/srv/irclogs.ubuntu.com/2011/09/22/#launchpad.txt

lifelessmmm00:13
wgrantWas this added as part of the inline comment stuff recently?00:13
wgrantI thought this used to work.00:13
wgrantAnd we recently started getting message page timeouts.00:13
wgrantThey appeared from nowhere on top of the timeout reports a couple of weeks ago.00:14
lifelessmwhudson: whats up ?00:19
Fudgehi looking for help to contact launchpad to ask for intervention or guidance on how to prevent a ppa from being deleted00:19
jelmerFudge: what is the problem exactly?00:24
Fudgejelmer  someone on vinux deleted our natty ppa00:24
Fudgei am not linked with launchpad but as i am a contributor to vinux ive come to ask, the owner hasnt asked and thinks it cant be  stopped and may have to get all our users to chagne the ppa to another00:25
Fudgechange00:25
Fudgei thought it may be possible to ask for the deletion to be stopped and to find out who marked it for deletion00:25
Fudgeor if i can no tbe told for tony the owner to be emailed in regards to it00:26
jelmerFudge: As far as I know PPAs can't really be undeleted00:26
Fudgeit has not been deleted yet, just marked for deletion00:27
jelmerFudge: One of the Launchpad help contacts might be able to help further. If they don't show up here, I would recommend filing a question about this: https://answers.launchpad.net/launchpad00:27
wgrantThere will be no help contacts for at least 12 hours, so that won't work.00:27
wgrantFudge: Which PPA?00:27
wgrantFudge: The files can be recovered for a few days after deletion, if required.00:27
Fudgevinux/natty-/ubuntu natty00:27
wgrantAnd we can sort of revive a PPA sort of.00:28
Fudgeits still there currently00:28
wgrantBut I really wouldn't trust a PPA if you have people deleting it randomly.00:28
wgrantBecause anyone who can delete it can upload to it as well.00:28
Fudgeits been raised with dev team to limit who can access that but no action has been taken yet00:28
Fudgeclearly this will make it happen correctly, im told its  not an easy thing to do as a user is asked several times to confirm the action00:28
wgrantThe dev team is unconcerned that all these people have root access on users' systems? That is a worry.00:29
Fudgehell yeah00:29
Fudgei think anyone who joins vinux project on launchpad gets access, no idea why owner set it up like that00:29
Fudgedoesnt seem as if it would be default00:29
wgrantDo you have a link to the PPA page?00:29
wgrantI don't see a PPA named natty-00:29
wgrantThere is natty-testing.00:30
Fudgemaybe its already gone?00:30
wgrantNo.00:30
Fudgeim not very good with launchpad, havnt learned all about it yet, btw im visually impaired00:30
wgrantWell, it may already but gone, but I'd still be able to see it.00:30
wgrantAnd I can't see it, so the name is probably wrong. Hmm.00:30
Fudgemay i show you my source url?00:30
wgrantSure.00:30
wgrantThat would tell me what I need to know.00:31
Fudgedeb http://ppa.launchpad.net/vinux/natty-/ubuntu natty main00:31
wgrantAh, there it is.00:31
wgrantThanks.00:31
Fudgethank you00:31
wgrantI was looking in the wrong place.00:31
Fudge:) i do that a lot loL00:31
wgrantIt's odd. If someone clicked the delete button, it should be gone from ppa.launchpad.net by now.00:33
wgrantI wonder if someone has just disabled it.00:33
wgrantI've asked a sysadmin to check what state it is in.00:33
Fudgethank you so much mate00:33
FudgeI emailed Tony who is in the UK the creator of the ppa to add me to the vinux team on launchpad so I can make enquries too as well as asked him what steps he has taken to prevent it from being deleted. but think he has done little00:34
wgrantFudge: Good news, it's not actually deleted, just disabled. Anyone in ~vinux can go to https://launchpad.net/~vinux/natty-/+edit and check the 'Enabled' flag to bring it back.00:35
Fudgewoohoo ill get someone onto it :D00:35
Fudgeare you able to see who marked it disablexd00:36
Fudgedisabled00:36
wgrantI can try to grep through our logs, but may not be able to tell for sure. I'll see what I can find.00:36
Fudgethanx00:36
wgrantFudge: Do you have any idea when this might have happened?00:38
