[00:00] manjo: it's pretty easy to change that [00:01] kklimonda, yeah edit files .. [00:20] kklimonda: Careful, I've just noticed there are host specific peripheral settings in gconf [00:21] (although bizarely the only thing under periphers->keyboard->host-major is the default seetting of numlock) [00:29] penguin42: but they should still work after hostname is changed. [00:29] kklimonda: I just wonder what other settings are actually keyed off hostname [00:30] penguin42: It's not really a good idea to do that. [00:30] penguin42: if any application uses host in its settings then bug report should be filled imo [00:30] kklimonda: It is in an envrionment with a networked home directory; you want peripheral config to be local to where you are logged in [00:31] penguin42: then they should be stored somewhere local imo [00:31] admittedly those environments are rarer these days [00:32] penguin42: and in those environments you don't use a shiny graphical installer anyway. [00:32] probably true [01:29] can someone with a vague knowledge of gnome eyeball some code in mango-lassi for me - it looks broken to me, I just can't understand why it only segs on one of two machines [03:32] hay [07:36] hi cjwatson [07:42] i would suggest executing dpkg-reconfigure -phigh console-setup in casper after setting the locale [07:43] and when you use your installer then too [07:43] otherwise the console font is incorrect for the locale [07:43] then you can not see umlauts for example when you boot with german settings live === Claudinux_ is now known as Claudinux === zyga-gone is now known as zyga === hunger_ is now known as hunger [09:55] doko: do you mind if I upload a python-central crash fix? or do you want to review first (is there a vcs to use?) [09:56] mvo: no, please go ahead [09:59] thanks doko [10:23] kenvandine: libdbusmenu_0.3.13-0ubuntu1.dsc extracts fine for me. can you help me get a test case so that I can understand your problem? [10:23] mneptok: btrfs /boot> no, held up on licensing [10:25] kenvandine: hm, well, somebody seems to have solved the problem since there's a libdbusmenu_0.3.13-0ubuntu1.dsc in the archive now [10:29] cjwatson, I think he remade the tarball [10:30] the one on https://edge.launchpad.net/dbusmenu/+download didn't work for him [10:38] ok [11:09] cjwatson, do you know how the ddeb service works? [11:09] ie where the collector is running? [11:12] not really [11:12] I think I figured it out [11:12] the service seems stucked since yesterday, could be due to the launchpad downtime [11:12] I will just restart it and see === nenolod_ is now known as nenolod [11:16] enrico: it looks like apt-xapian-index 0.39 was never released. should I just append my change to the 0.39 changelog, and mark it UNRELEASED in git? [11:22] cjwatson: I just commited the "apt-cdrom looks into .disk for calculating the ident string" change, do we need to worry about backward comatibility? I guess not as it was never really working with usb-sticks anyway? [11:30] mvo: can you point me at the commit and I'll have a look? [11:31] * Daviey screams. [11:31] .. and sobs [11:36] * hyperair gives Daviey a strange look [11:37] * soren pats Daviey on the head [11:37] * davmor2 passes Daviey a cookie and tells him to eat it, life will be better afterwards [11:38] \o/ [11:38] That's more like it. [11:38] Daviey: now get back to work and fix that issue :P [11:39] yes sir! [11:39] enrico: (I'm ready to commit a fix for the xapian api change as well, once I know the answer to that question :-) ) === fabbione is now known as fabbione_vac [12:13] mdz, hi [12:14] cjwatson: did you get my notice about the wrong fonts in console === MacSlow is now known as MacSlow|lunch [12:28] tkamppeter: hi [12:31] cjwatson: here is the commit http://bazaar.launchpad.net/~ubuntu-core-dev/apt/ubuntu/revision/1796 (sorry for the delay, I was at lunch) [12:31] damn you for EATING [12:31] hm, is access() a reliable test for that then? [12:33] mdz, it is about bug 633987. [12:33] Launchpad bug 633987 in foomatic-db (Ubuntu) "foomatic-db-compressed-ppds holds back ubuntu-desktop" [Undecided,Incomplete] https://launchpad.net/bugs/633987 [12:33] tkamppeter: yes, I answered your question in the bug. did you and mvo talk further about it yesterday? [12:33] cjwatson: I'm not 100% sure, but in my tests it seems to be ok [12:33] tkamppeter: there were a few other people on IRC yesterday who confirmed the bug [12:33] mdz: no, sorry. but s-c/aptdaemon are uploaded now, so I have breathing space again for this [12:34] access("/mnt", W_OK) = -1 EROFS (Read-only file system) [12:34] apparently so. neat [12:34] mdz, I do not really understand why the update is not working for you. It probably works for many users as I did not get any duplicate. [12:34] mvo: you might want to e-mail the debian-live folks and ask, but given that it basically always breaks at the moment I think it should be OK [12:34] cjwatson: I tried with usb-stick (both rw,ro) and cdrom and got the answers I was expecting [12:34] cjwatson: good idea, I will do that [12:34] tkamppeter: mvo is our resident expert on this type of issue [12:35] mdz, I can try to replace the Conflicts: by Breaks:, but this is simply a guess. [12:35] tkamppeter: give me some minutes, I will try to reproduce the setup that mdz has [12:35] mvo, what exactly is the difference between Conflicts: and Breaks:? [12:36] enrico: looking at the history, I guess you don't normally use UNRELEASED, so I'll go ahead and put these changes in 0.39 [12:36] tkamppeter: internally they are handled similar, but for breaks apt will try a little bit harder to upgrade the package before it gets moved to the remove-list [12:36] tkamppeter: the policy manual documents this clearly [12:37] http://www.debian.org/doc/debian-policy/ch-relationships.html#s-breaks [12:37] http://www.debian.org/doc/debian-policy/ch-relationships.html#s-conflicts [12:37] the latter has specific instructions for when each field should be used [12:38] cjwatson: did you fix the iso build tool to get hybrid iso images [12:43] cjwatson, thanks. [12:56] is launchpad broke at the moment? #launchpad seems pretty dead and I'm having problems reporting an oops with a 'sorry, there was a problem connecting to the launchpad server.' and a guy I was helping last night had the same problems [12:59] cjwatson: I updated the diff yet again (noticed another bit that needed fixing) and mailed debian-live. I guess its best to wait for feedback from debian-live before uploading so that the ident id does not change/become incompatible if we decide to change the code again. what do you think? [13:00] mdz: could you mail me a tarfile with /var/lib/dpkg/status and /var/lib/apt/lists please ? I will dig into it then [13:05] mvo: seems reasonabe [13:05] +l [13:05] cool, thanks [13:11] mdz: nevermind, I can reproduce it here now [13:14] enrico: I've uploaded 0.39, and will sync that into Ubuntu [13:15] tkamppeter, cjwatson: the eaisest fix for the foomatic-db problem is to make foomatic-db a lower priority then foomatic-db-compressed. so swtichting foomatic-db to "Priority: extra" should do it. I belive this needs to be done by a archive-admin === MacSlow|lunch is now known as MacSlow [13:18] cjwatson: excellent, thanks! (just came online, I was at a customer) [13:21] Suggest a cross platform IDE,..... I know about eclipse and netbeans... tell me which is good or other if you know any other [13:21] mvo,tkamppeter: I've set the priority of foomatic-db to extra, as well as foomatic-db-engine [13:22] thanks cjwatson [13:22] * mvo updates the bug [13:27] cjwatson: and thanks for the xapian fix for software-center :) [13:29] mvo, thanks. So you have uploaded foomatic-db_20100906-0ubuntu2 now? [13:30] s/mvo/cjwatson/ ^^^ [13:31] tkamppeter: I think cjwatson meant that he changed the priority of -0ubuntu1 [13:32] mdz, where are these priorities set? How can I set them if I encounter a similar problem in the future? [13:33] tkamppeter: they are set centrally by archive administrators; it requires special privileges [13:35] mdz, mvo, cjwatson, looks a little like a workaround for a shortcoming (bug?) in apt for me. In this case foomatic-db-compressed-ppds should get priority against foomatic-db as it solves the dependencies of the new ubuntu-desktop and so no artificial prioritizing is needed. [13:38] maybe so, but in this matter I bow to mvo's expertise :) [13:38] and yes, mdz is correct [13:38] tkamppeter: yes, I'm inclined to agree that it's a workaround, but we're not likely to try to fix this in apt for 10.10 ;-) [13:40] mvo, mdz, cjwatson, are these package priorities somewhere visible in Launchpad? [13:41] tkamppeter: it is, the resolver is not always the most smartest in the world. fortunately its very predictable. I look into a solution in code next, but for mav we can not really change the resolver anymore [13:41] (too