=== asac_ is now known as asac [01:08] so ia32-libs is 560MB of so-called source, and we're ok with that beast? I mean, I know it's "just temporary until multiarch is done in 2006" and all that, but really...couldn't we break it up even a little? [01:11] lamont: I believe there is a plan to reduce it somewhat. I know for one that pulseaudio and its dependencies will probably have bi-arch packages in the coming months. Not elligant by any means, but better than nothing. [01:11] TheMuso: it was more one of we could split it up into ia32-libs-XX for some value of XX [01:20] lamont: right [01:21] lamont: I think cjwatson was working on reducing ia32-libs somewhat, not sure what he has in mind though. [01:22] it's certainly not worth splitting it up by any means other than creating individual biarch packages (which I haven't got round to yet, but may well do for pulseaudio). I don't know the current state but slangasek is pretty gung-ho about getting multiarch actually in place for 9.10 [01:22] yay!!! [01:23] cjwatson: Well if user bug reports re audio/pulse are anything to go by, bi-arch packages for pulseaudio are the best bet if multi-arch doesn't happen. [01:24] right, I just didn't get time for a4 [01:24] right thats fine. [01:35] cjwatson, i think he's off his rocker given i've been watching zero happen on that front since 2004. but hey, let's see! [01:38] directhex: there's a serious design that actually got widespread consensus at debconf (I nearly fell off my chair), and actual code being written [01:38] blimey [01:38] whatever next, HURD being usable? === nellery_ is now known as nellery [01:40] not to say it isn't still pretty challenging as a target, but it isn't quite the outer-space thing I thought it was going to be when he proposed it for karmic :) [01:48] so when's alpha 4 ready? :p [02:26] ctrl-alt-bkspace works i guess === nellery_ is now known as nellery [03:28] cd .. [03:28] woops === nxvl_ is now known as nxvl === emmy is now known as emmm === emmm is now known as emmy [04:51] when preparing an upload for a PPA, is there an easy (ideally scriptable) way to tell if the orig source needs to be uploaded? [04:52] it seems I can just always upload it, but that seems a bit wasteful [05:10] ojwb, for ppa's question, better ask in #launchpad [05:11] ta [06:22] c === cprov is now known as cprov-zzz [06:52] good morning === tkamppeter_ is now known as tkamppeter [07:58] is the intel graphic driver issue in jaunty has been fixed ? [08:57] is the intel graphic driver issue in jaunty has been fixed ? -> is there a patch ? [09:28] cjwatson, so it looks like rtg built the linux-fsl package with a full set of udebs (https://launchpad.net/ubuntu/+source/linux-fsl-imx51/2.6.31-0.2/+build/1157092) do you think i can save my energy and not hack around in debian-cd now ? [09:29] (i guess it still requires d-i adjustments) [09:31] ogra: sure, if they're about the right shape you shouldn't need further hacks [09:32] * cjwatson pokes [09:32] great, i'll only change livecd-rootfs then ... once i know for sure how they want to call the new metapackage [09:33] you'll have to make d-i or debian-cd or whatever build from universe too, I'm not really comfortable with putting this in main just yet [09:33] oh [09:33] that shouldn't be difficult [09:33] i guess that requires some tinkering in livecd-rootfs too [09:33] hardly major [09:33] * ogra goes to look [09:34] binaries accepted [09:34] can a buildd admin please give back "bogl", "cheese" (both CHROOTWAIT) and "liblauncher" (Failed to upload). thanks [09:34] thanks :) [09:34] geser: doesn't need a buildd admin, any uploader can give stuff back [09:35] * ogra wonders why he gets a Conflicting tags: 0.69 with bzr pull of livecd-rootfs [09:35] right [09:35] I'm not comfortable giving liblauncher back, I don't understand the error Soyuz is throwing [09:37] cheese appears to be successfully built everywhere already [09:37] I've retried bogl [09:47] does that diff look ok for livecd-rootfs ? http://paste.ubuntu.com/251232/ [09:50] well, it wont break anything but my own images anyway .... [09:50] StevenK: liblauncher got promoted to main while the buildd tried to upload the debs, so for soyuz liblauncher was in both universe and main till the next publisher run (as far as I understand