either will probably be neglegable [00:29] yea, but ice ice is rare. I more want to go play when 6" falls [00:29] ohh haha [00:29] I don't like getting trapped in the house. I always need to prove I can go out and get a coffee :) [00:29] 12" came down during the day, we need to get some milk from the farthest store for the morning hehe [00:30] but it's the whole family safety thing. If I spend decent $$ and it performs better when I'm dodging the idiot spinning his wheels on the curve running into me...priceless [00:30] tires make such a difference. More so than a lot of vehicle/traction control systems [00:32] well it's the only thing connecting those couple tonnes to the road :) [00:32] holy smokes winter tires have gone up in price [00:32] ohhhh nm heh [00:32] I was reading it wrong [00:32] yea, tires + wheels are going to run me close to 1500-2000 :/ [00:32] ahh - yeah I already have another set of wheels so I'm lucky that way [00:32] though getting the 17" tires will save a chunk. Started out looking at 18" ones [00:33] also saves me from having to pay to switch them over [00:34] yea, bell tire will do storage for $60 ish a season [00:34] don't you have a huge yard? [00:34] lol [00:34] so that's worth it to me to just swap the wheels out, put them into storage, and I'll see you in 6mo [00:34] no, not really. My garage is full of woodworking gear [00:34] if I had built my dream shed I'd be cool, but didn't do that this year. [00:35] yeah ours goes into the shed [00:35] I've got a baby plastic shed full of lawn gear. At least that's out of the woodshop now. [00:35] ohhh - yeah a couple years ago I picked up a nice big tin one [00:36] used [00:36] nice [00:36] yeah it was a steal, but we had to take it apart [00:36] but I think I saved like $800 from new [00:36] so I was ok with that :) [00:39] yea, that's good stuff. You put it on dirt or pour a pad down? [00:40] built a 2x6 floor and put it on that so if I need to I can just drag it wherever I want [00:40] and I had the spare 2x6's [00:40] haha [00:40] nice! [00:40] honestly though, it'll probably never move [00:42] my $/km that I spent on that truck is just nuts lol [00:42] I need to get in it every morning and drive it around or something [00:42] haha === hazmat` is now known as hazmat [11:09] thanks frankban [11:10] rick_h_: thanks for QA and suggestions. /login/data/wordpress-api-response.json 404 is very annoying [11:25] frankban: the /test/data? [11:27] rick_h_: make test-prod (and therefore lbox propose/submit) fails intermittently here with this error: http://pastebin.ubuntu.com/6183447/ [11:28] frankban: hmmm, usually when I would get that there'd be some error where a navigate would occur breaking the loading of test data. I'm guessing some /login test is doing something bad [11:28] strange that it's intermittent [11:28] rick_h_: yeah [12:59] I'm going to take my "lunch" early. back in about an hour. [13:00] gary_poster: have fun === TheRealMue is now known as TheMue [13:52] mornin [13:53] morn [14:02] is the design meeting on? No one's there? luca__ gary_poster ? [14:02] rick_h_: yup, coming now [15:10] aiiiie! [15:10] I closed the thing and lost my chat! [15:10] boohoo hoo [15:10] ??? [15:11] I had info I wanted n the google hangout chat [15:11] and forgot to copy it out [15:11] :-( [15:11] yeah, info in hangout chat is so easy to loose [15:11] IRC logging for the win [15:11] yeah, that or a google doc. hopefully I will learn my lesson :-( [15:12] they finally added a 'do not mute' button when typing [15:12] so that's cool [15:12] no more js console hacks [15:16] To solve this issue I create an ssh tunnel to it, but I'm guessing that's not the real story for this guy...or is it? http://askubuntu.com/questions/346991/juju-gui-public-ip [15:16] I'll reply in kind if it is [15:23] rick_h_: looks like you can upgrade your favourite tech toy http://www.engadget.com/2013/10/02/fitbit-force-wristband-watch/?utm_medium=feed&utm_source=feedly lol [15:26] hatch: ugh, stupid thing. [15:26] hatch: pebble > fitbit [15:27] not going to go that route [15:27] hatch: as to the SO question, check with jcastro but I think you can configure lxc to not use nat, but bridged networking? Maybe they can tweak a config to get the IP to be exposed naturually. Otherwise, yea it needs a tunnel [15:31] we need a Best Practice for LXC networking with Juju [15:31] jcastro: sounds good [15:32] in the meantime....? shall I reply with the tunnel instructions? [15:32] :) [15:32] sure [15:32] I need to do a call on the list for common workflows [15:32] hatch: there's a question on the bridge side here with an accepted answer http://askubuntu.com/questions/231666/how-do-i-setup-an-lxc-guest-so-that-it-gets-a-dhcp-address-so-i-can-access-it-on [15:32] hatch: might at least link to it as a potential method to follow up on [15:33] sure thanks [15:33] which links to http://askubuntu.com/questions/256304/public-ip-address-for-lxc-container/311003#311003 [15:33] <_mup_> Bug #311003: [Patch] Fix linking when with --as-needed --no-undefined [15:33] lol, welcome to the internet [15:33] link -> link -> ...which one do I want!!! [15:35] so the UK is censoring the internet, our city is trying to zone for the newly allowed strip clubs...man people are afraid of nekkid ppl!! [15:35] abentley, benji: could one of you have a look at this MP: https://code.launchpad.net/~adeuring/charmworld/proof-lib-path/+merge/188871 ? (tiny diff) [15:36] adeuring: sure [15:36] thanks! [15:41] replied http://askubuntu.com/questions/346991/juju-gui-public-ip/352764#352764 [15:41] <_mup_> Bug #352764: If kwallet is disabled, network manager plasma widget does not connect to wirelessly

[15:41] thanks mup but that wasn't a bug [15:48] hey frankban what happened to https://codereview.appspot.com/13900044/ ? looks like it never landed? [15:48] but I thought this was fixed [15:48] adeuring: your branch looks good [15:48] gary_poster: looking [15:48] benji: thanks! [15:50] jujugui call in 10 kanban now [15:50] jujugui call in 10 [15:50] darn! :-P [15:50] \o/ [15:50] lol [15:51] gary_poster: ah, do you remember that we discussed the switching to the ServiceUpdate call and realized it was not backward compatible? I created and landed another branch to fix that bug: https://codereview.appspot.com/13917043/ [15:51] frankban, ah right! could you mark those (MP and RV) as dead so future generations are not confused? :-) [15:52] gary_poster: sure :-), sorry for the confusion [15:52] np [15:52] ooh, someone soon will be able to have the bug number 1234567 [15:52] * gary_poster starts filing bugs... [15:53] :-) [15:58] jujugui call in 2 [16:00] Makyo, wassup? [16:01] gary_poster, juju is eating my machine [16:01] nom nom nom [16:15] Going to run a quick errand so we actually have food for lunch besides eggs, won't be long. [16:28] mmm foood [16:28] i forgot breakfast again [16:28] damn [16:30] I love it when auth fails while lboxing AFTER it's run lint and tests...ugh lol [16:41] jujugui anyone else having issues lboxing? [16:41] error: use of closed network connection [16:41] 3 times in a row now? [16:42] hatch: not an issue early this morning when I landed mine, but that sounds like LP issues. [16:44] yeah, well will try again I guess!! [16:44] hatch: I've gotten that error recently (but it cleared up when re-run); I too suspect an LP issue (probably a timeout) [16:48] here it goes [16:48] 5th time is the charm :) [16:49] does anyone know how to copy text in the terminal with the mouse in tmux when you have `setw -g mose-mouse on` ? [16:52] btw gary_poster using `bzr mv` worked properly this time - not sure what happened last time [16:52] hatch: middle-click is what I use to copy all text into the terminal [16:52] hmm no middle click [16:53] oh into the terminal [16:53] shift-insert then I think? [16:53] hatch, good. yeah, has always worked for me [16:53] I need out of the terminal [16:53] tmux now captures the mouse highlight it seems [16:53] hatch: then highlight with the mouse and then middle-click it where you want it to go [16:53] as soon as I release the highlight is gone [16:54] not sure [16:54] yeah it's pretty odd [16:55] rick_h_: would ya mind? https://codereview.appspot.com/14175046/ [16:55] hatch: sure ting [16:55] thing [16:55] coolio [17:08] hatch: comments and qa issue inbound [17:10] rick_h_: reading thanks [17:37] rick_h_: fixed - reproposing [17:38] hatch: k, sec and will relook [17:39] rick_h_: sorry have test failures again [17:46] lboxing again [17:47] gary_poster: thought you would want to know that Dell just upgraded the XPS 15 laptops.... :P [17:47] * hatch ducks [17:49] hatch, heh :-P [17:49] rick_h_: ok reitveld has the new stuff [17:51] hatch: k, looking [17:51] hatch: so then did you bzr grep all