[00:01] <verterok> drascus321: the problem to get new tokens it's related to the fact that the service is changing the domain name, from ubuntuone.com to one.ubuntu.com...
[00:02] <verterok> drascus321: and the oauth client can't handle the redirect
[00:03] <verterok> drascus321: and sadly the url it's hardcoded :(
[00:03] <drascus321> so really no way to fix it then?
[00:03] <verterok> drascus321: I'm looking into it, give me a minute ;)
[00:04] <drascus321> Alright thanks!!
[01:05] <urbanape> statik, you around?
[04:01] <Brennydoogles> Hello
[04:02] <Brennydoogles> How is everyone tonight?
[04:04] <Brennydoogles> Is anyone actually here?
[13:49] <thisfred> aquarius: can you confirm that it's just me/my machine that's crazy and that in fact the tests in dc do *not* mess with the real user's couchdb? (by throwing in a pdb in the middle of a test or so , because the tests *might* be smart enough to clean up after themselves)
[14:00] <aquarius> thisfred, I'm confused by this
[14:00] <aquarius> what are you testing? trunk?
[14:01] <thisfred> aquarius: yes, have tested trunk
[14:01] <thisfred> yesterday, but I  assume nothing's changed
[14:02] <thisfred> my couch is still weird in that I seem to always start two, not one (both with start_master_couch, and in desktopcouch)
[14:03] <thisfred> aquarius: so maybe this, and the weird ini chaining and everything is just a screwy set up here
[14:03] <aquarius> ok, odd. I will give it a go shortly
[14:03] <thisfred> aquarius: I'm trying to migrate to my new laptop
[14:03] <thisfred> but our deps won't install without java5, which isn't in karmic anymore :S
[14:04] <thisfred> aquarius: there appears to have been a real problem in the .ini chaining, and adam has patched trunk and the 0.10 branch, but for me that doesn't help either
[14:04] <thisfred> aquarius: so when you're testing, could you also stuff an admin user in your system couch for good measure?
[14:11] <ia> hello. ubuntuone devels, could you look at #426804 bug, please? does this bug is correct, or reason somewhere else?
[14:23] <dobey> ia: i believe it's been fixed already in trunk :)
[14:54] <thisfred> aquarius, CardinalFang: I see in desktopcouch.records.server in CouchDatabase.__init__:
[14:54] <thisfred>             desktopcouch.find_pid()
[14:54] <thisfred>             port = desktopcouch.find_port()
[14:54] <thisfred> should that not now be:
[14:54] <thisfred>             pid = desktopcouch.find_pid()
[14:54] <thisfred>             port = desktopcouch.find_port(pid=pid)
[14:54] <thisfred> ?
[14:55] <aquarius> ooooo.
[14:55] <aquarius> perhaps so, yes indeed
[14:55] <thisfred> Maybe merge order of branches screwed this up
[14:55] <thisfred> but it would 'splain  my current pickle
[14:56] <aquarius> although...if you call find_port with no pid, it calls find_pid itself
[14:56] <aquarius> so that shoudln't be the problem. Probably
[14:56] <aquarius> Having it print out the pid it's found might be good though :(
[14:56] <aquarius> it looks at local_files.FIND_PID
[14:57] <aquarius> er, FILE_PID
[14:57] <CardinalFang> thisfred, Riger..
[14:57] <aquarius> having local_files print that out would show which one it's reading
[14:57] <CardinalFang> That's probably the problem.
[14:57] <thisfred> aquarius: I'll run my local tests (for local people) with that.
[15:00] <dobey> o/~ lookin in my gucci, it's about that time o/~
[15:01] <aquarius> after the meeting begins line, we should paste a line of everyone we expect to attend so everyone gets notified, I reckon
[15:01] <jblount> aquarius: CardinalFang has a watch thing set up in his IRC for "MEETING BEGINS" in this channel
[15:02]  * dobey doesn't want that responsibility. has enough already
[15:02] <jblount> aquarius: But that's a good idea also :)
[15:02]  * aquarius pokes around to work out if xchat-gnome can do that.
