/srv/irclogs.ubuntu.com/2009/03/20/#bzr.txt

poolieglyph: i don't see it  there00:04
poolieigc, hi00:04
=== lifeless_ is now known as lifeless
=== TheMuso_ is now known as TheMuso
lifelessjam: I think you should land your gc change00:38
lifelessjam: I haven't fully reviewed, but its definitely conceptually fine00:39
RaceConditionis there a way to, instead of unshelved changes, commit shelved changes only?01:06
lifelessRaceCondition: shelve you current changes, unsehlve the ones you want to commit, commit. ?01:09
RaceConditionlifeless: um, yeah, but you didn't answer my questio01:09
RaceConditionis it possible or not to commit shelved changes01:10
RaceConditionsmth like bzr commit -m "msg" --shelved01:10
RaceConditionapparently not, though, google gives me nothing on this01:11
lifelessno there is no direct way to do this today01:12
spivRandom observation: bzrlib.errors.AmbiguousBase has been deprecated since 0.801:15
spivWe should delete it...01:15
pooliespiv, i have a branch outstanding that removes lots of old code01:28
pooliei have review comments from robert still to do01:28
=== ampelbein is now known as Ampelbein
jamlifeless: streaming or the change to the delta_index code?02:16
abentleythumper: pong02:19
thumperabentley: I was wanting to talk to you about the bzr sprint at uds02:20
abentleythumper: sure thing.02:20
abentleythumper: did you mean on skype?02:21
thumperabentley: :) yes02:22
* igc lunch02:31
=== abentley1 is now known as abentley
lifelessjam: both03:13
=== abentley1 is now known as abentley
=== Ampelbein is now known as ampelbein
thumperpoolie: what do we need to do to kick the backport effort in ubuntu?04:24
pooliejames_w said something in bne about driving it04:26
pooliebut he's on holiday until monday04:27
thumperbne?04:27
thumpernm04:27
thumperfriday afternoon and I'm knackered04:27
igccontent filtering patch submitted to pqm04:29
* igc off to hospital for the rest of the day04:29
igchave a good weekend all04:29
lifelessigc: good luck04:36
macobzr push needs some better error messages04:38
maco"bzr push lp:~maco.m/seahorse-plugins" tells me that ~maco.m isnt a valid project name instead of telling me i need a branch name on the end of it04:39
macomaybe this is launchpad-specific though....? should i bug them instead?04:39
lifelessmaco: its launchpad specific04:44
macook ill go bug them04:44
lifeless:)04:44
poolieidea:06:06
pooliemaybe we could take all these people who like bikeshedding06:07
poolieand rent them out to actual power plants that need their bikesheds painted?06:07
pooliestrictly one per shed06:07
lifelesshopefully with big bikesheds06:07
pooliemm06:08
=== r0bby_ is now known as r0bby
pooliekfogel: don't you sleep? :)06:50
kfogelpoolie: long story :-)06:51
kfogelpoolie: I'm actually an early-to-bed person, but... sigh, something came up, so here I ma.06:51
kfogelam06:51
vilahi all06:56
kfogelhi vila06:56
pooliehello vila06:57
pooliereadjusted to europe?06:58
vilapoolie: fully :-)06:58
vilaI even add a ~2hours protest under my windows yesterday :)06:58
vilas/add/had/06:59
vilanot specifically targeted at me of course...06:59
poolieheh07:00
poolienot even at slow push to 1.13? :)07:00
vilanot even :-) But some wanted a faster selftest though... :-P07:01
lifelessvila: you begged :P07:01
lifelessvila: and we have now07:01
vilalifeless: it rocks ! I toyed a bit yesterday evening trying to spawn 256/2048 processes instead of just 8 and obvisouly the kernel just can't handle that (I have plenty of memory so no problem from that side)07:02
lifelessvila: :P07:03
lifelessvila: its fixable07:03
vilalol07:03
Peng_Possibly stupid question: How does doing BzrDir.find_branches() on a local transport wind up returning remote branches?07:03
lifelessits just in the wrong spot, the ec2 version has it moved to run()07:03
lifelessPeng_: branch references07:03