Fudgelast couple of days I believe00:38
Fudgeis that narrow enough field00:38
Fudgetimeframe i mean00:38
wgrantYep.00:38
Fudgeout of luck wgrant ?00:49
Fudgewgrant  im a member as well now01:09
wgrantFudge: I can't find any requests that could have disabled that PPA in the last three days, unfortunately.01:10
Fudgemm i guess its not as simple as grep vinux | grep disable lol01:11
wgrantIt is pretty much that simple, but there are a *lot* of logs.01:11
FudgeI appreciate you trying01:11
Fudgewhen the main guy wakes we will look at who has access, is it possible to have members that can not make changes, such as only uplaod to testing ppa where it has to be tested and aprooved01:12
Fudgelike some kind of flag system based per user01:12
wgrantAnybody in the team that owns a PPA can upload to or disable or delete the PPA. However, you can use the API's newComponentUploader method (see https://help.launchpad.net/API/) to allow other people or teams to upload to particular PPAs, in addition to the owner.01:14
Fudgeah gr8801:14
mwhudsonlifeless: https://bugs.launchpad.net/launchpad/+bug/85595101:22
ubot5Ubuntu bug 855951 in Launchpad itself "hide comment action is displayed when it cannot be used" [Low,Triaged]01:22
wgrantmwhudson: Ah, heh, this is your fault, sort of.01:24
lifelesss/sort of/entirely/ :P01:24
wgrantmwhudson: We had memcached disabled on BugTask:+index01:25
mwhudsonwgrant: because of feature flags?01:25
mwhudsonwgrant: aah01:25
wgrantmwhudson: But your flag changes broke the pageid and team scopes.01:25
wgrantMust check if that's deployable yet...01:25
lifelessI've commented to this effect01:25
wgrantIt is deployable!01:25
wgrantBut let's QA a bit more first.01:25
wgrantStevenK: ^^01:25
mwhudsonwgrant: ah, i relanded my branch but didn't do anything to fix the pageid scope01:25
wgrantmwhudson: I didn't realise it was broken too until now :/01:26
mwhudsonoh well, it might be easy to fix01:26
StevenKwgrant: Working on it, qas is timing out a lot01:26
lifelessmwhudson: pageid scope was broken too ? howso ?01:27
mwhudsonlifeless: that's what wgrant just said01:27
wgrantlifeless: I don't know how, but it no longer works.01:27
lifelessI'm asking not asserting ;)01:27
mwhudsonit's the same sort of thing though, reading things out of the request before they are set01:27
lifelessI assumed that the bust scope for team broke everything01:28
mwhudsonno01:28
lifelessok01:28
lifelessso this is critical then01:28
wgrantTeamScope was broken by the move away from LaunchBag.01:28
lifelessmwhudson: want to rollback your new landing01:28
mwhudsonwgrant: no01:28
wgrantmwhudson: Oh?01:28
mwhudsonwgrant: it was broken by trying to find the person during featurecontroller construction01:28
wgrantWell, yes.01:29
mwhudsonrather than when the scope was checked01:29
wgrantWhich was done when LaunchBag was removed.01:29
mwhudsonwgrant: yes, but that's irrelevant01:29
mwhudsonreally01:29
mwhudsonbecause the user is added to the launchbag when request.setPrincipal is called01:30
mwhudsons/when/at the same time as/01:30
wgrantSure, but to remove LaunchBag you had to pass in the user from elsewhere, which meant working it out early.01:31
mwhudsonwgrant: no01:31
mwhudsonwgrant: i didn't want TeamScope to reference the request, that's why i read it early01:32
mwhudson(implicitly reference in this case, to be fair)01:32
lifelessmwhudson: -> -dev?01:32
mwhudsonlifeless: yes01:32
jo-erlendif I delete my PPA and then recreate it later, will it be completely gone, or will it remember the package versions, etc?02:12
mwhudsonyou won't be able to recreate it with the same name, iiuc02:13
jo-erlendoh02:13
jo-erlendhow do I mark a bug as fixed?02:43
RAOFSet the status to ?fix released?02:44
jo-erlend"how do I set a bugs status to fixed released"?02:44
RAOFBe logged in, click on the existing status, click on ?fix released? in the popup?02:46
jo-erlendah, there you go. Thanks. :)02:46
jo-erlendactually... When should I set it fixed release and when do I set it fix committed?02:46