[15:02] <CardinalFang> "highlight words"
[15:02] <jblount> MEETING BEGINS
[15:02] <jblount> Hello! Please say "me" if you are here for the Desktop+ meeting. Format is DONE / TODO / BLOCKED
[15:03] <aquarius> done it. rawk.
[15:03] <aquarius> good one xchat-gnome
[15:03] <CardinalFang> me
[15:03] <jblount> me
[15:03] <aquarius> er, me :)
[15:03] <teknico> me
[15:03] <rodrigo_> me
[15:03] <urbanape> me
[15:03] <dobey> me
[15:04] <aquarius> vds?
[15:04] <vds> me
[15:05] <jblount> CardinalFang: Your it, I think :)
[15:06] <CardinalFang> DONE: Committed test code for couchdb_io.  Now approaching one huge patch, which is bad.
[15:06] <CardinalFang> TODO: Get it reviewed and included today.
[15:06] <CardinalFang> BLOCKED: None
[15:06] <CardinalFang> jblount, want mudkips?
[15:06] <jblount> DONE: Work on template tags to accomplish #402837
[15:06] <jblount> TODO: Still finishing header refresh and tabs, need to keep working on  #402837 and talk to teknico about front end stuff
[15:06] <jblount> BLOCKED: Nope
[15:06] <jblount> aquarius: Tag!
[15:06] <aquarius> ⚀ DONE: allow unpairing of servers (bug #419975); got U1fs working on my machine :)
[15:06] <aquarius> ⚁ TODO: get dc-records-oauth tested and merged into trunk (bug #415375); fix UnknownLoginError and make it be known (bug #376087); turn on oauth for desktopcouch by default (bug #416413); talk to thisfred and CardinalFang about that last one; test android phone sync
[15:06] <aquarius> ⚂ BLOCKED: well confused by the big bzr format change thing for ubuntuone; can't replicate thisfred's start-up-the-wrong-couch problem
[15:06] <aquarius> ⚃ BUG COUNT:https://bugs.edge.launchpad.net/~sil/+assignedbugs?field.tag=ubuntuone-karmic - 4
[15:06] <aquarius> t to the e to the k to the teknico
[15:06] <teknico> DONE: check mattg's problem with multiple contacts, upgrade branches to 2a repo format, reviews
[15:06] <teknico> TODO: more reviews, talk with jblount (!) about the new contacts web ui, implement adding and editing contacts
[15:06] <teknico> BLOCK: none
[15:06] <teknico> next: rodrigo_
[15:06] <rodrigo_> • DONE: Working oauth support in couchdb-glib. IM addresses in top level contact record
[15:06] <rodrigo_> • TODO: Start upstream discussion for adding social services accounts config to about-me. Talk to Ara about writing mago tests for evo-couchdb. Propose couchdb-glib/evo-couchdb for GNOME 2.29. Store UUIDs for postal addresses. Conflict resolver tool in pair tool. Look at becoming a MOTU (https://wiki.ubuntu.com/UbuntuDevelopers). openSUSE/Fedora packaging with aquarius. More tomboy syncing fixes.
[15:06] <rodrigo_> • BLOCKED: no
[15:06] <rodrigo_> next urbanape
[15:06] <urbanape> DONE: Finally got Bindwood speaking OAuth. Waiting on branch merge with trunk. Started a branch for complete sync of bookmarks.
[15:06] <urbanape> TODO: Finish new branch that ensures complete sync on start, and make sure we don't back-propagate nulls on pull. Investigate whether we can use the batch processing.
[15:06] <urbanape> BLOCK: None, thank the maker.
[15:06] <urbanape> dobey: hit it
[15:07] <urbanape> (not sure what the bug number is looking up)
[15:07] <dobey> ☺ DONE: Worked on #403243 #419365
[15:07] <dobey> ☹ TODO: Finish #403243 #419365, Fix #397331, Release 0.95.0
[15:07] <dobey> 〠 BLCK: None.
[15:07] <dobey> vds: ciao
[15:07] <vds> DONE: landed branch to improve funambol exchange logs, investigated problems with the logs in production, discussed with the others about security in couchdb
[15:07] <vds> TODO:
[15:07] <vds> BLOCKED: no
[15:07] <vds> and I guess that's it
[15:07] <CardinalFang> aquarius, I can help with android sync test!