risky) [13:41] tkamppeter: yes [13:42] tkamppeter: though the easiest way to see them is to use apt-cache [13:43] mizar:[~] apt-cache show foomatic-db | grep Priority [13:43] Priority: optional [13:43] the top three priorities (required, important, standard) matter to varying degrees and are managed semi-automatically, but mostly we don't bother much with the distinction between optional and extra in Ubuntu [13:43] it's just useful to resolve occasional bugs like this one [13:44] mdz, I have done [13:45] sudo apt-get update [13:45] * cjwatson wonders how sudo got to be Priority: optional, and fixes it back to important [13:45] tkamppeter: it's not instant, you need to wait until the next publisher run. give it an hour [13:45] now and checked the priorities, both foomatic-db-compressed-ppds and foomatic-db have "optional". [13:46] cjwatson, OK. [13:46] an hour> assuming you're using a mirror that's pushed fairly quickly that is [13:46] cjwatson, mvo, mdz, thank you very much for the help on this. [13:47] cjwatson, I am using the German mirror. === ogra_ is now known as ogra_panda [13:58] tkamppeter: hi, me again. trying to work around bug 612578 I tried to forward-port cups from jaunty, but the build fails http://pastebin.com/aMeWijTY [13:58] Launchpad bug 612578 in cups (Ubuntu) "Printing fails unless the IPP's are purged, but fails again after "some time"" [Undecided,New] https://launchpad.net/bugs/612578 [13:58] tkamppeter: ideas what makes it fail? === ivoks-afk is now known as ivoks === almaisan-away is now known as al-maisan [14:29] please can someone look at sponsoring the tar sru in bug 539814 [14:29] Launchpad bug 539814 in tar (Ubuntu Lucid) "tar: futimens() with a bad file descriptor (AT_FDCWD) causes bootstrapping failure with kernels < 2.6.22" [Medium,Triaged] https://launchpad.net/bugs/539814 [14:42] hi guys is this the right channel for packaging help? [14:45] rhlee: There is a #ubuntu-packaging [14:46] penguin42: cheers mate === dendro-afk is now known as dendrobates [15:03] tedg: hey! So is there any other problem with my merge(s) ? [15:05] tedg: Thanks for the comment on the bug, I'll be off now [15:05] bilalakhtar, No problems other than we need to figure out the symbolic/color part [15:06] Uhg, the IRC equivalent of starting a conversation as you're jumping out of the plane. [15:07] lol [15:11] cjwatson, around and have some time to talk about grub and that annoying ec2 ? [15:12] anyone finding alt-sysrq-b is being ignored? s and u are both working but B is doing nothing [15:16] mvo_, cjwatson, mdz, on my mirror priorities have changed to extra for foomatic-db and optional for foomatic-db-compressed-ppds. Is this OK? Can you try? [15:17] tkamppeter: please remove me from the list of people you're asking; all I did was the archive-side change [15:17] smoser: let me read through it all again; I'm having real trouble cramming all this into my head [15:18] cjwatson, ok. i've got a bout 45 minutes, then i need to be afk for 30. [15:18] so whichever hunk works better for you. [15:20] smoser: is there some way that I can try this out personally? I'm never going to really understand it otherwise [15:20] both on EC2 and UEC [15:20] nb: must not involve having to set up a cloud in my house :) [15:20] cjwatson, i can get you an instance on ec2 easy. [15:21] euca is more difficult, but in general, euca almost "just works" [15:21] the only issue with it is that i'd like to be able to tell grub "don't worry about installing the bootloader anywhere" [15:21] can I please not have pre-specified solutions [15:21] I want to inspect the problem :) [15:22] comon. [15:22] ok. i'll get you an ec2 instance, and will work on getting you access to a uec also. [15:23] I mean, you *can* tell grub-pc.postinst not to install the bootloader anywhere, and I know this because d-i does it [15:23] but I'm not clear yet whether this is quite the right thing [15:23] in particular creating core.img is considered part of installing the bootloader [15:23] from grub-pc.postinst's point of view [15:23] (yes, within grub-install they're two separate steps) === bjf[afk] is now known as bjf [15:25] * penguin42 is just testing afix for the lshw hang - I can see why it oops's, I just can't see why lshw manages to trigger it === amitk is now known as amitk-afk [15:26] cjwatson, ok, you can ssh to ec2-75-101-239-180.compute-1.amazonaws.com [15:27] and that system is "all yours" until you tell