[08:37] <JamesTait> Happy Friday, folks! :)
[09:13] <AlanBell> I am having problems with U1 sync, it is taking days to upload files (a folder of stuff took from Sunday to Tuesday)
[09:13] <AlanBell> I added a bunch more files yesterday and it is still just starting
[09:15] <AlanBell> I think it broke inotify as well, tail-f tells me: tail: inotify resources exhausted tail: inotify cannot be used, reverting to polling
[09:21] <AlanBell> maybe it relates to bug 761578
[09:21] <AlanBell> fs.inotify.max_user_watches=8192 maybe having more files than that in the directory breaks things
[09:25] <czajkowski> morning
[12:07] <ralsina> good morning!
[12:47] <ralsina> dobey: if I can get a re-review of https://code.launchpad.net/~ralsina/ubuntuone-control-panel/sort-shares/+merge/108594 that has a NEEDSFIXING from you it would be great. Thanks!
[13:11] <dobey> ralsina: done
[13:13] <ralsina> dobey: thanks!
[13:15] <dobey> sigh. launchpad keeps timing out :(
[13:24] <dobey> ralsina: heh. i wouldn't say we don't have the expertise to do it, but more we don't have good enough reason to do it. ;)
[13:24] <ralsina> dobey: ever packaged for slackware? I haven't! ;-)
[13:25] <ralsina> I could package for arch but we are already in extras there
[13:25] <ralsina> I don't know if I am missing any other interesting distros... ;-)
[13:25] <dobey> ralsina: slackware is easy. it doesn't have dependency tracking at all :)
[13:26] <dobey> ralsina: hell, i even wrote a package manager.
[13:29] <dobey> heh
[13:29] <dobey> interesting fallback nick
[13:29]  * mandel back
[13:30] <mandel> ralsina, do I still own you the https://code.launchpad.net/~ralsina/ubuntuone-control-panel/sort-shares/+merge/108594 ?
[13:30]  * catdog is finally getting around to setting up irc on the phone
[13:31] <ralsina> mandel: nah, it's approved now
[13:31] <alecu> and by catdog, I mean, /me
[13:31] <dobey> heh
[13:32] <dobey> alecu: can you review https://code.launchpad.net/~dobey/ubuntuone-storage-protocol/validate-ssl-cert/+merge/109362 ?
[13:32] <mandel> alecu, I found the reason for the bug! :)
[13:32] <mandel> alecu, the one related with tcpactivation on mac
[13:34] <alecu> mandel, awesome!
[13:34] <alecu> mandel, tell me
[13:34] <alecu> dobey, sure
[13:35] <mandel> alecu, the NullProtocol does on connection self.transport.loseConnection() this makes the ServerFactory no to create a protocol (I don't know why in mac happened more often) which means that we are waiting for a deferred to be called by a protocol that was never created
[13:36] <mandel> alecu, the amount of time wasted on this has been stupid.. but I have a branch that fixes this in the root (u1-dev-tools)
[13:36] <dobey> alecu: i made a small change to get_ssl_context from your original patch, so we get a better error message when the client breaks if both aren't upgraded simultaneously
[13:37] <alecu> dobey, nice!
[13:40] <alecu> dobey, perhaps you've forgotten to add tests/test_context.py ?
[13:40] <mandel> dobey, alecu, can I have a review for: https://code.launchpad.net/~mandel/ubuntuone-dev-tools/darwin-cleanup/+merge/109363
[13:41] <dobey> alecu: i guess you did too
[13:41] <ralsina> dobey: review please? https://code.launchpad.net/~ralsina/ubuntu-sso-client/pretty-little-liar/+merge/109364
[13:41] <ralsina> and since gatox is not working this morning, I can use a volunteer, it's a very small branch
[13:41] <alecu> ralsina, I'll take a look
[13:41] <dobey> alecu: https://code.launchpad.net/~alecu/ubuntuone-storage-protocol/validate-ssl-cert/+merge/108962 doesn't have it :)
[13:42] <alecu> doh
[13:42] <dobey> i merged the change from there. :P
[13:43] <dobey> mandel: why removing the \ in some statements?
[13:44] <mandel> dobey, pylint was complaining that \ was redundant when used inside ()
[13:44] <alecu> dobey, I've pushed it now, sorry.
[13:51] <mandel> mmcc, dobey: improved tests for tcpactivation (this don't get stuck in mac/windows/linux: https://code.launchpad.net/~mandel/ubuntu-sso-client/fix-tcp-activation/+merge/109367
[13:52] <dobey> somehow i think i need t onot be doing reviews all day today
[13:52] <ralsina> alecu: thanks
[13:52] <mandel> dobey, I agree as long as you do mine hehe
[14:04] <alecu> ralsina, +1
[14:05] <ralsina> alecu: awesome, thanks
[14:11] <dobey> alecu: also pushed to my branch, and added a test for my change as well :)
[14:12] <alecu> dobey, great, I'll review it.