[15:07] <teknico> vds, wow, lucky you ;-)
[15:07] <vds> cool
[15:07] <aquarius> CardinalFang, are you running android 1.5?
[15:07] <CardinalFang> aquarius, Yes.
[15:08] <aquarius> CardinalFang, and the funambol android client didn't, e.g., delete all your contacts or wipe the phone or anythning? :)
[15:08] <urbanape> Actually, I don't think we had a bug for enabling OAuth in Bindwood. Shame.
[15:08] <CardinalFang> I have not tried it.
[15:09] <aquarius> ah. How then is your help to be manifest? :)
[15:09] <CardinalFang> I have a phone.  I have emulators.  I am willing to back up my stuff and install.
[15:09] <aquarius> ah, OK. I plan to do something similar -- need to back up my contacts
[15:10] <dobey> UnknownLoginError: Unbeknownst to user, but knownst to curl, there was a problem.
[15:10] <aquarius> dobey, yeah, that's what I'm going to look at next, but I have to go through a million confusing bzr things first, as I understand it
[15:10] <aquarius> CardinalFang, I am less impressed with the automatic sync-to-google thing than I might be, since it occurred to me that if I screw up the contacts it will merrily sync the screwed ones to google.
[15:11] <dobey> yeah
[15:12] <CardinalFang> aquarius, Roger.  I will test in a sandbox first.  Then, export contacts out of gmail and try on physical phone.
[15:12] <dobey> i've got plenty of bzr upgrading to do as well
[15:12] <thisfred> aquarius: I have found a syncml export app for android
[15:12] <thisfred> aquarius: have not gone much further than installed it
[15:12] <CardinalFang> gmail -> contacts -> import/export
[15:12] <thisfred> aquarius: VCard IO be its name
[15:13] <thisfred> CardinalFang: ah, yes
[15:13] <thisfred> probably massively easier
[15:13] <aquarius> thisfred, ah, I've got that. I bet it doesn't export all my linking-people-to-facebook adn so on though :(
[15:13] <dobey> ugh
[15:13] <urbanape> aquarius, I don't see anything about dbus and bindwood in your branch, I just see the diff that I submitted as well.
[15:14] <urbanape> ah, I see it
[15:14] <aquarius> urbanape, it was a *minute* fix :)
[15:14] <urbanape> yeah, but since my branch hadn't landed, your branch included all the rest of the diff.
[15:15] <aquarius> yeah, I know
[15:16] <urbanape> anyhoo, does anything need to happen in the keyring for that to take effect? Do I need to delete my existing entry?
[15:17] <urbanape> doesn't look like it
[15:18] <aquarius> no. It's purely there so the prompt which says "this application wants access to the keyring: Bindwood" actually says "Bindwood".
[15:18] <dto> is ubuntu one broken on Karmic Koala? i made an account, but can't seem to get anywhere with the applet. it crashes after starting.
[15:18] <aquarius> rather than "Application"
[15:19] <dobey> dto: can you elaborate on the meaning of 'crashes'? :)
[15:20] <dto> sure. the icon comes up in the tray, is clickable, can go to preferences etc, but the icon shows a red "can't connect" type of box on it. then a moment later i get a dialog box saying ubuntu one crashes.
[15:21] <dobey> please report the issue then :)
[15:21] <dobey> it should work, but maybe it's having a problem on your system
[15:24] <urbanape> aquarius, coolio. I'll give it a thumbs up.
[15:25] <dto> the bug is apparently already on there. seems like time and date settings may have something to do with it.
[15:26] <dobey> oh
[15:26] <dobey> your time is off > 5 minutes?
[15:27] <dto> hmm. i apparently have no timezone and the time+date dialog is crashing
[15:27] <dobey> weird
[15:27] <dobey> i suppose that could cause problems, yes :)
[15:29] <dto> brb
[16:04] <dto> well, fixing my timezone and NTP time didn't help with ubuntuone crashing on karmic. i guess i will have to wait for now
[16:05] <CardinalFang> dto, how does it crash?  If you run it in a terminal, what happens?  What output?