me to kill it. if you shut it down (/sbin/halt, it will go away). [15:27] your ssh keys from launchpad are installed on that host [15:39] smoser: so, if /dev/sda1 is really a disk image, then I wonder if it really makes sense for GRUB to regard it as a partition [15:39] (aside from whatever Xen is doing) [15:39] maybe we should go "huh, weird Xen thing, treat it as a disk" [15:39] grub should not regard it as a partition. [15:40] it is a disk for all purposes , just with a funny name. [15:40] right, good. it's not as if there's any particular relationship between /dev/sda1 and /dev/sda2 here [15:40] note, also, that this disk is funny in that it has no partition table. [15:40] GRUB will be OK withthat [15:40] kees: ping [15:40] tkamppeter: confirmed, it works now [15:42] i poked trhough some of grub source, and hacked at trying to convince it that was a full disk. my hack was to check if this 'name' existed in /sys/block, then it was a disk, not a partition. [15:47] doko: Is i686 without cmov supposed to be supported in maverick or not? [15:48] LP #632232 [15:48] Launchpad bug 632232 in debootstrap (Ubuntu) "lucid pbuilder-dist unable to build for maverick" [Undecided,New] https://launchpad.net/bugs/632232 [15:51] mdz, OK. Thanks. I will close the bug then. [15:52] smoser: I'm thinking of instead simply observing that if the partition device exists but the disk device does not, then we might as well treat it as a disk since what else can we do? [15:52] james_w: ping [15:52] hi barry [15:53] james_w: hi. have a few minutes to talk about a udd issue? [15:53] yeah. i can't imagine other scenarios where that would fail. [15:53] barry: of course [15:54] james_w: cool. so i'm working on an update to the gtimelog package. it's been basically abandoned and i have upstream commit privs now. marius released 0.4.0 and i'm redoing the packaging to more modern standards... [15:55] james_w: upstream trunk is in bzr on lp. so i 'bzr branch trunk packaging' and create a loom, add a thread for the debianization work... [15:55] james_w: convert the patches i need (and the still relevant ones from the old packaging) to quilt, commit all that and 'bzr bd -S' [15:55] james_w: but i get a lintian warning: patch-system-but-direct-changes-in-diff [15:56] might I ask that anyone with a free moment run `sudo dmidecode` on their machine and pastebin me the output? Thanks! [15:56] barry: what's an lsdiff -z of the .diff.gz say? [15:56] james_w: now, the reason for that is that the trunk has moved a little bit from the released tgz. it's got a bit of extra NEWS.txt and a version bump in the __init__.py [15:56] ah [15:56] james_w: i can certainly ignore that, but i guess i'm looking for recommendations on how to handle this scenario in general [15:57] (ignore the lintian warning) [15:57] then base your packaging on the revision corresponding to the tarball if available [15:58] james_w: right, that's the direction i was thinking about, but my question is: what happens when a new version is released from trunk? what will be the best way to take my existing packaging loom and update it? just pull from trunk with a revision# or tag to the bottom thread? [15:58] barry: yeah [15:58] barry: it would be great if we could get the pristine-tar data in there too [15:59] james_w: hmm, how could that be done? [15:59] if you can use bzr import-upstream/merge-upstream at some point [16:01] james_w: yeah. i think the scenario where the upstream is in bzr on lp is a weird corner case that actually makes things a little more difficult in several ways. e.g. lp:ubuntu/gtimelog has no relationship to lp:gtimelog [16:01] barry: yeah, but e.g. the dx team have that, and we linked the branches [16:02] james_w: ah, didn't know that could be done. is that a manual request? === apparle_ is now known as apparle [16:02] barry: it's something you can do yourself [16:03] james_w: i'm probably just being dense. it's obvious? [16:04] barry: you pass an upstream branch and revision to merge-upstream [16:04] james_w: gotcha [16:05] james_w: cool, thanks. as always very helpful. i'll work this out and see if i can write something up on the wiki [16:05] great, thanks [16:05] let me know if you have any trouble [16:07] james_w: will do, thanks again [16:11] lool: no [16:23] doko: thanks [16:23] there should be an email [16:27] Hey guys [16:27] rmemeber me? [16:27] Randall: no. [16:28] Genesis 2 [16:28] 1Thus the heavens and the