[14:12] <alecu> mandel, I've approved this branch: https://code.launchpad.net/~mandel/ubuntuone-dev-tools/darwin-cleanup/+merge/109363
[14:12] <alecu> mandel, but it has a typo, so please correct it before merging it.
[14:13] <mandel> alecu, great! does the change make sense?
[14:13] <mandel> alecu, sure, where is the typo?
[14:13] <alecu> mandel, it's in the proposal
[14:13] <mandel> alecu, ok, fixing
[14:13] <alecu> mandel, and the change makes a lot of sense: the problem was that the client was started connecting, but the server was not connected yet, so the protocol created was only the client's protocol.
[14:14] <alecu> mandel, and the server had no protocol because it didn't got to do that yet.
[14:14] <gatox> hello everyone!! back from the doctor, sooner as expected! :D
[14:14] <mandel> alecu, exactly.. it makes soo much sense but it was hard to find (or consider it)
[14:14] <alecu> mandel, it makes a lot of sense that different reactor have different timing issues related to this, so that's why the osx reactor was showing it much often.
[14:15] <alecu> "much often" "more often"?
[14:15] <dobey> more
[14:15] <alecu> "much more often"!
[14:15] <alecu> thanks :-)
[14:15] <mandel> often more much?
[14:15] <gatox> alecu, when you have some free time, we can have the mumble that we talk about
[14:16] <alecu> hi gatox, welcome back! Are you completely healthy now? :-)
[14:17] <gatox> alecu, i don't know jeje i'm going to know in a couple of days... i'm doing a complete check
[14:17] <dobey> alecu: no, he has a hairball stuck in his throat
[14:19] <mmcc> Hello, World!
[14:20] <gatox> dobey, jejeje
[14:20] <alecu> lolz
[14:21] <mandel> gatox, rectal exam included? O_o
[14:22] <mandel> mmcc, morning! I finally fixed the tests! I pointed the code for you a few mins ago
[14:22] <ralsina> briancurtin is internet-less, will try to get back ASAP
[14:23] <gatox> mandel, jejeje no
[14:24] <mmcc> thanks mandel! I'm looking at it now, I see you added a FactoryFactory, so I'm going to fire up the JavaBadgeFactoryAdapter and get you a JavaBadge
[14:24] <mandel> mmcc, hahahah yes is within the twisted naming convention
[14:25] <ralsina> alecu: oftenest
[14:26] <dobey> ralsina: you lose at scrabble
[14:26] <dobey> unless it's the valley girl edition
[14:26] <ralsina> dobey: but I win at LIFE
[14:27] <ralsina> if one reactor showed the behaviour more than the mac os one, it would show it oftenester
[14:27] <ralsina> and so on, keep tacking er/est
[14:27] <dobey> i think Ramses III still has you beat, in that game too
[14:28] <ralsina> dobey: he lost like thousands of years ago!
[14:29] <dobey> his burial chamber is bigger than your house :P
[14:29] <ralsina> dobey: also, now property of the egyptian government
[14:29] <dobey> always has been
[14:29] <ralsina> also, has bad cell reception
[14:30] <ralsina> is in a dead neighborhood
[14:30] <ralsina> I can do this all day, so I better stop now ;-)
[14:31] <dobey> it's more fun than doing reviews and staring at launchpad timeout errors
[14:31] <mmcc> hah, dead neighborhood. great property values though
[14:31] <dobey> and quiet
[14:31] <ralsina> mmcc: it's beachfront, right onthe sand
[14:31] <dobey> not to mention the view of stars at night
[14:32] <mmcc> granite countertops?
[14:32] <ralsina> expensive interior decoration
[14:32] <dobey> mmcc: it's a fixer upper, but back in the day, the whole exterior was covered in granite
[14:32] <mmcc> built to last
[14:33] <ralsina> pets allowed
[14:33] <dobey> also, you can see your house from space
[14:33] <ralsina> and viceversa
[14:34] <dobey> hey look, it's Sirius B minor
[14:34]  * dobey wonders how to make squid3 work as a transparent caching proxy
[14:34] <ralsina> no plumbing though. Not the place, not ramses
[14:35] <ralsina> dobey: I used to know how to do that
[14:35] <ralsina> dobey: just set the option and use iptables to re-route all 80/443 traffic to it
[14:36] <ralsina> dobey: and also, usually not worth it, specially with your bandwidth
[14:36] <mmcc> hey dobey, gatox: review run-tests for macos in controlpanel? https://code.launchpad.net/~mikemc/ubuntuone-control-panel/fix-1010211-run-tests-on-darwin
[14:36] <ralsina> dobey: the added latency offsets the modest speed gain
[14:36] <mmcc> dobey this should be familiar
[14:36] <gatox> mmcc, ack.... on it!