vilalifeless: spawning the processes you mean ?07:04
lifelessyes07:04
Peng_lifeless: Okay.07:04
Peng_I wonder where I have one of those?07:04
Peng_Oh, I remember.07:04
vilalifeless: diff available somewhere ?07:04
lifelessvila: no :P but the branch I put up has a new snapshot07:05
lifelessvila: remember that none of this history will be merged, its all noise - I'm cutting and pasting bits for the relevant projects07:05
lifelesspoolie: 7m30 or so with 5 instances07:08
lifelessthough there are still, uhm, glitches07:08
vilalifeless: branch url ? And yes, I understand we don't care about history (though I'm a bit surprised we make an exception there ;-P)07:10
lifelesstests.parallel07:10
vilalifeless: got it07:12
pooliePeng_, would you be interested in coming to this bzr sprint at all?07:12
poolierobert suggested at least putting you on the list07:12
vilalifeless: wow, of course I should ignore all that ec2 related stuff07:14
vilalifeless: I'll manage, back to test_breakin07:15
poolieok, i think that's enough07:17
lifelessciao07:20
vilahave a nice week-end !07:21
CokeHi! Odd question perhaps, are there any free project hosting for free projects (like launchpad) that use bzr?07:25
Coke(except launchpad, ofcourse)07:25
cody-somervilleSourceforge.net07:26
lukssourceforge, savannah, any web hosting07:26
CokeThey do?07:26
CokeSavannah is using bzr these days?07:26
mrooneyhmm is this an important error? bzr: ERROR (ignored): The medium '<bzrlib.smart.medium.SmartSSHClientMedium object at 0x9d71d2c>' has reached its concurrent request limit. Be sure to finish_writing and finish_reading on the currently open request.07:30
mrooneyI can't seem to get through this push, it just stalls at walking content, though if I control+c that is what I get07:30
pooliemrooney: it's typically a knock-on effect of some other error07:30
poolieit's better in later versions07:31
pooliecheerio07:31
mrooneypoolie: how many later versions are there than 1.13~rc1-1build1 ?!07:31
poolienot many :)07:31
mrooneyhm I can't seem to push, alas07:31
mrooneyit just sticks on walking content07:32
pooliemrooney: if it's reproducible please do control-backslash07:32
pooliewhich should put you in a debugger then type 'bt'07:32
mrooneyyes it does seem to be07:32
poolieand then put that in a bug and show vila :)07:32
pooliecause unfortunately i have to go07:33
mrooneyhm okay well here it is for any other eyes http://paste2.org/p/16792107:33
mrooneyI really need to go also07:33
mrooneyI just wanted to push this code so someone else could work on it07:33
mrooneyI guess not :/07:33
vilapoolie: lol07:33
vilamrooney: looking07:33
mrooneyvila: thanks :)07:33
vilamrooney: ok, so, hpss requests shouldn't interrupted as they don't know how to restart, instead you should use -Dhpss and look at .bzr.log07:34
vilayou can even add 'debug_flags = hpss' in bazaar.conf07:36
mrooneyit will always just get stuck at some random walking content number07:36
vilamrooney: we need to look at that 'random' :-)07:36
vilabut first,07:36
mrooneyso where should that log be?07:37
vilawhere are you pushing to ? What format is used there, are you stacking ?07:37
mrooneyI have no idea :) I just tried to push to a fresh branch07:37
mrooney"bzr push lp:~mrooney/do-plugins/skreemr-plugin"07:37
vilamrooney: 'bzr version' will tell you where the .bzr.log file resides07:37
CokeHey, I can't find any evidence at all that sourceforge or savannah is using bazaar, any link to docs explaining how to use it?07:39
mrooneyvila: http://paste2.org/p/16792307:40
mrooneythe top part is the .bzr.log part, the bottom is where I am stuck07:40
bob2http://bzr.savannah.gnu.org/07:40
vilamrooney: wow, wow, what bzr version are you using ?07:40
bob2for small values of using07:40
mrooneyyes indeed07:40
mrooney1.13~rc1-1build107:41
CokeThere's nothing in the user guide nor http://savannah.gnu.org/userguide/07:41