earth were finished, and all the host of them. [16:28] 2And on the seventh day God ended his work which he had made; and he rested on the seventh day from all his work which he had made. [16:28] i do.. you're the asshole [16:28] 3And God blessed the seventh day, and sanctified it: because that in it he had rested from all his work which God created and made. [16:28] 4These are the generations of the heavens and of the earth when they were created, in the day that the LORD God made the earth and the heavens, [16:28] 5And every plant of the field before it was in the earth, and every herb of the field before it grew: for the LORD God had not caused it to rain upon the earth, and there was not a man to till the ground. [16:28] 6But there went up a mist from the earth, and watered the whole face of the ground. [16:29] 7And the LORD God formed man of the dust of the ground, and breathed into his nostrils the breath of life; and man became a living soul. [16:29] 8And the LORD God planted a garden eastward in Eden; and there he put the man whom he had formed. [16:29] 9And out of the ground made the LORD God to grow every tree that is pleasant to the sight, and good for food; the tree of life also in the midst of the garden, and the tree of knowledge of good and evil. [16:29] 10And a river went out of Eden to water the garden; and from thence it was parted, and became into four heads. [16:29] 11The name of the first is Pison: that is it which compasseth the whole land of Havilah, where there is gold; [16:29] 12And the gold of that land is good: there is bdellium and the onyx stone. [16:29] 13And the name of the second river is Gihon: the same is it that compasseth the whole land of Ethiopia. [16:29] 14And the name of the third river is Hiddekel: that is it which goeth toward the east of Assyria. And the fourth river is Euphrates. [16:29] 15And the LORD God took the man, and put him into the garden of Eden to dress it and to keep it. [16:29] nice [16:30] what a troll [16:30] * penguin42 can't imagine he would be good with daemons [16:30] was i the only one reading that? [16:30] brewmster: Yes [16:31] i think he might be a slow adult [16:32] there is no need to cause further noise by discussing it further., [16:32] particularly since I can see for myself that the two people who responded to Randall most quickly joined the channel immediately before him and haven't been here before ... [16:33] cjwatson: there is no need to cause further noise by discussing it further., [16:34] * jdong takes a pass through the SRU queue [16:41] Is LP behaving funky? https://edge.launchpad.net/ubuntu/lucid/+queue?queue_state=1&queue_text= has two identical looking php5 packages [16:42] jdong: that's permitted in queues [16:42] only one of them gets to be accepted, of course [16:42] cjwatson: oh, interesting; didn't know that [16:42] check that they have the same content? [16:43] EdwinGrubbs: hola! [16:44] yep, the diffs look the same [16:46] kees: hi, Launchpad is discussing whether to switch to the apport (problem_report) format for its oops tool, which puts the python traceback, cgi variables, sql query times, etc. for a single failed request in one file. [16:47] kees: there are certain limitations to the apport format, and I wonder whether we would be better served by wrapping problem_report, pushing changes upstream, or just using another format. [16:48] kees: I haven't seen pitti in IRC. Who all would be good to CC our oops tool discussion? [16:51] one issue I see with problem_report is that it only really supports parsing a single level of field names and values, so parsing a complex value is nonstandard. The other is that it doesn't seem to let you set the content-type for a file attachment. === yofel_ is now known as yofel === dendrobates is now known as dendro-afk [16:55] EdwinGrubbs: I think extending apport would be great. [16:55] EdwinGrubbs: pitti is very responsive to patches. you could email ubuntu-devel and CC him, maybe [16:56] kees: ok, thanks, I'll do that. [16:57] EdwinGrubbs: okay, cool. === deryck is now known as deryck[lunch] [17:12] EdwinGrubbs: pitti is on holiday currently [17:15] achiang: thanks for the info [17:15] sure === bjf is now known as ___bjf === ___bjf is now known as _b_j_f === dendro-afk is now known as dendrobates === beuno is now known as beuno-lunch === deryck[lunch] is now known as deryck [18:03] hm, is umask anywhere in /proc for a process? doesn't