jo-erlendmy fix has been merged into trunk of the project, but I don't think a new package have been created for it. Does that mean it's committed, but not released?02:48
RAOFThat's a project-specific policy.02:58
jo-erlendoh, ok.03:00
lifelessjo-erlend: many folk work this way:03:07
lifeless - if a developer should still know about it, it is not fix released (or any other 'hidden by default' status)03:07
jo-erlendwhich are hidden by default?03:12
lifelessfix released03:13
lifelessinvalid03:13
lifelesswontfix03:13
lifelessopinion03:13
lifelessI *think* thats it03:13
jo-erlendoh, ok, great. Thanks :)03:13
lifelessthe largest number of users of your bug tracker is your developers03:14
lifelesswith second largest being users :)03:14
lifelessso its a balancing act, but usually its better to optimise for developer time03:14
jo-erlendyes, that sounds reasonable :)03:15
=== stub1 is now known as stub
chrisccoulsonhi, i'm trying to set the contact address for ~mozillateam on launchpad to ubuntu-mozillateam@lists.ubuntu.com, but launchpad tells me that is already associated with https://launchpad.net/~ubuntu-mozillateam, which only seems to exist as a consequence of a package upload in gutsy08:43
chrisccoulsonis there any chance of rectifying that?08:43
xrg_Hello. What is the preferred setup for some automated bot pushing branches onto LP? with separate ssh keys? separate account?08:47
maxbchrisccoulson: We can merge the teams, but for some reason LP insists that you claim the ~ubuntu-mozillateam team before we can merge it08:51
chrisccoulsonmaxb, ok, just tried that. i'm waiting on an e-mail now though :)08:56
lagDoes anyone know what's going on with LP's unloading ability09:04
lagIt appears to be shot09:04
wgrantUnloading ability?09:05
lagI try to upload a tar.gz to "Download files for this release" milestone page09:06
lagIt uploads, then when it gets to 99% it says, "There as been a problem, please try again"09:07
lagAnd now I'm trying to dput into my PPA09:07
lagI get this:09:07
lagUploading to ppa (via ftp to ppa.launchpad.net):09:07
lag  Uploading linux-ux500_3.0.0-1000.0.dsc: done.09:07
lag  Uploading linux-ux500_3.0.0-1000.0.tar.gz: 98763k/98764k09:07
lagAnd it just hangs there with 1k to go09:07
lagThis has happened multiple times09:07
lag</rant>09:08
wgrantlag: Have you tried uploading to your PPA with SFTP instead of FTP?09:09
wgrantIt looks like some router between you and LP isn't playing nice with FTP.09:09
wgrantSFTP is a bit harder to mess with, so is more likely to be reliable.09:09
lagwgrant: How do I do that?09:12
wgrantlag: https://help.launchpad.net/Packaging/PPA/Uploading#Uploading_with_SFTP09:13
bigjoolshttps://answers.launchpad.net/launchpad/+faq/173809:14
lagWill that cover dput and the upload button on my milestone page?09:14
bigjoolsthe FAQ explains it better09:14
* lag looks09:15
wgrantlag: Release file uploads are via HTTP, and unrelated to FTP/SFTP.09:15
wgrantPerhaps something is timing your HTTP connection out?09:16
lagI don't know what - the uploader in the bottom right corner goes up to 99%09:16
bigjoolsit's a known problem with FTP, no idea other than to guess at dodgy router natting09:21
lagYes, that's made things much worse :)09:23
lagWhen it asks for login, is that my LP name 'lag', or one of my email addresses (that I use for SSO)09:23
wgrantYour LP name.09:24
lag  linux-ux500_3.0.0-1000.0.tar.gz:09:24
lag<hang>09:24
wgrantYeah, dput's SFTP implementation doesn't report progress information, unfortunately.09:24
lagAh09:25
lagThanks for the heads-up09:25
* lag waits09:25
xrg_up : how to push branches from a build machines to LP? with restricted ssh keys?09:52
=== matsubara-afk is now known as matsubara
=== EisNerd_ is now known as EisNerd
=== yofel_ is now known as yofel
=== Quintasan_ is now known as Quintasan
RFlemingGreetings and Salutations!13:50
RFlemingI need help deciding if LaunchPad is right for me.13:51
RFleminganyone available for a chat?13:51
dobeyRFleming: do you have a definition of what exactly is "right" for you?13:54