[14:36] <dobey> well i have it running, and set up, but even doing wget on that machine doesn't seem to alter the size of the cache dir
[14:37] <ralsina> dobey: hmmmm does anything appear on your storage log?
[14:37] <dobey> ralsina: i'm not doing it for speed.
[14:37] <dobey> ralsina: nope, nothing in cache.log or access.log
[14:37] <ralsina> dobey: the branch you reviewed for me today has a fix I would like to have on P, but is probably not worthy of a SRU. What should I do with it?
[14:37] <ralsina> dobey: then you are not using it
[14:37] <dobey> right. and i don't know why
[14:38] <ralsina> dobey: are you using wget on the proxy machine itself?
[14:38] <dobey> since i did the iptables
[14:38] <dobey> yes
[14:38] <ralsina> dobey: well, that's not going to work
[14:38] <mmcc> this morning I've tried using ctrl-a to switch emacs buffers, C-x o to switch tabs in Terminal.app, and shift-leftarrow in Chrome. I'm a mess.
[14:38] <dobey> well why not?
[14:38] <ralsina> dobey: the squid box will not have its trafic redirected, or no traffic would ever get out :-)
[14:38] <ralsina> dobey: unless you use the iptables process extension
[14:38] <dobey> iptables -t nat -A PREROUTING -i eth0 -p tcp --dport 80 -j REDIRECT --to-port 3128
[14:39] <dobey> is what i did
[14:39] <ralsina> dobey: in the router or in the squid box, or are they the same?
[14:39] <dobey> but i guess that only works connecting to the server on port 80
[14:39] <dobey> the squid box
[14:39] <dobey> which is not my gateway yet
[14:39] <ralsina> dobey: yes, if you are using https, that does nothing
[14:39] <mmcc> btw ralsina: reminder that I'm doing a half day today, and off Monday. I forget what I said about Monday when we talked on Tuesday...
[14:40] <ralsina> also, you used -i eth0 which means "incoming on eth0"
[14:40] <ralsina> mmcc: ack
[14:40] <alecu> gatox, mumble?
[14:40] <gatox> alecu, ack
[14:41] <dobey> ralsina: is there no good way to test it that doesn't involve immediately breaking the network?
[14:41] <ralsina> dobey: you want -t nat -A OUTPUT  but that will break squid
[14:41] <ralsina> dobey: nope
[14:41] <dobey> i guess i could boot a VM and set the gateway to that server instead
[14:41] <ralsina> dobey: ok, yes, a crossed cable and another box :-)
[14:41] <ralsina> or that
[14:42] <dobey> but will need to also set up iptables to masquerade all traffic first
[14:43] <ralsina> dobey: no need, if you redirect the tcp, you can still leave the rest of the traffic out without masquerading, if you want
[14:43] <ralsina> dobey: you can test it by only redirecting the traffic of wget
[14:43]  * ralsina looks for how to do that
[14:45] <ralsina> dobey: like, using -m owner --uid some-uid
[14:46] <dobey> oh hrmm. the vm fake interface thingy is a different subnet though
[14:48] <dobey> hrmm, and it doesn't like me trying to tweak the route
[14:50] <dobey> blah
[14:51] <mandel> mmcc, I also need a review from you for https://code.launchpad.net/~mandel/ubuntuone-fsevents-daemon/add-dispatcher-tests
[14:51] <mandel> please.. :)
[14:52] <mmcc> mandel, ok, I'll do that now
[14:53] <dobey> mmcc: hrmm, so on darwin, where are $u1trial/$u1lint originally set?
[14:53] <mandel> dobey, I still don't like that we use $u1trial and not just u1trial..
[14:53] <mmcc> dobey, they're set when you source a mac-env script from the windows-installer
[14:54] <dobey> mandel: and why do we do that?
[14:54] <ralsina> dobey: our status on bug #711413 is "WTF?" right?
[14:54] <mmcc> mandel, agreed, mac-env should just add path/to/u1trial to PATH
[14:54] <mandel> dobey, gatox has the explanation AFAIK
[14:54] <dobey> ralsina: oh is that the weird crash in qt?
[14:54] <gatox> mandel, just a sec.... on mumble
[14:54] <ralsina> dobey: no, it's the error connecting to DBus on SSO
[14:55] <dobey> ralsina: oh, the permission denied issue?
[14:55] <ralsina> dobey: right
[14:55] <dobey> (lp just times out for me)
[14:55] <ralsina> dobey: ubuntu-sso-login crashed with DBusException in __new__(): org.freedesktop.DBus.Error.NoServer: Failed to connect to socket /tmp/dbus-zPW5jjeWfI: Connection refused
[14:55] <dobey> yeah, though mvo was looking into it. i think it happens for some other apps as well
[14:55] <ralsina> I don't know if it's permissions or a race condition
[14:56] <dobey> i think it's a race
[14:56] <ralsina> yes, doesn't look like something that could be caused by SSO, really
[14:56] <ralsina> will ask mvo about it and see about reassigning then
[14:57] <mmcc> mandel, is SBJson OK to use licensing-wise? ISTR a question about that one too
[14:57] <dobey> launchpad is really unhappy with me this past week it seems
[14:58] <dobey> mmcc: what license is SBJson?