seem to be. === al-maisan is now known as almaisan-away === dendrobates is now known as dendro-afk === dendro-afk is now known as dendrobates === ogasawara_ is now known as ogasawara [18:38] cjwatson, this might be in the realm of 32-bit is undefined, but from my experiments of throwing the 64 bit grub efi on a 32 bit CD and making it multi-catalog, it looks like the 32-bit kernel doesn't load with efi, missing /sys/firmware/efi and all the accompanied dmesg listings about memory mapping with EFI === beuno-lunch is now known as beuno [18:47] looks like I need python-gconf-dbg and it isn't built. Is that a bug, or am I missing something? [18:48] superm1: I looked into the kernel a bit. It refuses to load EFI unless the boot loader signature passed to it says EL32 (for a 32-bit kernel) or EL64 (for a 64-bit kernel); grub's x86_64-efi build passes EL64 [18:48] superm1: but that is not the worst problem [18:49] superm1: the EFI runtime services calls are just function pointers. If you have a 64-bit EFI implementation, those functions are very likely to contain 64-bit code. You can't call them from a 32-bit kernel [18:49] cjwatson, oh yikes- yeah that could be quite troublesome then [18:49] I don't see any way we can cope with this [18:50] if you have 64-bit EFI and want to boot a kernel in EFI mode, as far as I can see it has to be a 64-bit kernel and that's really all there is to it [18:51] so we'll need to be addressing whatever problems you have that make that difficult [18:53] indeed I just checked the UEFI spec and it requires this [18:53] "For an operating system to use any UEFI runtime services, it must: [18:54] [be] In long mode, in 64-bit mode [18:54] " [18:54] that's in the "x64 Platforms" section, 2.3.4 [18:54] OK, i'll take that back to my team and see what we can come up with. [19:02] cjwatson: thanks for the reply. i answered my own question with a VM yesterday. [19:04] cjwatson: your late reply meant i could not be lazy, which creates a very interesting Slacker Feedback Loop. ;) === ivoks is now known as ivoks-afk === dendrobates is now known as dendro-afk [19:45] Does ubuntu have any way to extract an update priority in a similar fashion to Mint's update manager w/ its colour and number coding? Even a commandline thing would be fine [19:46] I'd just like to know if that is a capability it has [19:59] nemo: I think it's on the list. Eg. IIRC the colour was from some fairly simple heuristics (eg. is a core library) [20:01] sladen: oh... [20:01] sladen: I thought they were doing it based on "is this an exploit fix" [20:01] versus "is this a stability only issue" or "is an enhancement" [20:02] sladen: 'cause really, that's what our admins want to know - is this something crucial that has to be applied to the server right away [20:02] (and w/o reading each description to find them) [20:02] nemo: if it's post-release, it's a security fix. If it's pre-release it's a bug-fix enhancement [20:03] bug-fix or enhancement [20:03] nemo: if your admins are blindly applyin security updates, I would have my concerns about their skills... [20:05] Chipzz: well, these are windows admins, so I don't know their minds [20:05] but my instinct is their thought process is "we should update as little as possible on running servers, unless there is a security issue" [20:05] but, sounds like what sladen is saying is all post-release updates should be considered security issues and thus must be applied. [20:06] odd, I'd swear I've seen post-release updates that wouldn't really fit that [20:06] sladen: you're ignoring SRU's [20:06] nemo: https://wiki.ubuntu.com/StableReleaseUpdates [20:07] sladen: heh. "Bugs which do not fit under above categories, but (1) have an obviously safe patch and (2) affect an application rather than critical infrastructure packages (like X.org or the kernel). " [20:08] sladen: but anyway, sounds like you're saying the only method to differentiate right now is read the description and linked LP and use your judgement [20:09] nemo: or get them to sign up to http://www.ubuntu.com/usn [20:10] nemo: all security updates will come from RELEASE-security rather than RELEASE-updates [20:10] visible in apt-cache policy output, among others [20:10] cjwatson: thanks === almaisan-away is now known as al-maisan === al-maisan is now known as almaisan-away [23:48] what's the name of the program that pops the broadcast accounts dialog === Amaranth_ is now known as Amaranth