RFlemingdobey: I do.  I'm in private business, and need issue tracking on deploying a rather large app internally.  There are several contractors outside of our WAN who will need access to the issue tracking system.  We currently use bugzilla internally, but with me and a couple of other guys doing stuff with open source and familiar with launchpad... well we're wondering if it's possible to have a secure launchpad13:56
RFlemingproject for us to use externally.13:56
dobeyah ok. i don't think i can answer that then. i think you want to look at this: https://launchpad.net/+tour/join-launchpad13:59
RFlemingdobey: that's perfect!14:00
dobeyRFleming: particularly the box in the bottom left corner there :)14:00
RFlemingI have no problem paying for it :)14:00
RFlemingdobey: one other thing.14:00
RFleminginternally, we're using CVS and Bugzilla.  How can I install LaunchPad internally?14:01
dobeylaunchpad is really only designed to be run on launchpad.net afaik, and while i believe it is possible to run a local instance for testing; i don't think it is a supported means of use for production14:02
RFlemingahh14:03
dobeyi am not on the launchpad team, btw. i just use it very heavily :)14:03
dholbachhiya14:03
dobeyhi dholbach14:03
dholbachare launchpadlib hacking tips collected somewhere? :)14:04
dobeyi don't know if such a thing exists14:04
dobeythere is lp:lptools though :)14:04
dholbachI just need the number of bugs of a team, but something like launchpad.people["ubuntu-security-sponsors"].searchTasks(status=['Opinion', 'Invalid', "Won't Fix", 'Expired', 'Fix Released']) seems to cause an internal error it seems14:05
dholbachright, I could take a look at lp:ltools14:05
dholbachhttp://blog.launchpad.net/api/three-tips-for-faster-launchpadlib-api-clients was the only thing I found so far14:05
dobeydholbach: ah, it's probably causing the db query to timeout, since the query is probably very very large :)14:05
dholbachI nicked "int(collection._wadl_resource.representation['total_size'])" from somewhere and thought that'd help14:06
dobeydholbach: if you split that up into separate queries for each status, it might work better; but some of them might still timeout14:06
dholbachbut it might be the db query14:06
dobeydholbach: doing a query on the toplevel bugs object might also be more prudent14:07
dholbachdobey, what do you mean by toplevel bugs?14:07
dobeydholbach: https://launchpad.net/+apidoc/1.0.html#bugs but i guess it only has createBug() it seems :(14:08
dholbachah ok14:08
dobeydholbach: you might want to refine the searchTasks() also; with your query i think you're getting every bug related to that team in any manner; not sure if that's what you want14:10
dobeydholbach: omit_duplicates=True might be useful, for example ;)14:10
dholbachI'll try omit_duplicates=True - thanks dobey14:11
dholbachdoesn't work either, but at least I found out from https://dev.launchpad.net/Foundations/Webservice/Performance/CollectionLength that I can remove the 'total_size' code :)14:15
dholbachok, I guess it means, I need to go back to screen scraping for now14:15
ams_csI'm having lots of trouble with bzr integration in LP. Is there is known problem, or is it just me?14:19
ScottSanbarams_cs: What kind of problems?14:58
ams_csScottSanbar: the diff on branch and merge pages never appears, and (presumably) this leads to branch merges going unnoticed and the branches have to be tidies manually15:00
ams_csSee here: 7 branches "have not been scanned yet" .... https://code.launchpad.net/gcc-linaro15:01
ScottSanbarams_cs:  I see that.  What was the bzr command(s) you used that caused this?15:04
ams_csScottSanbar: nothing new or unusual - they're straight-forward "bzr push", sometimes with an explicit --stacked-on, perhaps, but probably not even that, in most cases15:06
ams_csScottSanbar: what is unusual, I'm told, is the size of the gcc repo15:06
ams_csbzr does not cope well with it, although it's far better than it was15:06
ScottSanbarams_cs: Ok.  Good luck.15:08
ScottSanbarams_cs:  Do you have a hint as to how to go forward to fix the problem, other than manually doing stuff, yet?15:09
ams_csScottSanbar: I'd guess that a bzr operation is crashing or timeing out15:10