[14:58] <mandel> mmcc, I believe is not and that is why I'm changing to JSONKit which is BSD (https://github.com/johnezang/JSONKit)
[14:58] <mandel> dobey, ^
[14:59] <dobey> JSONKit is the wrong type of BSD
[14:59] <ralsina> dobey: I am getting a tarmac bounce on my branch :-/
[14:59] <ralsina> dobey: just retry?
[14:59] <mandel> dobey, and the apache license 2?
[14:59] <dobey> ralsina: fix the tests
[15:00] <dobey> mandel: apache should be ok for us to depend on
[15:00] <dobey> http://www.apache.org/licenses/GPL-compatibility.html
[15:00] <ralsina> dobey: eh?
[15:00] <mandel> dobey, JSONKit has dual license BSD or Apache, so we are ok, right?
[15:00] <ralsina> dobey: it's failing with a GIO error, and no tests fail
[15:00] <thisfred> me
[15:00] <dobey> twisted.trial.unittest.FailTest: ('setButtonLayout', (([0, 9],), {})) not in [('setButtonLayout', (([],), {}))]
[15:01] <dobey> ralsina: is what i see
[15:01] <dobey> ralsina:  there are 2 failing tests
[15:01] <mmcc> ME
[15:01] <gatox> me
[15:01] <mandel> me
[15:01] <dobey> ralsina: http://pastebin.ubuntu.com/1030495/
[15:01] <ralsina> dobey: weeeeelllll ok
[15:01] <ralsina> dobey: will fix then
[15:01] <ralsina> dobey: but I also see PASSED (skips=9, successes=820)
[15:02] <ralsina> Oh, the second pass. Silly me
[15:02] <dobey> ralsina: what are you looking at? the wrong test suite?
[15:02] <dobey> :)
[15:02] <dobey> meh
[15:02] <ralsina> dobey: runing run-tests on my box shows me the good one last
[15:03] <ralsina> which is strange
[15:03] <dobey> ralsina: then they probably all passed running on your box
[15:03] <ralsina> me
[15:03] <ralsina> dobey: makes no sense, will look at it anyway :-(
[15:04] <thisfred> alecu? whoelse?
[15:04]  * thisfred starts
[15:04] <dobey> brian has no internet
[15:05] <ralsina> go ahead thisfred
[15:05] <thisfred> DONE: bug #999585 bug #999574 TODO: bug #1009505, u1db sync integration tests BLOCKED: no NEXT: mmcc
[15:05] <mmcc>  DONE: packaging stuff, merged qt plugin bugfix, controlpanel tests running on darwin, many pass
[15:05] <mmcc>  TODO: polish off & propose setup-mac, work on controlpanel fails, half day
[15:05] <mmcc> BLOCK: none
[15:05] <mmcc>  NEXT: gatox
[15:05] <gatox> DONE:
[15:05] <gatox> Fix macfsevents to create a snapshot of the folder before returning the control to the caller. Half-sick day. Mumble with alecu
[15:05] <gatox> TODO:
[15:05] <gatox> Fix the remaining tests for u1-client filesystem_notifications
[15:05] <gatox> BLOCKED:
[15:05] <gatox> No
[15:05] <gatox> mandel, go
[15:05] <mandel> DONE: Fixed bug 1009071 and bug 1009408. Started working on integration tests for fsevents.
[15:05] <mandel> TODO: Finish integration tests. Start integration for fsevetns and u1-client. Write email for elopio about proxy settings.
[15:05] <mandel> BLOCKED: no
[15:05] <mandel> dobey, please
[15:05] <alecu> me
[15:05] <dobey> λ DONE: some triage, poking through the security fix stuff, branch for sec fix in protocol trunk
[15:05] <dobey> λ TODO: reviews, triage, finish releases/uploads, tarmac tweakery
[15:05] <dobey> ralsina
[15:05] <dobey> λ BLCK: None.
[15:05] <ralsina> DONE: dash call, team call, administrivia, merged sort-shares, fixing the extra back buttons (need to fix tests), studying, not BLOCKED
[15:06] <mandel> oh, and 1-1 with ralsina!
[15:06] <mandel> after this?
[15:06] <dobey> alecu
[15:06] <ralsina> mandel: I have 10 minutes of battery and will have lunch.
[15:06] <ralsina> mandel: is in 80 minutes ok for you?
[15:06] <dobey> lol. 1-1
[15:07] <dobey> guess we should do those or something
[15:07] <mandel> ralsina, would be close to my 7 pm..