[16:06]  * jblount waits patiently while running fsck and hoping his machine isn't borked.
[16:08] <CardinalFang> jblount, what happened?
[16:08] <jblount> CardinalFang: Unsure, random abort on this vm dumped me to a root shell asking me to fsck
[16:08] <dto> CardinalFang: http://paste2.org/p/419851
[16:09] <dto> CardinalFang: the Application Problem dialog box comes up right after ubuntuone starts
[16:10] <CardinalFang> dto, do you have a  ~/.config/ubuntuone/ubuntuone-client.conf  file?
[16:12] <dto> yes.
[16:12]  * Chipaca looks at the paste
[16:13] <dto> http://paste2.org/p/419854
[16:13] <Chipaca> dobey: ping? dto is getting "ttributeError: 'NoneType' object has no attribute 'key'" in storageprotocol.oauth
[16:13] <CardinalFang> dto, run "seahorse", look in "passwords" for "UbuntuOne token for ...".
[16:14] <dto> nothing like that in the list.
[16:14] <CardinalFang> Ah, good.
[16:14] <dobey> Chipaca: something is trying to create a token from an empty string then i guess
[16:14] <CardinalFang> dto, what kind of things are in that list?
[16:15] <Chipaca> dobey: ah, this isn't something you've touched? then it's probably me :)
[16:15] <dto> pop and smtp passwords, probably from evolution. network secret for my wireless.
[16:15] <Chipaca> dto: could you file a bug, with the log?
[16:15] <dto> well i don't recall there being any place to enter my ubuntuone account info
[16:15] <dobey> Chipaca: it's probably due to the domain name move
[16:15] <dto> somebody else already filed a basically identical bug.
[16:15] <CardinalFang> dto, right.  That's what we need to know, I think.
[16:15] <dobey> dto: what version of the ubuntuone-client-gnome package do you have?
[16:16] <Chipaca> dto: have you ever used it before?
[16:16] <Chipaca> dto: or is this the first time?
[16:16] <dto> first time. i set up the account the other day so i could try it.
[16:16] <dobey> well it obviously failed to get the token :)
[16:17] <dto> dobey: 0.94.0-0ubuntu1
[16:17] <dto> i notice ubuntuone-client-tools is not installed
[16:17] <dto> should it be?
[16:17] <dto> brb
[16:17] <dobey> no
[16:18] <dobey> probably an issue with the moving of the domain name
[16:18] <dobey> dto: should be fixed in 0.95.0 though
[16:19] <dto> is there a ppa i can use in the meantime to get it working?
[16:20] <CardinalFang> dto, What happens when you run  xdg-open http://ubuntuone.com/  ?
[16:21] <dto> the webpage opens.
[16:22] <CardinalFang> Okay, good.  I thought maybe you had the system setting for web-browser wonky, perhaps.  That's good.  That means it's all our fault.  We'll try to get it fixed soon.
[16:22] <dto> ok :)
[16:23] <dobey> dto: it's not fixed yet in trunk, so no :-/
[16:24] <dobey> CardinalFang: no, web browser failure should pop up an error dialog on its own :)
[16:24] <dobey> CardinalFang: this is just failing to get the request token, because it's getting a redirect which it's not handling
[16:53] <ia> can anyone tell me, please, what dns_srv.default record should be in /etc/xdg/ubuntuone/syncdaemon.conf instead of default _https._tcp.fs.ubuntuone.com (since canonical change domain name) for works?
[17:25] <CardinalFang> ia, Hrm.  statik?
[17:25] <CardinalFang> $ host -t ANY  _https._tcp.fs.one.ubuntu.com
[17:25] <CardinalFang> Host _https._tcp.fs.one.ubuntu.com not found: 3(NXDOMAIN)
[17:29] <statik> CardinalFang, smells like a mistake in the domain move or in the DNS config for file sharing. I bet pfibiger and lucio or dobey can figure out what to fix
[17:30] <CardinalFang> I don't know if SRV records are allowed to point to radically different names, but those would be good to keep around if they are.