ams_csbut I have no access to the logs, or whatever to know15:10
ScottSanbarams_cs: is there anyway to do the necessary automated merges, etc. locally then just push the final "pre-tidied" stuff up?15:11
* ams_cs does not understand15:12
ams_csThe problem is, besides the missing diffs, that when I do "bzr merge; bzr commit; bzr push" it doesn't detect the merge and update the branch status15:13
ams_csI have to go into the launchpad site, and manually alter the branch attributes15:13
ScottSanbarams_cs:  can you replicate the launchpad repo on your local PC in a local bzr "repo" then do all your merging and commiting locally, then just simply push up the pre-merged, committed (ie, "tidy'ed") stuff up?15:14
ams_csScottSanbar: of course, that's what I just said ....15:15
ams_csScottSanbar: I don't expect the site to do my merges for me!15:15
ams_csScottSanbar: I just expect it to notice I've done it, as it always has done, and remove the branches from the active branches page15:16
ScottSanbarams_cs:  Oh, I see how it works - so you make all your changes locally, then when you push them up, you expect bazaar to notice all the changes correctly and update the launchpad repo to reflect the changes you made on your local repo?15:17
ams_csScottSanbar: exactly so15:17
ScottSanbarams_cs:  Ok, thanks for helping me learn.  So, you have and everyone else working on this project has a complete copy of the launchpad repo on their local PCs?15:18
ams_cswhen it's working, it normally changes the merge proposal to state "Merged", and the branch status similarly15:18
ams_csyes, that's how bzr works15:19
ScottSanbarams_cs:  Ok, thanks again for the info.  So, if you are working on your local repo, and someone else has changed their repo and updated launchpad, how do you get their changes into your local repo?15:19
ams_csbzr merge would be one way15:21
ams_csor start again with bzr pull and reapply your patch15:21
ams_csbzr doesn't have "rebase" :(15:22
exarkunthere's a plugin15:22
ScottSanbarsounds like kind of a pain15:22
exarkunworks pretty well15:22
ScottSanbarwhat's the plugin name?15:23
ScottSanbarexarkun15:23
ams_csexarkun: yes, but it takes so long to run that it's actually quicker to do it manually, and once it trashed my repo15:23
exarkun"rewrite" I guess15:23
exarkunams_cs: That sounds pretty different from 'bzr doesn't have "rebase"'.15:24
ams_csexarkun: it has a plugin that I no longer trust or find useful15:24
exarkunams_cs: That's unfortunate.  You should have said that in the first place though.15:25
exarkunThen I could have kept ignoring the conversation :)15:25
ScottSanbarams_cs, exarkun: I can see where manually doing pulls could be a very good solution vs. an automated solution that might be non-transparent15:26
ams_csexarkun: to be fair to the rewrite plugin, it probably wasn't written with a project the scale of gcc in mind. It has 10's of thousands of files, and 100,000+ revisions, and makes bzr creak15:26
=== eyes_ is now known as EyesIsServer
exarkunams_cs: I am all for software that always works correctly, or at least fails without destroying data.15:27
ams_csScottSanbar: yes, there are always conflicts in the ChangeLog15:27
exarkun"there's a lot of data" is a bad excuse for failure15:27
ams_csexarkun: true. I don't know what bzr's excuse it for not scaling well. git works fast and well, but doesn't integrate with LP, more's the pity15:28
exarkunFortunately I've been lucky, it seems, with my use of the rewrite plugin, and haven't had issues (and I've only used it with small projects).  I really am sad to hear that it has problems, particularly repository corrupting problems.15:28
exarkunBut I am glad to learn about the issue, rather than continuing to be ignorant of it.  I'll think twice before I use rewrite in the future.15:28
exarkunAnd I look forward to the day when bzr doesn't corrupt itself so readily (I've had _plenty_ of that sort of problem with bzr in other contexts).15:29
ScottSanbarams_cs, exarkun:  Thanks for taking the time to help me learn.15:29
=== verterok` is now known as verterok