[15:07] <alecu> DONE: team mumble, various mumbles, misc reviews, 1-1, ssl branches
[15:07] <alecu> TODO: moar ssl-related branches
[15:07] <alecu> BLOCKED: no
[15:07] <ralsina> mandel: then monday morning
[15:07] <mandel> ralsina, ok, I'll change the calendar
[15:07] <ralsina> mandel: in fact, let's reschedule it for monday morning from now on
[15:08] <mandel> ralsina, sure, you set the time then :)
[15:08] <alecu> gatox, shall we get back to mumble?
[15:08] <gatox> alecu, ack
[15:08] <ralsina> mandel: will do!
[15:09] <mandel> ralsina, cool :)
[15:10] <ralsina> dobey: I am concerned that obviously broken tests are not failing on my box
[15:11] <ralsina> dobey: and dn't be jealous, I will try to give you more 1-1s, it's just that I feel like I know what you are doing already :-)
[15:14] <dobey> firefox failbar strikes again
[15:14] <thisfred> is that an extension?
[15:14] <dobey> it's a default feature
[15:16] <thisfred> I'm using vimperator + lesschrome. what does the failbar do? ;)
[15:18] <ralsina> ok, 3 minutes of battery left. time to suspend. Will have IRC on phone, but don't expect me to type there ;-)
[15:18] <ralsina> and... luch
[15:18] <ralsina> luNch
[15:18] <dobey> thisfred: i think the official marketing term from mozilla is "awesomebar"
[15:19] <dobey> thisfred: but since it is in fact, not awesome, and more consistently fails and is horribly slow for me. i gave it a proper name
[15:20] <dobey> like, when i type "tw<downarrow><enter>" and end up on a google search page for "tw" rather than on twitter (which is the first result in the drop-down if i wait several seconds for firefox to catch up with my typing)
[15:26] <mmcc> mandel, this isn't new in this merge, but I just noticed that sometimes you spell "cookie" as "cookey"... is that on purpose? it's just in the COOKEY_KEY constant, but it's used lots of places
[15:27] <dobey> ok, need to get lunch. bbiab
[15:31] <mandel> mmcc, is me being an ass..
[15:32] <mmcc> mandel: heh, ok. don't care, just wondered if there was some weird name clash or something
[15:32] <mandel> mmcc, I mix eng with america a lot..
[15:33] <mmcc> mandel: also, for future reference, when comparing equality of two strings the docs say you should use isEqualToString: instead of isEqual: -- it's "faster"
[15:33] <mandel> mmcc, cool, can you add that as comments and I'll fix them :)
[15:33] <mmcc> (noticed that in ConfigMessage.m:49)
[15:33] <mmcc> ok
[15:33] <mmcc> I'm planning to, just thought that was worth mentioning in public too
[15:34] <mandel> :)
[15:38] <alecu> mmcc, also keep in mind that mandel is a bit slightly more than slight dyslexic. So don't worry about rejecting his branches when you find any of his orthographic nightmares.
[15:38] <gatox> mmcc, mandel back...... what happend?
[15:38] <mandel> mmcc, yes, I'm very dyslexic in all langs.. (includes eng, spanish and catalan) so I'm certainly not offended when this things are mentioned :)
[15:39] <mmcc> ah, ok mandel. I'll keep that in mind
[15:40] <gatox> mmcc, mandel ahhhhhhh i see...... i tried that..... but the problem was that with u1trial/u1lint inn the path (and not an alias pointing to the file), when you execute that, it tries to execute it with the python from the system, not the one in the buildout..... and obviusly that fails badly..... maybe i'm missing something.... but that was my conclusion after trying a lot of stuff
[15:40] <mmcc> oh right- becaues u1trial and u1lint have a #!/usr/bin/python
[15:41] <gatox> right
[15:41] <gatox> so now is "python $u1trial" and you ensure that you are going to execute u1trial with the proper "python"
[15:41] <mmcc> right
[15:42] <mmcc> maybe dobey can comment on how he'd like to fix that when he gets back from lunch...
[15:43] <mmcc> or *if* he'd like to fix it :)
[15:51]  * gatox lunch
[16:19] <mandel> mmcc, I'm off to enjoy the weekend and forget about twisted, please remember to review: https://code.launchpad.net/~mandel/ubuntu-sso-client/fix-tcp-activation/+merge/109367
[16:20] <mandel> so that it gets merge and sso has all tests green on mac os x
[16:20] <dobey> hrmm
[16:20] <mmcc> ok mandel, I will try to get to it, but I'm also leaving for the weekend in ~40 min and I'm still reviewing your other branch ...
[16:21] <mandel> ok
[16:21] <mandel> mmcc, no worried :)
[16:21] <mmcc> also will be gone Monday, which I mentioned but might have gotten lost
[16:22] <dobey> oh
[16:22] <dobey> mmcc: i guess we are installing stuff wrong in the buildout then
[16:23] <mmcc> dobey, how so?