[17:33] <pfibiger> statik: i had mentioned SRV records to mthaddon during the switchover, we should probably ask him.
[17:33] <pfibiger> err, CardinalFang: ^^
[17:34] <mthaddon> pfibiger: I don't think we switched over fs.one.ubuntu.com, only fs-1, fs-2
[17:34] <mthaddon> pfibiger: we were told it wasn't used any more
[17:34] <pfibiger> ah ha.
[17:35] <statik> right, we dropped api.ubuntuone.com, it was probably my mistake about fs.ubuntuone.com then
[17:35] <pfibiger> verterok: ping.
[17:35] <CardinalFang> mthaddon, I think you should update the contents of those records, and leave some at ...fs.ubuntuone.com .
[17:35]  * CardinalFang reads up on SRV.
[17:36] <mthaddon> but fs. points to fs-1 - what's that actually used for? would prefer not to have traffic going to one specifically when it's not using the specific hostname
[17:37] <mthaddon> I mean the current DNS for fs.ubuntuone.com actually just points to the same IP as fs-1.one.ubuntu.com
[17:37] <CardinalFang> mthaddon, We're talking about service discovery records.  SRV type.
[17:38] <CardinalFang> mthaddon, and the names inside must be canonical names, not CNAMEs or anything.
[17:38] <statik> i'm not sure which name the syncdaemon uses to lookup what storage server to connect to, verterok knows though
[17:39] <mthaddon> CardinalFang: yeah, but I'm just questioning what's looking for fs. vs. either fs-1. or fs-2.
[17:39] <thisfred> hmm: CardinalFang: the aforementioned fix in desktopcouch does not seem to do anything, actually. Tests on my machine seem to talk to the real desktopcouch couchdb no matter what I do. I can't figure out what changed between friday and yesterday that caused this.
[17:39] <verterok> pfibiger: pong
[17:39] <CardinalFang> mthaddon, Well, we would never have a _https._tcp.fs-1.one.ubuntu.com, right?
[17:40] <CardinalFang> mthaddon, We look up  _https._tcp.fs.one.ubuntu.com and find out about fs-1, fs-2 and maybe eventually fs-41826341823
[17:41] <verterok> statik, pfibiger: as CardinalFang points ^ _https._tcp.fs.ubuntuone.com is used for ther srv lookup
[17:41] <pfibiger> verterok: i was just looking for your input on the above. :)
[17:41] <verterok> s/is/was//
[17:42] <verterok> pfibiger: I didn't changed that in my update-domainname branch as it wasn't listen in the email  :/
[17:42] <mthaddon> CardinalFang: ah, okay, so we do need fs.one.ubuntu.com
[17:42] <verterok> pfibiger, CardinalFang: so ATM, we still use _https._tcp.fs.ubuntuone.com for the srv lookup
[17:44] <verterok> pfibiger: should I change it to _https._tcp.fs.one.ubuntu.com ?
[17:45] <pfibiger> verterok: i guess that yes, as soon as fs.one.ubuntu.com is set up by the losas, then we could change it, right?
[17:45] <mthaddon> pfibiger: can has RT?
[17:45] <pfibiger> mthaddon: yes.
[17:45] <mthaddon> thx :)
[17:46] <verterok> pfibiger: I can change it in the branch as it wasn't landed yet
[17:46] <verterok> pfibiger: anyway, it's just a config change :)
[17:46] <pfibiger> verterok: ok, cool, we'll just have to make sure it is set up dns-wise before we roll that out :)
[18:15] <ia> i'm sorry for annoying, but i still can't get connect with _https._tcp.fs.one.ubuntu.com and _https._tcp.fs.ubuntuone.com either :-/
[18:16] <dobey> ia: hrmm?
[18:16] <dobey> ia: those aren't valid hosts... they are SRV records for resolving to a valid host
[18:17] <CardinalFang> ia, I don't think those exist yet, anyway.
[18:17] <dobey> oh they exist
[18:17] <dobey> oh
[18:18] <dobey> the one.ubuntu.com ones don't
[18:19] <ia> dobey: AFAIU, syncdaemon.conf must contains correct dns_srv.default option for ubuntuone-client syncing, right?