exarkunSpeaking of changelog conflicts, here's one solution to that problem, http://twistedmatrix.com/trac/browser/trunk/twisted/python/test/test_release.py#L1523 and http://twistedmatrix.com/trac/browser/trunk/twisted/python/_release.py#L60515:32
exarkunAnd an example of its use in practice, http://twistedmatrix.com/trac/browser/trunk/twisted/topfiles15:32
exarkunAnother solution is that it's possible to write custom merge algorithms for bzr to use.  It probably wouldn't be too difficult to implement a changelog merge algorithm that could deal with the conflict that always arises.15:33
ams_csthat's a nice feature!15:34
ScottSanbarhello-debhelper recipe fails in launchpad: https://launchpadlibrarian.net/80630411/buildlog.txt.gz https://code.launchpad.net/~ubuntu-branches/ubuntu/natty/hello-debhelper/natty16:02
czajkowskiAloha16:09
=== beuno is now known as beuno-lunch
=== matsubara is now known as matsubara-lunch
chmrrI'm one of the upstream maintainers of https://launchpad.net/rt/ -- https://launchpad.net/rt/+packages shows that request-tracker3.8 packages list rt/4.0 as upstream, instead of rt/3.8  What's the right way to fix this on launchpad, as I don't seem to have rights to change them myself?17:07
chmrrNevermind, #ubuntu-devel managed to set things right.17:08
=== beuno-lunch is now known as beuno
dobeyexarkun: or better solution is to just not maintain a separate changelog file in vcs; vcs log messages should be high enough quality to be in a ChangeLog too. :)17:34
exarkundobey: Perhaps.  But there are different audiences for the information.  Certainly a ChangeLog which duplicates the vcs history is a waste of effort.  A ChangeLog that is suitable for consumption by a slightly less technical audience can be valuable though.17:36
dobeyexarkun: are most of your issues with "bzr corrupting itself so readily" with conflicts in such files? i haven't had any real issues with bzr in quite some time17:42
exarkundobey: no, they're mostly with shared repositories17:42
exarkunconcurrent "bzr pull" processes stepping on each other and producing a state that can't be recovered from automatically17:43
dobeyah17:46
chmrrOut of curiosity, how much work is left to be able to import non-HEAD branches from git?  I know https://bugs.launchpad.net/launchpad/+bug/380871 is relevant, but only looks to be the backend part17:54
ubot5Ubuntu bug 380871 in Bazaar Hg Plugin "support for colocated branches" [Medium,Fix committed]17:54
=== matsubara-lunch is now known as matsubara
chrissbxHello. How do I access launchpad with git?19:04
TheEvilPhoenixchrissbx:  launchpad doesnt support git19:04
TheEvilPhoenixchrissbx:  they support bzr directly, but you can link to code in git, or request it be pulled into bzr19:05
chrissbx(How do I clone a repo at all? Can't find any url on https://code.launchpad.net/~jonls/redshift/trunk)19:05
chrissbxWell, I want to take a look at history etc.; I'm fine with converting it locally, what would be the way to go?19:05
cndis there a way to get ddebs built and published in ppas?19:07
chrissbxOk, so the answer is: git init --bare trunk.git && bzr fast-export --export-marks trunk.git/marks trunk/.bzr/ | (cd trunk.git/ && git fast-import --export-marks=marks )19:29
chrissbx(not sure about the marks file path, seems to have to be inside the git dir, but dunno why)19:29
chrissbxAnd the answer to the first question is: bzr branch https://code.launchpad.net/~jonls/redshift/trunk19:30
=== doko_ is now known as doko
dokospam at https://bugs.launchpad.net/bugs/65319619:51
ubot5Ubuntu bug 653196 in openjdk-6 (Ubuntu) "when i compile using javac programname.java in my ubuntu 10.04 yerminal i get the following error: plz help me out or mail me.." [Undecided,Expired]19:51
maxbhidden20:00
=== matsubara is now known as matsubara-afk
RFlemingGreetings and salutations!20:26
RFlemingDoes anyone know how long staging.launchpad.net will be offline?20:26
cndwgrant, I've been trying to find info on how to enable ddebs in ppa builds23:11
cndbased on bugs, it looks like you might know what's going on there?23:11
=== yofel_ is now known as yofel

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!