[16:23] <mmcc> should u1lint be getting edited to change the shebang during installation?
[16:24] <dobey> mmcc: my understanding was that doing python setup.py install with a different python would cause the #! lines to be tweaked, when they are #!/usr/bin/python in the scripts
[16:24] <dobey> or whoever told me that didn't fully explain it
[16:25] <mmcc> dobey: aha, ok.  we're not doing python setup.py install here.
[16:25] <mandel> mmcc, is not a big deal because I can ask gatox to run the tests on mac
[16:25] <dobey> mmcc: how are we installing into the buildout?
[16:25] <mmcc> dobey, gatox's mac-env script ends up setting u1lint=/Users/mmccrack/Documents/Canonical/Source/buildout-env/scripts/devsetup/parts/ubuntuone-dev-tools/bin/u1lint
[16:25] <mmcc> (for me, obs)
[16:27] <mmcc> see http://bazaar.launchpad.net/~ubuntuone-control-tower/ubuntuone-windows-installer/trunk/view/head:/scripts/devsetup/env-mac
[16:27] <dobey> oh
[16:27] <dobey> it's not even installing stuff?
[16:27] <mmcc> well, it's not installing those two
[16:28] <dobey> well why the hell not :)
[16:28] <dobey> and he's at lunch
[16:28] <dobey> and brian is networkless
[16:29] <mmcc> well, the osx buildout googledoc has a step that says 'buildout install windows' (skip  this)
[16:29] <mmcc> maybe that's where it'll install the dev tools? and we need a mac version of that step?
[16:30] <mmcc> dobey: I just sent you an invite to that googledoc in case you are curious
[16:30] <mmcc> also, OT, but s/google docs/google drive/ <-- whoa
[16:31] <dobey> i am not sure it's properly installed on windows either
[16:32] <dobey> also, OT, but i ♥ my new server hardware
[16:34] <mmcc> why?
[16:36] <dobey> well, it's a 1U rack box, nearly slient, and it and my 24 port gigabit switch together, are only using about 28W of power on average
[16:36] <mmcc> not bad...
[16:37] <dobey> yeah. and when i upgrade the cpu (when the cpu i wanted to get, is more easily available), it'll drop down to probably about 10W
[16:37] <dobey> though, adding a raid 10 enclosure with 4 disks, might knock it back up a bit
[16:38] <mmcc> all this is in your house?
[16:39] <dobey> yeah
[16:39] <dobey> well, in my desk. it has rack space :)
[16:42] <mmcc> well alright then.
[16:42] <mmcc> I'd make a comment about office noise, but I have the worlds loudest AC vent just above my head
[16:42] <mandel> ok, EOW for me, have a great weekend!
[16:43]  * mandel goes to forget about twisted..
[16:43] <mmcc> bye mandel, see you tuesday
[16:43] <dobey> yeah, my hvac isn't the quietest either. and my workstation is really loud. but switching my server to this new hardware made a very noticeable difference in noise level
[16:46] <dobey> whoot!
[16:47] <dobey> ubuntuond-dev-tools MIR has been approved in quantal
[16:49] <lgc> Hi. I wonder why my two computers just won't sync.
[16:50] <dobey> what do the logs say?
[16:50] <lgc> dobey, where are the logs?
[16:50] <ralsina> dobey: who filed a MIR for that?
[16:51] <dobey> lgc: ~/.cache/ubuntuone/log/ assuming you're talking about the Ubuntu client
[16:51] <dobey> ralsina: i did, the other day
[16:51] <ralsina> dobey: ok!
[16:51] <dobey> ralsina: so we can fix the packages in ubuntu to run the tests :)
[16:51] <ralsina> dobey: awesome then!
[16:53] <lgc> dobey, anything too look for, especially?
[16:53] <dobey> lgc: syncdaemon-exceptions.log being non-empty, or errors mentioned in syncdaemon.log
[16:55] <dobey> gatox: can you review https://code.launchpad.net/~dobey/ubuntuone-storage-protocol/validate-ssl-cert/+merge/109362 please?
[16:55] <gatox> dobey, yes!
[16:55] <dobey> alecu: btw, did you also get to run those tests with libsoup you mentioned you would run yesterday evening?
[16:56] <alecu> dobey, no, didn't get around to doing that. I've got them on my todo for today.
[16:56] <lgc> dobey, plenty.
[16:57] <dobey> alecu: are you just doing all the branches for those fixes today?
[16:57] <dobey> lgc: plenty of errors in syncdaemon-exceptions.log ?
[16:58] <lgc> Yes.
[16:58] <dobey> lgc: can you pastebin a bit?
[17:02] <alecu> dobey, yes, and also I need to see about having jenkins do our windows builds with a patched twisted.
[17:03] <alecu> but right now I'm off to have some lunch.