it) [09:50] * ogra commits and uploads [09:53] ogra: yes, except temporary -> temporarily [09:53] whoops [09:53] ogra: though surely multiverse is overkill ... [09:53] i just copied the xubuntu line :) [09:53] * ogra fixes [09:59] pushed [10:24] warp10: [10:24] _ _ _ ____ ______ __ ____ ___ ____ _____ _ _ ____ _ __ ___ [10:24] | | | | / \ | _ \| _ \ \ / / | __ )_ _| _ \_ _| | | | _ \ / \\ \ / / | [10:24] | |_| | / _ \ | |_) | |_) \ V / | _ \| || |_) || | | |_| | | | |/ _ \\ V /| | [10:24] | _ |/ ___ \| __/| __/ | | | |_) | || _ < | | | _ | |_| / ___ \| | |_| [10:24] |_| |_/_/ \_\_| |_| |_| |____/___|_| \_\|_| |_| |_|____/_/ \_\_| (_) [10:24] [10:24] O_O [10:25] StevenK: "Duplicated ancestry" really means "I am terribly confused by the override that occurred while the build was doing its thing" [10:25] dholbach: :D man, that's awesome! Thank you very much! :) [10:25] for those who didn't notice, dholbach is back ;-) [10:26] haha [10:26] :-D [10:26] :D [10:26] wb dholbach :) [10:26] hey highvoltage :) [10:28] ugh, who did give back qt4-x11 on armel [10:28] mcasadevall, ^^^ was that you ? [10:29] oh, there was an actual upload, nevermind me then === cprov-zzz is now known as cprov [10:51] wgrant: So a giveback will clear it? [10:51] StevenK: Yes. [10:52] In this case the override actually occurred just before the build started, but the key bit is that the publisher didn't run between the override and the build finishing. [10:53] The publisher has hopefully run in the past three days, so it should be fine now. [11:35] Hey could someone with an ATI card on amd64 do me a quick favor and ldd /usr/lib32/libGL.so.1 (with ia32-libs installed) ? [12:14] has anyone else seen bug 411889 happen? [12:14] Launchpad bug 411889 in desktopcouch "package python-desktopcouch (not installed) failed to install/upgrade: trying to overwrite `/usr/share/pyshared/desktopcouch/__init__.py', which is also in package python-desktopcouch-records" [Undecided,New] https://launchpad.net/bugs/411889 [12:16] not, but one should depend on the other if it's enhancing it [12:17] (that __init__ file should only be present in python-desktopcouch, not python-desktopcouch-records aswell [12:18] Who would be the right person to talk to about upload.ubuntu.com -- I want to use sftp rather than anonymous ftp (because dput keeps failing with ftp due to my internet provider, but sftp works) ? [12:24] YokoZar, is your dput configured to use "passive ftp"? [12:24] how do i blacklist a udev module in ubuntu? [12:25] i dont want ubuntu to claim my usb printer [12:30] YokoZar: upload.ubuntu.com is Launchpad. But SFTP is a completely different protocol from FTP, and LP uses a custom FTP server, so it's not at all easy to change. [12:31] But it is intended that the FTP upload mechanism will be replaced soonish, I believe. [12:38] Does anybody know this sort of bug with VirtualBox 3.0.4 and recent Karmic -> http://people.canonical.com/~mmueller/VirtualBox-3.0.4_bug.png [12:38] Initial install went fine [12:39] Updating to Guest-Additions worked fine [12:39] then updated to the latest packages I got this garbage once gdm starts [12:40] it's with Karmic using kernel 2.6.31-5-generic === arand_ is now known as arand [12:54] * cjwatson debugs root-on-iSCSI into existence. Woo [13:05] liw: dput is at default, the issue is my comcastic internet connection that likes to forge reset packets [13:43] soren: do these tasks look plausible to you? http://paste.ubuntu.com/251338/ [13:44] soren: hmm, perhaps openssh-server on the node too [13:45] cjwatson: The node controller shouldn't need a dhcp server. [13:46] Err.. [13:46] I must admit, though, that the current packaging seems to suggest otherwise. [13:46] Now I'm confused. :) [13:48] cjwatson: I'll check up on that. It looks odd. Other than that, it looks reaonable. [13:49] I'm pretty sure dhcp3-server is unnecessary at best on a node controller [13:49] soren: it doesn't actually seem to use it, so seems like a packaging bug [13:50] I wonder why a) it ended up there to begin with and b) why noone has complained :) [13:50] soren: eucalyptus-cc needs it though, doesn't it? [13:50] cjwatson: I believe so. [13:50] yes [13:50] and it doesn't seem to be listed among its dependencies ... [13:51] No. That's why I'm wondering why noone