BrowserCharmView and look for the data passed in? [17:52] yeah did I miss one? [17:53] hatch: no, just asking [17:53] doesn't look like it [17:53] oh ok [17:53] but I also use `ack-grep` :) [17:53] does bzr grep work better? [17:53] hatch: cool, sorry. My review is to ask questions and prod thinking vs just "here's my list" [17:53] hatch: bzr grep just makes sure to only limit to files in the bzr content [17:54] oh ok - that might be better because then it'll skip any of the built files [17:54] git has the same thing. Handy for ignoring files you don't care about easily [17:54] yep [17:54] I'll try and remember to use that [17:54] I used to have 'ack' (OSX) set up to ignore anything in the .gitignore so I guess that just stuck :) [17:59] rick_h_: found out how to copy in tmux with my settings hold 'shift' or in iterm(OSX) hold 'option' [17:59] just fyi [17:59] bah, I want to bank my bug credit for my next QA day :P [17:59] hatch: cool, not a problem here so must be some mac-ism or something. Good to know I guess [18:00] hatch: and LGTM thanks for the updates [18:00] thanks! [18:00] rick_h_: do you have tmux capturing your mouse scroll? [18:00] so you can scroll up/down in tmux with the mouse? [18:00] hatch: hmm, I do that in my terminal I think. [18:00] that setting appears to trigger the issue [18:00] so no, guess I don't [19:02] benji: ping, got a sec? [19:02] rick_h_: sure [19:02] rick_h_: video chat or here? [19:02] benji: video please, sec will set up [19:02] rick_h_: try this benjiyork.com/chat [19:03] connecting [19:14] rick_h_: you were my first chat using that thing, other than not having a mute it worked well [19:14] what is it? [19:15] benji: yea, worked fine. Backgroud noise on the road there but other than that sounded good and loaded up quick [19:16] <3 the nice url [19:17] heh [19:17] yeah, me too [19:21] https://code.launchpad.net/~rharding/charmworld/bundle-metadata/+merge/188905 for a WIP with notes on why some stuff is changed. Feel free to smack my hands with the rule where I'm going against the new grain [19:21] appreciate the time to peek at it [19:24] rick_h_: looking [19:25] heh, there's some silly stuff in the mid-mode. #185 for instance is meant to be the upcoming icon helper method but it looks just like the next block in the if [19:33] rick_h_: why the new constant empty dict being returned from _extract_charm_id? [19:34] benji: confusion. Typo going back/forth between methods since I added the third param to the bundle things to help find bundles [19:34] benji: chalk it up to mid-work [19:34] k [19:35] benji: so get_file returns the bytes, but none of the extra info for serving the file like content_type and such? [19:36] or nvm, that's the gridfs object so should have the content-type in there [19:36] ok [19:43] need to get kids from school. back soon [19:53] rick_h_: your branch looks fine to me [19:53] benji: cool [19:53] benji: the file hashes in the basket db can be yaml? [19:55] rick_h_: I don't understand the question. [19:56] benji: I'm looking at how the files are getched and noticed in the unquote_yaml. nvm, that's the period escaping stuff. Ignore me [19:56] man, is today monday or something? [19:56] heh [19:56] It's Monday somewhere. [19:57] that and I keep typing backet vs basket [19:57] I can't seem to hit 'as' in a row [20:02] * gary_poster here [20:03] oh rick_h_...you available to talk in 5? you are probably past your EoD, yeah? [20:04] gary_poster: I can talk in 5 [20:04] cool thanks [20:15] rick_h_, uh "five minutes"... https://plus.google.com/hangouts/_/d909f7c1b9c8db56be05af4887e38eef6587a85c?hl=en if you are still ok :-) [20:42] benji: still around? [20:42] rick_h_: yep [20:42] does fs.put(content, contentType=content_type) [20:42] *** TypeError: must specify an encoding for file in order to write unicode [20:43] ring a bell? Trying to manually put a file into fs. Looks like all we do is shove a string of content in but getting this when trying to do it via pdb [20:43] the code is currently just silently passing without any files added to the gridfs [20:43] content is just a u'Test Content' for a test file [20:44] rick_h_: I don't think I've seen that error in that context, but the problem is that it wants a string and it doesn't know what encoding it hould use. de-unicode your string and you should be fine. [20:44] hmm, guess it just doesn't like the unicode. I see it in the docs now. Ignore me some more [20:45] but still isn't there...ugh. after an fs.put() and then an fs.list() is empty [20:46] hmm, there it goes. Ok. maybe I just timed it wrong. Stupid 'eventually consistent' not liking my pdb points [20:48] heh [20:48] yay! test passes. [20:50] thanks for the help today benji. [20:50] my pleasure [20:58] benji: #1190944 is the orig of your markdown bug [20:58] <_mup_> Bug #1190944: The copyright header of README.md in the charm is shown