[17:04] <dobey> alecu: ok. buen provecho
[17:04] <alecu> gracias!
[17:04] <lgc> dobey, be my guest: http://pastebin.com/tKB21hKq.
[17:05] <dobey> lgc: ah, i think you are hitting a small bug, which we have fixed and is waiting in SRU queue.
[17:05] <dobey> lgc: can you enable the precise-proposed repository, and try the package in there?
[17:05] <dobey> lgc: the ubuntuone-client in there, that is
[17:08] <gatox> dobey, +1
[17:12] <lgc> dobey, please bear with me. What is the precise-proposed repository?
[17:15] <dobey> lgc: https://wiki.ubuntu.com/Testing/EnableProposed
[17:15] <dobey> lgc: it is an additional repository where packages sit while being tested, before being pushed out as updates.
[17:16] <dobey> lgc: it's not enabled by default
[17:16] <lgc> Will the installation from these repositories take precedence over the others?
[17:17] <dobey> precedence is determined by version, not location. but yes, the ubuntuone-client package in there, is a newer version than in current precise
[17:17] <dobey> though it should appear as an update next thursday or so
[17:19] <dobey> mmcc: hrmm. i'm not sure i understand everything in the buildout.cfg exactly. what's the difference between [development] and [sources] ?
[17:20] <mmcc> dobey: I'm not sure - they're different steps in the instructions...
[17:20] <mmcc> looking
[17:20] <dobey> oh
[17:21] <mmcc> yeah, development installs dependencies and sources updates the sources
[17:21] <mmcc> so maybe you'd want to do them separately
[17:22] <mmcc> the notes say that 'sources' is optional, but IIRC some things assume that you do that step
[17:22] <mmcc> eg, by default my setup-mac.py looks for sources there, but it's configurable
[17:22] <dobey> on mac i guess because it requires the dev-tools bits to be there
[17:22] <dobey> at least, that's what i understand from the env-mac script
[17:23] <mmcc> I don't quite get what you're saying
[17:23] <dobey> probably because it's pulling the old tarball of dev-tools
[17:23] <dobey> the u1trial=$DEVPATH/ubuntuone-dev-tools/bin/u1trial bit
[17:24] <dobey> although, ubuntuone-dev-tools isn't listed in [sources] so also a bit confusing
[17:24] <mmcc> yeah, 'development' should lose the tarball and we should put dev-tools in sources
[17:24] <mmcc> that's what I had to do manually anyway
[17:24] <dobey> i'm not sure that's correct
[17:24] <dobey> well, it should lose the tarball. but just having it under sources probably doesn't help significantly
[17:25] <dobey> it's still not clear to me how exactly things get installed though
[17:26] <mmcc> why doesn't having it in sources help?
[17:26] <dobey> because it needs to be installed
[17:27] <dobey> having it in sources means we still just tweak the PYTHONPATH and run it from there, which isn't what we want
[17:27] <mmcc> I remember having to build it manually, that'd be nice to have automatic
[17:27] <dobey> at least, it's not what we want, if we want to get rid of the "python $u1trial" stuff
[17:28] <mmcc> so, we want to install it with --prefix=buildout/path/to/bin right?
[17:28] <mmcc> yeah, not sure how to do that with buildout, but there's got to be a way
[17:28] <dobey> well we want to do $buildoutpath/bin/python setup.py install --whatever
[17:28] <mmcc> yes
[17:28] <mmcc> might need an extra step after [sources]
[17:28] <lgc> dobey, I can't even find 'Edit Software Sources' on the Ubuntu Software Center.
[17:29] <mmcc> lgc, it's the last item in the "Edit" menu up in the menu bar at the top of the screen
[17:29] <dobey> what he said
[17:30] <lgc> (I'm suffering the comeback after a couple of years: I know no sh*t).
[17:30] <mmcc> if you're not used to the new desktop ui, you have to hover your mouse to see the menu items. I had the same problem!
[17:31] <mmcc> or tap the Super key, to bring up the dash and type 'sources'
[17:31] <mmcc> on my keyboard, Super is alt
[17:31] <dobey> crazy mac keyboards
[17:32] <mmcc> nope! Sun type 7
[17:32] <dobey> crazy oracle people
[17:32] <mmcc> what can I say, the control key is in the right place
[17:33] <lgc> mmcc,  thanks. I'm in sort of a panic here.
[17:34] <dobey> don't panic. or get a brown paper bag and slow your breathing down, at least
[17:34] <mmcc> lgc, if you're new to unity I found holding down the alt key to bring up the quick reference help is useful
[17:35] <dobey> not alt
[17:35] <dobey> super
[17:35] <mmcc> I'm afraid my keymapping is all messed up, dobey can you... thanks :)
[17:35] <dobey> which on most peoples' keyboards, is the windows key
[17:35] <mmcc> OK, on that note, I have to head out for the weekend.