complained. [13:51] I'll bung it into the eucalyptus-simple-cluster task for now then [13:51] hmm, you could argue it's not a hard dependency though, the default networking mode just needs a dhcp server to be present on your network [13:51] Ng: The default networking mode in Jaunty, yes. [13:51] Ng: Karmic will default to MANAGED. [13:51] soren: bold :) [13:52] how about I take it out of the task for now then since we don't have the new euca yet [13:54] anyone fancy looking at the cdbs ftbfs? bug 409863 [13:54] Launchpad bug 409863 in cdbs "FTBFS: FAIL: recursive.sh" [Undecided,New] https://launchpad.net/bugs/409863 [13:54] cjwatson: Makes sense. [13:55] cjwatson: At least we're consistent in our buggy dependencies then :) [13:56] soren: hrm, shouldn't -cc depend on bridge-utils too? [13:56] Ng: I would have thought so, yes. [14:13] cjwatson: I'm looking at updating kubuntu-meta and you've TIL. It appears you updated it using the Sid deboostrap because I get "/usr/bin/germinate-update-metapackage: Installed debootstrap is older than in the previous version! (1.0.13 < 1.0.15)". What's the best way to proceed from here? [14:14] Imagine I spelled debootstrap correctly in the last line please. [14:21] ScottK: oh, bah, good catch. I don't want to sync that into Karmic right now for various reasons - how about you just manually edit the debootstrap-version file? [14:22] cjwatson: I can do that, just thought I should check with you first in case there was a reason not to. [14:22] Thanks. [14:23] no, it should be ok [14:23] in this case anyway [14:24] It's running. Thanks again. [15:09] seb128, gah, you are to fast ... (/me wanted to ask if all of sbalneavs ubuntu fixed went into sabayon before that gets uploaded) [15:10] (i didnt see his name mentioned in the changelog, but i know all edubuntu users use his ppa packages nowadays) [15:11] not sure if his changes are in the new version I didn't know there was a ppa [15:11] i know he did the xephyr transition and sent that upstream [15:11] but i'm not sure all of his patches are in, i'll ask him to check if i see him around [15:12] i only follow edu stuff with half an eye nowadays === ivoks_ is now known as ivoks === bdrung_ is now known as bdrung [16:27] StevenK, poke, can i ask for de-NEWing linux-meta-fsl-imx51 ? [16:39] * ogra assumes StevenK went to bed (which would be valid at nearly 2am :) ) [16:39] could any archive adimn take a look at linux-meta-fsl-imx51 and wave it through the NEW queue ? [16:39] *admin [16:48] * ogra jumps up and down waving his arms to get archive admin attention [16:49] no #archive ? [16:50] nope [16:50] i know colin would help immediately if i pinged him directly, but i want the load being shared [16:53] didn't you just ping him directly? [16:53] nope [16:53] i doubt he highlights on his real name :) [16:55] ahaha [16:59] * ogra starts to glow in rotating colors to get some more attention, jumping and waving doesnt seem to help ... [17:03] so ping someone else directly? [17:07] Not sure if you guys saw this yet: http://lwn.net/Articles/346400/ (etch and feisty comparison in terms of common packages) [17:10] 1079 Ubuntu has Vebian bersion with ubuntuXX patch [17:10] ???? [17:10] funny typoing :) [17:18] Two of those updates only make sense if you update clamav to a newer version in Etch, which Debian won't do. [17:19] Oddly the analysis misses a bunch of the other clamav rdepends. [17:19] ScottK, you dont feel like taking a look at linux-meta-fsl-imx51 ? (since you made the mistake to speak here *and* are archive admin :) ) [17:20] ogra: There is specialness about kernel packages that the LP U/I can't handle. [17:20] Sorry. [17:20] it will stay in universe and vanish after alpha [17:20] oh, its only meta though [17:30] seb128, you dont happen to have some spare minutes to let linux-meta-fsl-imx51 out of NEW ? (its only a metapackage) [17:31] no [17:31] we have a meeting starting right now [17:34] * amitk is reminded of Scrat and his acorn from Ice Age when seeing ogra and his linux-meta-fsl-imx51 package :) [17:34] heh [17:35] amitk, well, if nothing helps i'll ping cjwatson but he does so much already that i wanted to put the load on someone else [17:35] you just did :-p [17:35] though a meta review is really trivial i think [17:35] amitk, yes, deliberately :) [17:36] since all the other archive admins are busy or not around [17:36] ogra: is it usually easy the bug the archive admin who is assigned for the day. [17:37] slangasek, ping [17:38] slytherin, meh, i forgot to look at the schedule on the wikipage, thanks for reminding [17:38] Riddell, poke, can you unleash linux-meta-fsl-imx51 from NEW ? === luisbg_ is now known as luisbg [17:41] kenvandine: hi [17:41] slangasek, so can you get xsplash on the desktop seed? [17:42] kenvandine: working on some other critical-path stuff at the moment, but I'll have a look at it later this morning [17:42] ok [17:42] please do [17:42] kenvandine, is there artwork already ? [17:42] pitti had verbally approved it [17:42] ogra, there is better temp artwork [17:42] (if i install it do i get the shiny stuff ?) [17:43] right now it is the default wallpaper [17:43] :) [17:43] * ogra would love to test it [17:43] ah [17:43] there are a couple of flickers and gdm needs to start sooner [17:43] so it isn't the final experience of course [17:44] but it is important that we get it on alpha4 :) [17:44] yeah, i wasnt expecting anything final :) [17:44] we really want bug reports, et [17:44] +c [17:45] what is xsplash? [17:45] ogra: I don't think there's a need for it to explain the history of linux in debian/copyright given that it's just a metapackage [17:45] so we still have a splash, no < 3s to gdm? :-\ [17:45] plus, Vcs-Git is invalid [17:46] james_w, i'll tell rtg, the package will vanish after A4 again anyway [17:46] kenvandine: or is the < 3s to gdm still expected once it has moved? [17:46] ogra: ok, thanks [17:46] james_w, no, thanks to you for unleashing it :) [17:46] ccheney, not sure how fast it will be [17:47] kenvandine: ok [17:47] iirc the plan was to not show splash if gdm came up in ~ 3s [17:47] slytherin, the new boot experience stuff for karmic [17:47] ccheney: isn't 3s overly optmistic? [17:47] ccheney, it is covering the black screen X had by default on startup [17:47] ccheney, well you shouldn't see usplash [17:47] but you should see xsplash [17:47] kenvandine: using kms? [17:47] * ccheney wasn't at the sprint though so plans might have changed [17:47] the idea is xsplash start with gdm and fades out when gdm is completely ready for you to login [17:48] then when you login, xsplash raises again until the desktop is completely loaded and fades out [17:48] so a nice smooth experience [17:48] kenvandine: ah ok, i think that was a change from uds, but sounds more realistic anyway :) [17:48] it is a change [17:48] ahh ,good [17:48] i thought it was something like plymouth [17:49] no... no point [17:49] our boot will be so fast :) [17:49] pfft [17:49] my laptop still takes 25sec [17:49] kenvandine, so usplash remains seeded/installed right? [17:49] yes [17:50] ogra, mine too... but people have great goals [17:50] will it actually show up before xsplash takes over, or is the intention to not even run usplash when xsplash is available to be used? [17:51] kenvandine, well, for me modprobe hangs in a 10sec loop, i doubt that will be fixed (hw doesnt respond) so i'll always have +10sec [17:51] superm1, you shouldn't see usplash at all [17:51] unless it fails to start X or something [17:51] basically usplash should only be seen if something is wrong :) [17:52] right now you will still see usplash [17:52] kenvandine, ah okay [17:52] until gdm gets moved to the beginning of the boot sequence [17:52] that will happen post alpha4 [17:52] it is incremental :) [17:52] we just want to get it out there now so we can get bug reports, etc [17:53] james_w, can i get a binary NEW as well from you ? (yes it already built :) ) [17:55] ogra: for main? [17:55] no, universe [17:55] that package will never enter main [17:56] it is only temporary for A4 [17:56] (we build the arm image from universe atm) [17:57] the arm kernel naming is still being discussed, its likely to change right after A4 [17:58] * slytherin wonders if any archive admin can review jmeter (arch all). === beuno is now known as beuno-lunch === rmcbride_ is now known as rmcbride === JanC_ is now known as JanC [18:07] ogra: eh, why are the kernel names changing again [18:07] ? [18:08] slangasek, bug 411968 [18:08] Launchpad bug 411968 in linux-fsl-imx51 "Please rename babbage to imx51, just