mrooneyjaunty07:41
Cokeand the faq is lacking bzr commentary07:41
Peng_You think it's a good idea to set public_branch on all of your, well, public branches on the server?07:42
vilamrooney: What 'bzr info -v' shows in your local branch ?07:42
CokeThe menus include git, cvs, svn and arch, no bzr07:42
Peng_It's a bit redundant..07:42
mrooneyvila: http://paste2.org/p/16792407:43
mrooneyvila: thanks for helping me with this by the way :)07:43
mrooneyhopefully I can get this up07:43
vilamrooney: hmm, assuming you put 'debug_flags = hpss' in bazaar.conf, can you try deleting the branch and pushing again ?07:46
mrooneyoh I didn't let me do that07:46
vilamrooney: I suspect a stacking problem but upgrading all do-plugins branches to 1.9 may not be an option :-/07:47
vilamrooney: I can't parse "I didn't let me do that", surely you can be gentle with you and authorize yourself to delete a branch no ? ;-P07:48
mrooneyhaha07:48
mrooney"I didn't, let me do that"07:48
mrooneynow I am getting tons of output in the log07:48
mrooneyafter doing that07:48
mrooneyvila: http://paste2.org/p/16792707:50
mrooneyit does that forever on and on07:50
vilamrooney: on purpose, that's a known bug, you're using a slow code path :-(07:51
vilafor excessive values of slow :(07:51
mrooneyvila: how long will it take?07:51
vilamrooney: the main factor is your ping time for lp,  with ~20ms I found it hardly tolerable, above that, well, it depends on whether you are patient or very patient :-/07:53
mrooney~170 on average here07:53
mrooney(bazaaar.launchpad.net)07:53
vilaI think lifeless just posted a patch against bzr.dev to address situations like yours (and spiv already posted one for some of them last week), it's definitively high on the list to have that bug fixed07:54
mrooneyso what is my situation though07:54
mrooneyall I did was branch from lp and push07:54
mrooneywhat is special about it?07:54
vilamrooney: nothing, that's the bug :)07:55
mrooneyhm okay haha07:55
mrooneyI'll guess I'll burn through some kilowatts and leave my computer on overnight07:55
vilamrooney: hmm, just checked, lifeless patch hasn't landed yet07:56
vilamrooney: if that's an option, go for it and let us know how it ends up :-/07:56
mrooneyeh I guess I will try pushing tomorrow at work on a better connection07:56
vilamrooney: the laternative will be to update all the do-plugins branches to format 1.9 which doesn't suffer from this bug07:57
mrooneybut I have 1Mb up, I guess the ping is bad07:57
mrooneyhow do I do that?07:57
vila'bzr upgrade lp:do-plugins --format 1.9' for all branches07:57
mrooneycan I do it on just mine?07:58
vilamrooney: possibly, but if stacking is enabled on lp for do-plugins, I'm not sure this will avoid the bug... It may be faster to try though07:59
mrooneyit will be faster to give up and sleep07:59
mrooney:)07:59
mrooneythanks for your help though07:59
vilamrooney: sure, sorry about that :-/07:59
mrooneyI will hope this magically works tomorrow or something07:59
vilamrooney: things should be better once the initial push suceeds...08:02
mrooneyall I need is the initial push :)08:02
mrooneyit is only for review so it can be merged haha08:02
mrooneyas I don't have permission to push elsewhere08:02
vilamrooney: hmm, looks like lifeless patch landed after all (BB didn't acknowledged it yet, but it's in bzr.dev), so you may want to try that too (please report if you do)08:04
mrooneybzr: 1 me: 008:05
mrooneyI'll try to even the score tomorrow :)08:05
mrooneygood night thanks for your help!08:05
vilamrooney: you're welcome08:05
mrooneyps I upgraded the remote branch and my local although it didn't seem to fix it, I suppose your stacking suspicion was correct08:06
vilamrooney: which remote one ?08:07
mrooney~mrooney/do-plugins/skreemr-plugin08:07
mrooneythe unpushed to one08:07
