=== kadams54-away is now known as kadams54 === kadams54 is now known as kadams54-away [15:10] rick_h_: do you have 2 minutes for me? [15:10] lazyPower: sure thing [15:10] i just got a mail from castro about the big data topic page, our second listed bundle 404's, but the repository exists in lp [15:11] https://jujucharms.com/big-data - the high performance view details link [15:11] lazyPower: looking [15:11] any insight? i'll file a bug @ github - but it makes no sense to me [15:12] lazyPower: is that https://jujucharms.com/high-performance-batch-processing/4 ? [15:12] thats the one [15:12] its ina different namespace hwoever [15:12] lazyPower: looking to see the diff in the charmers one that shows up vs the other one with the bigdata-charmers [15:13] you know rick_h_, if i remember correctly - we are warehousing teh bundles in both places - but only /~charmers/bundle link is the store link (otherwise its marked as development focus in launchpad) [15:13] i'm wondering if there's a collision going on here, or something similar [15:14] but i bug reported it so we can track progress https://github.com/CanonicalLtd/jujucharms.com/issues/44 [15:14] lazyPower: yea looking [15:17] jcsackett: can you check out and see why ^ would 404? It looks like it should load via the api here: https://api.jujucharms.com/v4/~bigdata-charmers/bundle/high-performance-batch-processing-6/meta/any?include=bundle-metadata [15:18] jcsackett: wondering if the -charmers is causing us to not just remove dupes, but 404 the details [15:18] rick_h_: looking. [15:18] lazyPower: I'm going to be this is because the team name ends in -charmers and we've got a temp hack in to remove dupe entries [15:18] lazyPower: so it's 404'ing this one so that the promulgated one is the one that shows/is used [15:18] lazyPower: jcsackett can confirm if that's the reason for the 404 and get back to you [15:19] lazyPower: we're working on a back end model change to remove the dupes properly, but it'll take some time this month for that to complete and get rolled out [15:19] thanks rick_h_, i appreciate you taking a look [15:19] yeah, murphy's law + time = current state of things [15:19] totally understood [15:19] lazyPower: so my 'answer' right now is to update that page to point to the promulgated bundle which should be the best source anyway [15:19] ok, thats not something I can do right? [15:19] lazyPower: we're working on a release today and can update that url if you can confirm it's ok to do so [15:20] lazyPower: no, we have to adjust the url in the template file but it's a one line change jcsackett can get in before release today [15:20] i +1 that notion. having a bundle listed vs a 404 is a 100% better experience. [15:20] jcsackett: follow ^ ? [15:20] if we need to do anything charmer side to update, i'll make that happen [15:20] lazyPower: right, so the thing you all can do is update the ~charmers version with the latest code form the other team [15:21] it looks a bit behind using r4 of the charm vs r7 [15:21] jcsackett: http://bazaar.launchpad.net/~bigdata-charmers/charms/bundles/high-performance-batch-processing/bundle/view/head:/bundles.yaml and http://bazaar.launchpad.net/~charmers/charms/bundles/high-performance-batch-processing/bundle/view/head:/bundles.yaml [15:21] rick_h_: the promulgated version should just be the bundle absent the /u/..../ stuff, right? [15:21] jcsackett: correct [15:21] b/c that's *also* 404. [15:21] rick_h_: https://jujucharms.com/high-peformance-batch-processing/ [15:21] jcsackett: https://jujucharms.com/high-performance-batch-processing/4 [15:21] jcsackett: https://jujucharms.com/high-performance-batch-processing/ [15:22] those are 404s for me. [15:22] https://jujucharms.com/high-performance-batch-processing/ is a 404? [15:22] rick_h_: also, we only filter out the *-charmers stuff on search results--we don't *not* display them. [15:22] jcsackett: that's what I wanted to verify [15:22] jcsackett: since it shows up in the api I was wondering why it would 404 [15:22] jcsackett: so we might need to runa local storefront and debug the page load through the api [15:25] rick_h_: no need. [15:25] the answer is far simpler. [15:25] jcsackett: hah [15:25] the link on big-data is to high-peformance-..., not high-performance-... [15:25] that 'r' is pretty important. [15:25] i'll go ship a trivial. [15:25] jcsackett: ty kindly [15:25] lazyPower: ^ [15:25] lazyPower: typo ftw [15:25] simple fix, we'll have it up soon. :) [15:26] well FML [15:26] \o/ [15:26] i like simple fixes === kadams54 is now known as kadams54-away === kadams54-away is now known as kadams54 === whit is now known as whit|review-q === kadams54_ is now known as kadams54-away [16:49] frankban_: so the uuid endpoint, will it be required for the juju-core endpoint as well or just the wss requests? [16:50] hatch: juju-core endpoint? oh, you mean the HTTP proxy? [16:50] yeah [16:50] ^/juju-core/(.*) [16:50] hatch: not sure, let me check [16:50] that's for the icons right? [16:51] thanks [16:51] hatch: that's both for the icons and for uploading local charms IIRC [16:51] ahh [16:51] so then it does make sense that it should require it...in some capacity at least [16:54] hatch: I confirm they also use the new endpoint: https://github.com/juju/juju/blob/master/apiserver/apiserver.go#L323 [16:55] frankban_: alright thanks! [16:57] hatch: IIRC the ProxyHandler should already redirect everything under /juju-core [16:59] hatch: so this is actually also a change to be made in the GUI later [16:59] frankban_: correct, it's looking like it simply redirects what's been sent to it [17:11] frankban_: with a deployed juju-gui charm, what's the best way to modify the live source/pdb etc [17:11] for the guiserver that is [17:15] hatch: hangout? [17:15] frankban_: sure, joining standup === kadams54-away is now known as kadams54_ [18:30] uiteam looking for reviews on the charm update to support v4 bundles: https://code.launchpad.net/~makyo/charms/trusty/juju-gui/guiserver-v4bundle/+merge/249236 === kadams54_ is now known as kadams54-away === kadams54-away is now known as kadams54_ [20:46] woah nodejitsu was bought by godaddy [20:47] what is nodejitsu? [20:47] oh. hosting. [20:47] kind of like heroku for node [20:47] sortof [20:48] godaddy is evil, so sounds like a new nodejitsu work alike is needed. [20:48] doesn't node work on heroku? [20:48] if you get a node gear [20:48] or whatever they call their vm's [20:48] heh [20:48] buildpack [20:48] that's it [20:49] a while ago nodejitsu was far superior [20:49] not sure anymore [20:49] it's been probably 2 years since I looked heh [20:49] I can't imagine their new overlords will be very good for the company [20:50] any company that supports a closed internet can shoveit === kadams54_ is now known as kadams54-away [21:46] I have this simple sublime plugin that marks new/changed lines when working with git - I do not have this functionality with bzr and I miss it! [21:59] Morning === whit|review-q is now known as whit === kadams54 is now known as kadams54-away === kadams54-away is now known as kadams54