[20:59] rick_h_: thanks, I'll mark mine as a duplicate [21:01] I changed 1190944 to "Low", it was "Medium" [21:30] jujugui I'm investigating the bug -re gary_poster email [21:31] * hatch wishes everyone replied to emails from the top :D [21:37] gary_poster: I'm a little confused here - it appears that it -works- in the gui but does not in the console.... [21:37] the email conflicts itself [21:58] nm I was able to reproduce [22:09] bcsaller: ping [22:09] whats up? [22:09] avail for a quick call? [22:09] re this email [22:10] https://plus.google.com/hangouts/_/4cec75a250f0b68cdf45f3c060a997dbdccb9c3b?hl=en [22:30] bcsaller: hangout is not working lol [22:31] nope [22:31] hatch: happy to say its a charm issue if we can get away with it, but we should make an attempt to detect it somewhere in the pipeline then [22:32] and update the docs [22:32] its a bit sad that there isn't an easy fix on our side though [22:32] go javascript? ;) [22:43] bcsaller: so can you think of any other ideas besides formatting the keys? [22:44] we could restrict to a single level of recusion? [22:44] nah that could still break the YUI ones [22:44] so basically only formatting the keys will help [22:45] hatch: I think we detect the case and inform the user. I would like to test that you can actually get/set those keys in a core env though [22:45] yeah i'm spinning up one right now [22:45] in which case its really matter of enforcing proper policy that it doesn't work [22:45] takes a long time :) [22:45] yeah [22:51] hatch: I suppose we could fork, fix and offer up the charm with s/\./\-/ in the key access [22:52] heh - yeah that would probably be the best way [22:53] bcsaller: although I wonder if it requires the keys in that format [22:53] so we would need to format them again in the charm [22:53] hah [22:53] we'd have to change config.yaml and the hooks [22:53] right - I mean, I wonder if it uses the same keys to set the values in Hadoop [22:53] I Know nothing of how Hadoop works to know if those are valid keys [22:55] and of course launchpad is showing the files are empty... [22:56] http://manage.jujucharms.com/charms/precise/hadoop/hooks/config-changed ::170 ish [22:56] so yes [22:56] heh [22:57] :/ [22:57] hatch: well, it looks like it uses the name twice, once in the dot form and once to lookup the juju value, which could change [22:58] right, we could change the key value in juju [22:58] ok I'll update the ticket with this [22:58] see where we are going with this [22:58] mostly because it's STILL starting [22:58] lol [23:00] bcsaller: yuss it breaks on juju-core [23:00] :) [23:00] * hatch does a happy dance [23:00] ha, good [23:01] Morning [23:01] at least my understanding of the world is correct. So we can do a number of things now. Warn in the GUI, warn on the CLI, and warn when we proof/ingest the charm [23:01] well, on ingest it would be a fail [23:02] yeah ok I'll write this all up in the ticket [23:02] hatch: it does look simple enough to fix in the wild, but still a pain [23:02] yep [23:03] moring huwshimi [23:03] if I could type [23:03] morning huwshimi [23:03] heh [23:03] ahh just call him sashimi [23:03] :P [23:06] I probably spelt that wrong [23:19] hatch: what's up? What bug? You working around '.' in something? Charmworld does the escaping/dealing with that usually [23:20] * rick_h_ doesn't see a gary email/bug at the top of the email [23:20] rick_h_: ttps://bugs.launchpad.net/charms/+source/hadoop/+bug/1234365 [23:20] <_mup_> Bug #1234365: juju-gui does not give option to pass configure paramter [23:21] hatch: ah, so this is purely on the gui talking to juju side and not around charmworld [23:21] hatch: nvm, ignore me then [23:21] already done [23:21] :P [23:21] jk [23:21] :P [23:22] rick_h_: honestly though, shouldn't charmworld have seen the config options using .'s and failed the injest? [23:23] hatch: no, we escape the . in storage