mrooneyI don't know if that actually does anything but it took about 10 minutes so I presume it does08:08
vilaI'm surprised you can update it if it's not pushed to yet... what does .bzr.log says ?08:08
mrooneyvila: http://paste2.org/p/16794008:10
mrooneyapparently it took only 3 minutes08:10
vilamrooney: hmm, sounds good, what does bzr push reports now ?08:10
mrooneystill hanging out at walking content08:11
mrooneylast thing in the log there is 8.569  fetch up to rev {mrooney@ubuntu.com-20090320073443-2g8kssdqgnwhyfnw}08:11
vilawell, so indeed, you may need bzr.dev (sorry about the trials, I don't have all the combinations clearly in head)08:12
Peng_Another possibly-stupid question: What if I want to read from a branch and conditionally write to it? Should I always take out a write lock? Can I take out a read lock and "upgrade" it to a write lock if necessary?08:13
* Peng_ wonders why downloading the most recent revision of bzr.dev over http took 1.5 minutes and ~3.7 MB of bandwidth09:35
Peng_It's 14 lines of changes!09:35
LarstiQbeuno: https://edge.launchpad.net/%7Ebzr-beta-ppa/+archive/ppa/+sourcepub/530792/+listing-archive-extra10:07
LarstiQjam: thanks, uploaded10:45
beunoLarstiQ, hi. So I should install bzr-svn from that PPA?13:30
pygihi hi13:32
LarstiQbeuno: either that or download the deb from there13:34
=== ampelbein is now known as Ampelbein
=== Ampelbein is now known as ampelbein
=== ampelbein is now known as goodkarma
=== goodkarma is now known as Ampelbein
beunoLarstiQ, installing bzr-svn works great14:50
beunowould you like me to use it in any way?14:50
=== nevans1 is now known as nevans
mxpxpodI just shelved some changes using bzr 1.13 which included adding some files and now when it goes to unshelve them it gives me this error: bzr: ERROR: No such file: None15:12
mxpxpodany way to fix that?15:13
=== Ampelbein is now known as ampelbein
LarstiQbeuno: I wouldn't object to that :) I'm reasonably sure that if it gets through the phase where debian/control matters, that the actual code itself works.15:23
beunoLarstiQ, so I should just try and check out an svn branch with bzr?15:24
beunogot any handy?  :)15:24
mxpxpodbeuno: http://svn.dojotoolkit.org/src/dojo/trunk15:25
LarstiQbeuno: stolen from subvertpy examples: svn://svn.gnome.org/svn/gnome-specimen/trunk15:25
beunofetching....15:27
mxpxpodso, does anyone know what's going on with bzr unshelve?15:27
beunomalbisetti@pentaserv:~$ bzr co svn://svn.gnome.org/svn/gnome-specimen/trunk test_svn15:27
beunoInitialising Subversion metadata cache in /home/malbisetti/.bazaar/svn-cache/203ae883-c723-44c9-aabd-cb56e4f81c9a15:27
beunoUpgrade to Subversion 1.5 or higher for faster retrieving of revision properties.15:27
beunomxpxpod, maybe you're mixing shelve and shelve2?15:27
LarstiQmxpxpod: in what sense?15:27
LarstiQah, you pasted something above15:28
mxpxpod:)15:28
mxpxpodbeuno: nope, I did bzr shelve and then bzr unshelve15:28
LarstiQmxpxpod: doesn't look familiar to me, did you check the bugs page in launchpad?15:28
mxpxpodyeah, looking now15:28
mxpxpodI could swear I've seen this before... but I don't remember the fix15:29
LarstiQmxpxpod: there have been various shelve fixes the past releases15:29
mxpxpodLarstiQ: yeah, I figured... I'm using 1.1315:30
mxpxpodand I need these files :(15:31
LarstiQmxpxpod: so if you trigger something in 1.13, either it's something new, something regressed, or newly exposed15:31
mxpxpodfrick15:31
LarstiQmxpxpod: we should be able to help you get your files back15:31
LarstiQmxpxpod: but it would be easier if this was a known problem15:32
mxpxpodLarstiQ: fun ;)15:32
mxpxpodthe other problem is that I can't give out the shelf file because the files are internal to my company15:33
LarstiQthat's fine15:33
mxpxpoddo you want me to paste the bzr log somewhere?15:33
beunoLarstiQ, worked like a charm15:34