like in jaunty" [Undecided,New] https://launchpad.net/bugs/411968 [18:09] slangasek, in our (mobile team) opinion the naming is wrong, but we postponed the discussion until after A4 to get the image out for now [18:10] I had already spoken to rtg about that last week, and he says the hardware that worked with the jaunty kernel doesn't work with the karmic kernel and vice versa [18:10] so there would be no reason to provide a direct upgrade path [18:10] thats not true [18:10] thats not true either [18:10] then I suppose you should make that clear to rtg in the bug [18:10] the HW that worked in jaunty will go on working [18:10] plus there will be two more iterations of the SoC that will work with it [18:11] and we would like to have the ability to add two additional ones the oem team cares for later (karmic+1 probably) [18:12] so the SoC name should be the name of the image/metapackage as it was in jaunty [18:12] but we'll discuss that *after* A4 as i said [18:20] kklimonda: see my audio-fixes branch ubuntu-karmic.git branch if you want to test-compile [18:20] wow, completely screwed that one [18:30] slangasek: I noted that in the bug report [18:32] slangasek: IMO it's wrong to name the kernel flavour after the board instead of the soc because we might want to support multiple boards with the same kernel flavour; I also believe that Babbage 1 (supported in jaunty) will work in karmic albeit I don't think that's a hard requirement -- the bootloaders are incompatible though [18:32] well, we already have one bootloader that clearly supports 2 boards ... so you never know :) [18:32] (babbage 2.0 and 2.5) [18:37] ogra: But we know for sure b1 and b2.x bootloaders are incompatible just like lange versus babbage bootloaders [18:38] yes [18:38] So I _do_ know [18:38] ;) [18:38] but who knows, b3 might use the same as b2 and 2.5 [18:38] Yeah but b1 wont [18:38] indeed === beuno-lunch is now known as beuno === dpm_ is now known as dpm === mrooney1 is now known as mrooney [19:49] f-spot (0.6.0.0-1) unstable; urgency=low [19:53] directhex, right was uploaded to karmic before as a fake sync [19:55] it was? :o [19:57] directhex: f-spot (0.6.0.0-1~ubuntu1) karmic; urgency=low [19:57] directhex: https://edge.launchpad.net/ubuntu/+source/f-spot ;) [19:57] hax! [19:58] directhex, I did sync the new mono too today [19:59] seb128: still archive admin? then sync cli-common too please, directhex didn't need a main ACK, I got one :P [19:59] sebner, bug number? [20:00] seb128: 410744 [20:00] seb128: bug #410744 [20:00] Launchpad bug 410744 in cli-common "Please sync cli-common 0.7 from Debian(Unstable)" [Wishlist,Confirmed] https://launchpad.net/bugs/410744 [20:03] sebner, done === sbasuita_ is now known as sbasuita [20:04] seb128: my hero \o/, thanks :) [20:04] you're welcome [20:05] yay for seb128 [20:06] seb128, barring bugs, i think that should be it for mono in karmic... although i'm leaning on meebey to make a small change which should save us 200k [20:07] directhex, sebner: could one of you look if gnome-desktop-sharp2 can be synced? [20:07] the diff seems to be non ubuntu specific [20:07] ie could be fixed in debian now [20:19] hey [20:20] k [20:28] ogra: so why is arm being built from universe? that's not a good thing, and I don't think we should be propagating that further by putting the kernels in the wrong place [20:30] kenvandine: should xsplash replace usplash in the platform desktop-common seed, or is there a reason this needs to be Ubuntu-only right now? [20:30] slangasek, it doesn't replace [20:30] hmm? [20:30] it is desktop specific though, needs X and gdm [20:30] if all is good you won't see usplash [20:30] but we want usplash to be there if something goes wrong [20:31] usplash is only in the desktop-common seed, that already implies X (but doesn't imply gdm in the case of kubuntu) [20:31] ok [20:31] ok [20:31] alright, seeded & promoted [20:31] woot [20:32] thanks slangasek! [20:32] now I just need to get eglibc vetted, and we can start building us some ISOs [21:21] mcasadevall: Can you look at dbus on ia64? Looks like a missing include to me, but I've got no way to test. [21:23] ScottK, *groan* that's a linker error [21:23] ScottK, I'll put it on my TODO list, but it probably won't get looked at until post-A4 [21:24] mcasadevall: That's fine. [21:24] ScottK, anyway, its possible qt4-x11/lzma weirdness might be fixed [21:24] mcasadevall: Excellent. With the one that's building right now perhaps? [21:24] ScottK, that's our test [21:24] err .... was last I looked. [21:25] ScottK, its still going [21:25] No idea if it will finish or not [21:26] ScottK, if it works, woo, if not, then common sense has been let go on this bug [22:22] mcasadevall: qt4-x11 seems to have finished successfully, BTW. [22:25] I have a request? [22:26] When you are updating and it say to run sudo dpkg --configure Have the update manager run it automatically without telling you to... [22:28] sudo dpkg --configure -a [22:48] Hey folks, any idea why "passwd/make-user= false" question during preseed would be ignored? It's delivered in the http preseed file, but the system still asks to create the user (name, username, password etc..) [22:51] docelic: you'll have to create a root user; user-setup won't allow you to install with no user at all [22:52] so 'd-i passwd/root-login boolean true' 'd-i passwd/root-password password blah' 'd-i passwd/root-password-again password blah' [22:52] (or passwd/root-password-crypted. see the guide) [22:53] ah! many thanks, root-login was the thing I missed [22:54] also, the example-preseed.txt file (in the docs from 9.04) has a mistake in one field, let me tell you which one [22:55] Question clock-setup/ntp-server [22:55] the third field (type = string) is missing === slangasek changed the topic of #ubuntu-devel to: Archive: main frozen for alpha-4, DebianImportFreeze | Development of Ubuntu (not support, not app development on Ubuntu) | #ubuntu for support and general discussion for dapper-jaunty | #ubuntu-motu for getting involved in development | http://wiki.ubuntu.com/UbuntuDevelopment | See #ubuntu-bugs for http://wiki.ubuntu.com/HelpingWithBugs [23:08] docelic: thanks; that was fixed a little while back in Debian and is fixed in Karmic, although unfortunately we overlooked the fix for 9.04 === robbiew is now known as robbiew_away [23:18] just as a sanity-check, karmic won't update to automake 1.11, will it? pulseaudio git now requires 1.11. [23:19] didrocks: are there source changes needed to gir-repository to transition it to clutter-gtk 0.10? [23:19] last I checked we were waiting for http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=512704 to be fixed [23:19] Debian bug 512704 in automake "automake: New upstream version available" [Wishlist,Open] [23:19] rather than doing it ourselves [23:20] infinity, !\o/! [23:23] cjwatson: thanks [23:31] StevenK: gir-repository was recently MIRed, it build-depends on goocanvas which hasn't been; is that something that should be done, or is gir-repository sufficiently "temporary" that we should be getting rid of it, instead? [23:32] superm1: hi === Gh0sty_ is now known as Gh0sty [23:35] slangasek, gir-repository is not really "temporary" enough [23:35] alas [23:35] ie it will not be solved this cycle [23:35] so there's a bunch of MIRs which would be needed for its build-deps, then [23:35] that's a collection of things that each library should build [23:35] (clutter-cairo, goocanvas, ...) [23:35] but since upstream didn't start on this move yet ... [23:37] the whole gir-repository thing sucks [23:38] yes, I can tell by looking at it :) [23:38] it's just a workaround to have gir working without doing thing properly [23:39] ie it's fine for early testing etc but things should be move to proper libs [23:45] But should we promote goocanvas. [23:46] I guess promoting it for completeness would be nice, and if we need to actually rebuild/sync gir, but ... [23:52] Could someone possibly get the latest binaries for linux-rt out of new? They've been there for a few days, and are needed for the next studio alpha. [23:53] A mostly trivial merge of Subversion 1.6.3 (we currently have 1.6.1) from Debian is currently blocked owing to Debian's version now pulling in Serf as a builddep and hence requiring a new MIR. Serf is an optional builddep for Subversion which Debian just happened to enable now. There are important bugfixes in Subversion 1.6.3. Given the proximity of FeatureFreeze, would it be sensible for me to prepare an adjusted merge that drops the Serf [23:53] builddep? [23:58] maxb: 'yes' [23:58] :-)