[18:20] <dobey> ia: _https._tcp.fs.ubuntuone.com resolves correctly still
[18:20] <dobey> i can't actually test anything at the moment though, because bzr is using up most of the resources on my machine
[18:21] <CardinalFang> ia, it still resolves, but I think we've decided that the HTTP-level redirection doesn't work well for the client code.
[18:21] <dobey> redirection isn't the problem with that one :)
[18:21] <dobey> CardinalFang: the storage protocol doesn't use http :)
[18:22] <CardinalFang> dobey, what does the "https" mean then, in  _https._tcp.fs.ubuntuone.com ?
[18:23] <dobey> CardinalFang: we use port 443, so we can do some funky stuff to poke through proxies
[18:29] <Chipaca> CardinalFang: read up on SRV records, it's really instructive :)
[18:29] <CardinalFang> Chipaca, yeah, I did.  What in particular are you talking about?
[18:31] <Chipaca> maybe I misunderstood the conversation, I thought you were stumped as to why there was an _https in there
[18:32] <firfin> quick question i couldnt fin in faq ... what if 'add this computer' doesnt come up after first launching ubuntuone?
[18:32] <Chipaca> mine was a fly-by commentary (the worst kind, maybe?)
[18:33] <joshuahoover> firfin: We are having an issue currently where adding your computer does not work, it's related to the change in domain names
[18:33] <firfin> joshuahoover: ah okay .. thanks for your answer .. any page i can see the status/progress on this issue?
[18:34] <joshuahoover> firfin: let me try to find it for you
[18:34] <dobey> oh, looks like my cpu is available again
[18:36] <firfin> joshuahoover: I could just try again later. Any clue as to how much later would be enough info for me. Not that I don't appreciate you taking the time. :-)
[18:38] <CardinalFang> firfin, if it's not fixed in 48 hours, I would be astonished.  Likely before tomorrow, even.
[18:38] <joshuahoover> firfin: should be going into tomorrow's release
[18:39] <firfin> CardinalFang: joshuahoover: ok, thank to the both of you. I'll be trying again tomorrow
[18:39] <joshuahoover> firfin: yes, you should do a system update, install the client upgrade and then try again :)
[18:39] <CardinalFang> firfin, If it's not working by Friday, come back and make loud noises.
[18:40] <firfin> lol, I will try
[18:40] <CardinalFang> firfin, watch for updates.
[19:33] <rmcbride> FWIW, new test clients are building in teh Nightly PPA now. Once those are done I'll do a quick sanity on them before posting to the Beta PPA.
[19:36] <joshuahoover1> rmcbride: very good!
[19:41] <rmcbride> lpia packages won't start for 37 minutes, but that's not as bad a delay as we normally get :)
[19:45]  * dobey wonders why lpia still gets built for anything
[19:47] <rmcbride> I dunno about that but Im not waiting for them to get built before I test the others.
[19:47] <rmcbride> Won't be able to copy the packages until all platforms are finished buildign though
[19:49] <dobey> sure. but less platforms == faster builds :)
[19:54] <CardinalFang> jdo, Hi!  How far off is the  http...api/couchdb  data?
[19:58] <jdo> CardinalFang, its in the review queue
[19:58] <CardinalFang> jdo, Rgr.  I'll review!
[19:59] <CardinalFang> Oh, gosh.  You are not James Đỗ.
[20:06] <CardinalFang> Hrm, I will just assume that  os.path.join  will concatenate URL pieces.  The day we use Windows will be frightening anyway.
[20:31] <rmcbride> as the change in topic should indicate, new BETA clients should be in the PPA in mere moments. Enjoy
[20:48] <joshuahoover1> thanks rmcbride for getting that packaged up and put out there!
[22:11] <statik> hey urbanape, could you set the commit message on your pending bindwood branch?
[22:11] <statik> i set the one on aquarius branch
[22:19] <urbanape> sure thing
[22:20] <urbanape> done
[22:20] <urbanape> running out for a bit. I'll be back after dinner and Lex's bath.