mxpxpodLarstiQ: oh, the other thing is that I added a new directory to the shelf too15:35
mxpxpodand there was a file in that directory15:35
mxpxpodLarstiQ: hmm, it might be trying to add the file in the directory before adding the directory back15:36
jammorning vila (I have a conference call right now, but we can chat after if you want)15:38
vilajam: chat sounds good15:38
mxpxpodLarstiQ: it's dying in line 546 of bzrlib/transform.py15:39
mxpxpodLarstiQ: any clues?15:45
LarstiQmxpxpod: sorry, irl sidetracked me for a moment there15:53
LarstiQmxpxpod: your analysis sounds plausible15:53
LarstiQmxpxpod: with that sequence I get a different message15:55
LarstiQmxpxpod: could you cp -a your branch so we can look at the state as it was later again?15:56
mxpxpodLarstiQ: huh?15:56
LarstiQmxpxpod: .bzr/checkout/shelf is where shelf has stored your files15:56
LarstiQmxpxpod: I'm intending to get your file out first, and do bugreporting later15:57
jamvila: call has completed if you want to chat15:58
mxpxpodLarstiQ: ok, so do you want a copy of my branch?15:58
vilajam: sure15:59
LarstiQmxpxpod: no, I just want to make sure you have an extra copy that we're not going to touch15:59
mxpxpodLarstiQ: or, just for me to make a copy for later15:59
mxpxpodLarstiQ: gotcha15:59
mxpxpodLarstiQ: making the copy now16:01
mxpxpodLarstiQ: ok, done16:05
LarstiQmxpxpod: ok, `bzr shelve --list` only shows 1 shelve?16:06
mxpxpodLarstiQ: yessir16:06
mxpxpodand there's only one file in .bzr/checkout/shelf16:07
LarstiQmxpxpod: if you open that with an editor, do you see the contents of your file clearly?16:09
mxpxpodLarstiQ: yeah16:09
mxpxpodLarstiQ: but there are other changes in the file16:09
durin42jelmer: ping, can we bug you in #mercurial briefly?16:09
mthaddonhas bzr merge --force changed somehow in 1.13?16:14
mxpxpodLarstiQ: I guess I'll just have to apply all of those changes manually16:16
LarstiQmxpxpod: ok, let me try a couple of things16:16
jelmerdurin42: sure16:50
jamBasicOSX: ping16:56
BasicOSXjam:  hiyah16:58
jamhey, just checking the 1.13.1 release16:59
jamIt looks like you don't have the fix for the "bzr merge --force" regressino16:59
jamwhich was an issue for some other people16:59
jamDo you might getting that into 1.13.1 before we make it finalized?16:59
BasicOSXsec16:59
jpdsAnyone else having troulbe pushing to lp?17:00
jamBasicOSX: "bzr.dev 4146" I'm happy to create a cherrypick and submit it to PQM as long as you are signed off on it17:01
BasicOSXjam: I can squeeze it in :-) Looks like Vincent fixed my bzrtools bug17:02
BasicOSXso I was going to re-start the release as well17:02
BasicOSXcherry pick and PQM to the branch, I'll sign off on it17:02
jamk17:02
BasicOSXEeep! I missed 1.14rc1 -and- 1.14final according to mbp post!17:05
jamjpds: it seems to be working for me17:06
jamdid you try again?17:06
jamBasicOSX: yep. though if you read further he really did mean April :)17:07
BasicOSXI know, was just putting it in irc to tease17:07
fullermdShucks.  There I was thinking I'd finally gotten a good, long sleep...17:08
jpdsjam: Just made it.17:10
jamBasicOSX: change is on its way to PQM now17:10
jamShall I ping you when I get the confirmation?17:10
BasicOSXMy lunch hour almost over, drop me email on it?17:10
BasicOSXsince I'm going to loose wifi when I leave the cafe and drop back to the office :-) (yes, I bring my laptop to lunch)17:11
Takcan an uncommit be pushed to other branches?17:12
jamTak: with 'bzr push --overwrite'17:12
vilajam, BasicOSX : SHouldn't revno 4174 (bug #345169) be included in 1.13.1, it seems to address some pushing-to-lp-is-slooow bugs that some early 1.13.1 are experiencing17:13
ubottuLaunchpad bug 345169 in bzr "bzr push to a new lp branch of bzr from a local copy fails to stack" [Critical,New] https://launchpad.net/bugs/34516917:13
Takthen what happens if I pull from a branch that already has the revision as well?17:13