of mongo and unescape during api calls [23:23] hatch: as you say in your note, it's valid from the cli and such [23:23] hatch: it's up to the client to deal with the issue once we've worked around the '.' not being able to be stored in mongodb itself [23:23] rick_h_: well no the CLI throws an error if you try to access/set that config option [23:24] huwshimi: https://code.launchpad.net/~jcsackett/juju-gui/charm-slider/+merge/152418 is the branch that initially added it. I'd search for a branch afterwards that removed the widget from the codebase for the final version [23:24] huwshimi: well then yea, I'd talk to marcoceppi about having proof blow up in that case and then charmworld would have rejected it [23:24] huwshimi: and the proof tool will tell the author when they write the charm not to do it before they ever submit/get through any review process [23:24] sorry hatch ^^ [23:25] rick_h_: Thanks for that [23:25] rick_h_ huwshimi: if proof is missing functionality (hatch ?) feel free to open up a bug against charm-tools with a description. I'm gearing up for a 1.1 release before the cloud sprint and will try to include anything you guys need [23:25] huwshimi: np, it was in the code for at least a few weeks I think [23:25] rick_h_: haha [23:25] (also patches accepted) [23:25] marcoceppi: so does it sound reasonable/is it known that having config options in the charm with a '.' in the name is bad? [23:25] marcoceppi: ok I will do that, thanks https://bugs.launchpad.net/charms/+source/hadoop/+bug/1234365 to see the issue in question [23:26] <_mup_> Bug #1234365: juju-gui does not give option to pass configure paramter [23:26] rick_h_: I'm not awayre of an issue with a . [23:26] rick_h_: I know that's a problem with charm names [23:26] yea, that's the rub I guess. hatch so we should dupe that the cli hates charms with a . in a config name, maybe even email the juju list about it, then update proof to check and erorr on it [23:26] problem goes away :) [23:28] oh charmtools is written in python [23:28] yeah you wouldn't want me to issue a pr [23:28] :D [23:46] hey hatch thanks a lot. I read backlog and bug. looks like you have it in hand. if you have to go, send me an email to hand off. If you end up staying up late on this to resolve, send me an email and then take some corresponding time tomorrow [23:46] I have to go take care of kids and house meanwhile :-) [23:46] ttyl [23:46] gary_poster: waiiiiit [23:46] see your email [23:46] hatch here [23:47] i JUST sent a follow-up with Narinder [23:47] would you like me to do the mechanical change and 'hope' it works? or wait for Kapil? [23:49] I'm ok with either options [23:49] hatch ack. so the mechanical change would be to change the values in config and then try to figure out where the charm consumes the config values and change them? [23:49] hatch did you verify that juju set fails? [23:49] right - it does a sort-of mapping to build the XML config file [23:50] yes, it does fail [23:51] so I could change the hook to use the new config values and hope that it still works - but like I mentioned in the email, I don't even know what constitutes 'running' for Hadoop so I wouldn't know how to test it beyond a basic juju deploy [23:51] I'm sure I could learn, but not in a couple hours haha [23:52] cool. we should clarify that. I'll shoot a quick email reply to confirm that. hatch, how quick would it be for you to make that change? easy? or easier for Ben if he is around? If it's a matter of 30 minutes, you could simply offer a mechanical branch and let Kapil or Narinder know its existence. They would then see the issue [23:52] I could probably do it in <1H if it's simply a case of doing a mechanical change and pushing it back up for someone else to take from there [23:54] hatch do it, if you are willing. Thank you. I will have a reply for you to review in just a sec. then I really need to run [23:54] sure - I'll just need to find where the source is kept [23:54] http://bazaar.launchpad.net/~charmers/charms/precise/hadoop/trunk/files/head:/hooks/ [23:55] all the files are empty? [23:55] jcastro: are you around? [23:55] hatch they are all symlinks [23:55] its a common pattern [23:56] hatch see hadoop-common and ganglia-common [23:56] ohhhhh [23:56] interesting pattern [23:56] ok I'll do the mechanical change and then push it up [23:56] cool thabnks [23:57] have a good night