[17:35] <mmcc> best of luck lgc
[17:35] <dobey> half day?
[17:35] <mmcc> yep
[17:36] <dobey> enjoy
[17:36] <mmcc> my parents are visiting
[17:36] <mmcc> bye
[17:36] <dobey> don't drink too much
[17:37] <lgc> mmcc, thanks. But I't getting the better of me. I've heard I can switch to a typical GNOME 2 feel. I'll make up my mind in the next few hours.
[17:38] <lgc> dobey, I already did the change in the reps. Do I just go sudo apt-get reinstall ubuntuone, or something like that?
[17:39] <dobey> lgc: sudo apt-get update && sudo apt-get install ubuntuone-client should do. there are other packages in the repository as well, if you don't want them then i'd recommend just doing the install of ubuntuone-client, and then disabling the repository and running apt-get update again
[17:40] <lgc> I'll do that.
[17:43] <dobey> and after ubuntuone-client is installed, do "u1sdtool -q && u1sdtool -c"
[17:46] <lgc> I stopped the daemon. Do I start normally now?
[17:51] <dobey> yes
[17:52] <lgc> It is working now, but I can almost swear I didn't start it.
[17:54] <lgc> dobey, my machine signaled it's uploading some file, but the graphical part of U1 is not around. Will it go berserk if I start it (since it's seemingly working already)?
[17:55] <dobey> lgc: if you mean the control panel, no. starting it doesn't do anything to the syncdaemon
[17:56] <lgc> Oh, I see.
[17:58] <lgc> dobey, is u1sd made to handle machine suspends (i.e., closing the lid)?
[18:00] <dobey> yes
[18:01] <dobey> it should resume fine after suspend
[18:01] <dobey> if you find any problems with it in that case, please file bugs
[18:07] <lgc> It's doing the job now. Thanks, dobey.
[18:08] <dobey> sure
[18:08] <dobey> glad it's working :)
[18:16] <gatox> alecu, do you want to review the macfsevents fork after the tests or right now?
[18:16] <dobey> alecu: can you finish up your review of my protocol trunk branch? would like to get it in/released today, in 3.99.0
[18:25] <dobey> gatox: care to review https://code.launchpad.net/~rye/ubuntuone-client/ignore-in-close-write-in-dirs-stable-3-0/+merge/106629 ?
[18:33] <gatox> dobey, on it
[18:36] <gatox> rye, the branch has some conflicts
[18:37] <dobey> gatox: you didn't try to merge it into trunk did you?
[18:37] <gatox> dobey, ahhhhhh stable
[18:37] <gatox> my bad
[18:38] <dobey> yep, stable-3-0. the same fix is already landed in trunk :)
[18:40] <alecu> gatox, I'll finish dobey's review, and then I can review the macfsevents branch
[18:41] <dobey> thanks alecu
[18:41] <gatox> alecu, ok.... i'm working on the tests right now
[18:41] <dobey> doh. just accidentally hit c-q in firefox :-/
[18:52] <lamalex> is there any way to check how far a long my file sync is?
[18:55] <dobey> lamalex: install rye's indicator-ubuntuone from his PPA, or install magicicada from universe, i think; or use u1sdtool --current-transfers and a calculator maybe
[18:58] <lamalex> oh there's a progress bar on my launcher icon cool
[18:59] <dobey> oh right, or look at the progress bar :)
[18:59] <ralsina> lamalex: you are a lucky man, that thing has been intermittently broken since it was created :-)
[18:59] <lamalex> ha
[18:59] <lamalex> today appears to be my lucky day
[18:59] <dobey> ralsina: nah, it's only broken in oneiric
[18:59] <dobey> ralsina: it should work just fine in precise
[19:00] <ralsina> dobey: or if you open u1cp
[19:00] <ralsina> dobey: because it hijacks the unity connection from sd, IIRC
[19:00] <dobey> oh
[19:00] <gatox> mmcc, ping
[19:10]  * alecu leaves for kinder
[19:46] <dobey> gatox: he's gone for the day. took a half day. and he'll be off on monday as well.
[19:46] <gatox> dobey, ahhhhh right, thx
[19:46] <dobey> can someone give a quick review of https://code.launchpad.net/~dobey/ubuntuone-storage-protocol/validate-ssl-cert-4-0/+merge/109422 please?
[19:47] <dobey> same change as just landed in trunk
[19:48] <gatox> dobey, on it
[19:48] <dobey> thanks
[19:49] <gatox> dobey, i already review the previous one
[19:49] <dobey> right
[20:14] <gatox> ok...... i'm off for today..... it seems that maybe this will be ready for monday! bye all!
[20:15] <dobey> bye gatox
[20:15] <gatox> have a nice weekend!
[21:11] <ralsina> EOW for me. See you all on monday.
[21:35] <dobey> ah crap
[21:35] <dobey> oh well. i am calling it a week. later all