jamvila: it isn't a regression17:13
jam"source is 0.92 branch 7"17:14
BasicOSXvila:  I asked on the ... nm, jam answering17:14
vilaI know :-/17:14
jamvila: not worthy of a 1.13.1 release17:14
jamthey would get the same from 1.1217:14
jamTak: uncommit just rolls the pointer back, it doesn't mark it as rejected17:14
jamso pulling from a branch that had it will bring it back17:14
Takso what I really want is to unmerge that revision17:15
jamTak: bzr revert -r -2; bzr commit17:15
jambzr merge -r X..X-1 ; bzr commit17:15
jamsorry17:15
jambzr merge . -r X..X-1 ; bzr commit17:15
jamdepending on whether you want to reject the last commit, or something older17:16
jam(merge works in both cases, but revert is more obvious, IMO)17:16
* Tak nods17:16
Takthanks17:16
d6ghello, i have some problem to push to a branch today, it stops at17:20
d6g[|                   ] Transferring:Walking content. 1294/142317:20
d6gand when i stop it using ctrl-c, i got this error: http://paste.pocoo.org/show/108824/17:20
jamd6g: I don't think it stopped, it is still transferring, just not giving progress messages.17:21
jamThis would be with at least bzr1.13 to another bzr.1.13 host, right?17:21
d6git's 1.3 and push to launchpad17:21
d6g1.1317:22
jambzr.dev is a little better in that you still get 'activity' indications on the bytes sent/received.17:22
d6gbut how long am i supposed to wait, it seems just stuck in there17:22
d6gand the number 1294 seems random, sometimes larger/smaller17:22
d6gjam: i think you are right, i saw the upload speed now, it still is around 50kb/s, so i will probably have to wait, thx.17:24
mrooneyvila: no more apparent luck today, hm17:40
awilkinsSo, if I used --stacked on a branch that doesn't support stacking, how well does it work?18:04
awilkinsIt's currently telling me it's using 1.6, inside a 1.9 repo, and it's doing stuff that I presume is stacking (building tree")18:05
=== ampelbein is now known as Ampelbein
=== abentley1 is now known as abentley
=== Ampelbein is now known as ampelbein
=== ampelbein is now known as Ampelbein
nekohayohey there, I'm getting a "Permission denied (publickey)." when trying to "bzr pull lp:~kiddo/specto/specto-jeff" .... what's up?20:21
nekohayousing bazaar 1.6 "bundled with intrepid"20:21
smoothiceDoes anyone know if bazaar version 1.5.0 supports Launchpad?20:23
asabilnekohayo: probably logged in, and bzr pull lp:... tries to use bzr+ssh20:24
nekohayofor a pull?20:25
nekohayobut then I still have no idea what to do in that case20:25
asabilbzr launchpad-login20:26
=== Ampelbein is now known as ampelbein
jelmerabentley: hi20:58
abentleyjelmer: hi20:58
jelmerabentley: What could cause a "Revision incompatible" error while merging a bundle?20:58
abentleyjelmer: I guess going from rich-root to non-rich-root?20:59
jelmerabentley: that was it, thanks!21:07
nekohayoasabil: returns "kiddo"21:10
nekohayoasabil: nevermind, I upgraded to bzr 1.13 and that fixed it21:18
hsn_how can i get around error message from apt-get public key is not available when i am trying to access bzr repos?21:19
beunohsn_,21:20
beunohttps://edge.launchpad.net/~bzr/+archive/ppa21:20
beunohttps://help.launchpad.net/Packaging/PPA#Adding%20a%20PPA%20to%20your%20Ubuntu%20repositories21:20
hsn_beuno: thank you it works now21:25
exarkunHow do I change the push branch of a working copy that already has one?21:50
fullermdpush --remember21:51
exarkunWith version 1.3.1 that doesn't seem to do it21:51
exarkunOh :(21:51
exarkunYes it does, I just have a comprehension problem.21:52
exarkunThanks21:52
fullermdCoffee helps with that   ;)21:52
exarkunHmm yea....21:52
=== abentley1 is now known as abentley
lifelessvila: hi23:39
lifelessvila: do you have a copy of tests.parallel with both our stuff in it?23:39

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