[00:01] mathiaz: okay, sent. [00:06] Just out of interest has anyone looked at being able to mount Vista UDF disks in Hardy - just tried it on my server vm and it barfed. [00:07] goodnight [00:08] nijaba: Thanks for the work you did on the surveys - sorry for all the reports :) [00:08] owh: hey, we need testers :) [00:08] You tarzan, me tester :) [00:20] mathiaz: you're right... message awaiting moderator approval [01:43] * faulkes- yawns [01:43] faulkes-: Does your fancy stats code handle IRC logs? === nealmcb is now known as nealmcb[CO] === nealmcb[CO] is now known as nealmcb [02:56] Jeeves_: ping? you about mate? [03:16] ScottK: debian bug 470852 [03:16] Debian bug 470852 in postfix "postfix: Wildcard virtual alias maps cause unwanted bounces" [Important,Open] http://bugs.debian.org/470852 [03:19] * lamont -> bed [03:30] lamont: IMO anyone using domain catch-all's in 2008 pretty well deserves whatever pain they get. OTOH, the behavior isn't ideal. [04:55] owh: no, fancy stats code does not handle irc but it could be adapted to do so [04:56] (yes I realize the reply is probably about 6 hours too late [06:28] kgoetz: I was sleeping. :) [06:28] kgoetz: What's up? [06:29] Jeeves_: The sun? [06:30] owh: Yeah, that did come up a about an hour ago. :) [06:30] Jeeves_: Over here it's already well and truly on its way down again :) [06:32] owh: 07:31, the clock says. I'm sitting here with sleepy eyes :) [06:33] Jeeves_: It depends on which clock you're looking, but given yours I'm thinking Holland :) [06:34] Jeeves_: kgoetz Is on his way home at the moment. [06:34] owh: Correct [06:34] pschulz01: Ah ok. Did he want to talk about the T1000's ? [06:34] (That you know of) [06:35] Jeeves_: Not sure.. we have an older sparc that he is trying to install.. we have passed the T1000 off to anotrher developer who is using Slowaris. [06:35] pschulz01: They're both pinging here. Still running solaris. [06:36] Jeeves_: Amsterdam die groote stad die is gebouwed op paalen ;-) [06:36] I'm thinking I'll give you guys access to my server, running dhcpd an tftpd and access to the serial console of the T1000 [06:36] So you can do your thing and try to make it boot [06:37] owh: :) [06:37] Jeeves_: I think kgoetz would appriciate that :-) [06:37] pschulz01: :) [06:37] Does he have pgp? [06:37] Jeeves_: I would like to know that I can reinstall either OS's. [06:37] Jeeves_: Yes.. [06:38] What's his emailaddress, I'll send him an email with login data [06:38] Now this is weird. Yesterday or so I came here to ask why a daily cron wouldn't run and for debugging purposes the suggestion was to install mail on this server - local delivery only. Now this cron job is running - there were no other changes. The mail has the log output of the cron job that was previously failing. This makes no sense to me. Any ideas? [06:39] owh: Did you restart cron? (I had to do that yesterday on a box after placing files in /etc/cron.d/ [06:39] Jeeves_: I'm trying to see if it's in launchpad. [06:39] You're not going to tell me that a cron job fails because there is no place for the output to go? [06:39] pschulz01: Ack [06:40] owh: Not that I know of [06:40] Jeeves_: Nope, no restart of cron. I'll check the history just to make sure. [06:40] https://launchpad.net/~kgoetz [06:41] Jeeves_: Oooh, you might be onto something there. invoke-rc.d cron reload was the last thing I did :) [06:42] So, cron doesn't detect new stuff in it's tree? [06:42] Hmm, but run-parts --test showed me everything. === joerlend_ is now known as XiXaQ [06:43] Is that a bug or a *feature*? [06:43] owh: That would be a bug [06:44] However, I've never seen it before, except for this week [06:44] I'm not even sure how it could even happen. [06:44] I mean run-parts runs every time cron runs. [06:44] Me neither [06:45] You know, come to think of it, I wonder if in the deeper past I've seen something similar. [06:45] * owh goes to RTFM for a bit. [06:46] It says: "Like /etc/crontab, the files in the /etc/cron.d directory are monitored for changes." [06:51] That makes no sense, the Debian policy manual says: " If a service reloads its configuration automatically (as in the case of cron, for example), the reload option of the init.d script should behave as if the configuration has been reloaded successfully." [06:51] Indicating that I shouldn't need to reload cron. [06:53] Jeeves_: Do you remember the circumstances in which you had to reload cron? [06:54] owh: Yes, we copied files in /etc/cron.d/ [06:54] And they were in crontab format, rather than scripts? [06:54] Jeeves_: I've email kgoetz.. expect an email from him at some stage. [06:54] It seems that .d wants one format, .daily, .weekly, .monthly seem to want scripts. [06:55] s/one/crontab/ [06:55] pschulz01: I'm allready emailing him :) [06:55] Jeeves_: Which version of OS? [06:56] gutsy [06:56] Ditto [06:56] Better test hardy :) - I'll run it on my vm. [07:00] Jeeves_: That will be interesting. I've created a demo cron job on both hardy and gutsy without reloading cron. I'll see what happens. [07:03] pschulz01: He's got email :) [07:18] Jeeves_: Well, my hourly cron job just ran without any problems at all. No reload needed - gutsy workstation. (In case you're wondering, I setup a cron.hourly to work the same as cron.daily, very handy) [07:23] <_ruben> hmm .. thought i started work early, but jeeves beat me to it for quite a bit it seems ;) [07:37] _ruben: Sorry bout that :) [07:39] Er... You never need to reload cron. [07:39] Regular users can have a crontab. It wouldn't be very useful for them if they had to e-mail their admin to get them to reload cron every time. [07:40] <_ruben> hehe [07:43] _ruben: no Jeeves_ at the office right now. so i guess you were earlier :) [07:43] soren: I realise that this is how it should be, but both Jeeves_ and I have experienced a situation where a job wouldn't run until cron was reloaded. [07:45] owh: I'm not sure how often it reloads or how it determines what to reload.. [07:45] owh: But if it's failing to reload at all, that's a *serious* bug. [07:46] Well, I had a daily cron job that hadn't run for 5 weeks. Now it magically does. [07:47] soren: According to suggestions here I installed postfix on the server in question *and* I reloaded cron. [07:47] soren: It's possible that the cron job didn't run because there was no mail, but the job itself does not use any mail at all. It does however generate backup logging output. [07:48] <_ruben> henkjan: guess he's mindcontrolling his computer from afar then ;) [07:48] soren: I'm uncomfortable with the notion that a cron job wouldn't run because output is not going anywhere and that would cause it to fail. [07:48] _ruben: irssi + screen on his colo server. And working from his computer at home :) [07:49] owh: I can understand that [07:49] soren: The very last thing I did three days ago on that server was reload cron, and low and behold, now the script is running. [07:51] I've just created daily scripts on three different machines, hardy-server, gutsy-server, gutsy-workstation to see what gives. [07:52] <_ruben> henkjan: guess "work" can be a wide subject .. irssi + screen here as well .. actually currently trying to find an irssi script/plugin i used to use ages ago which shows a horizontal line to indicate which part of a channel i've already 'read' [07:58] Heritage computers facing scrap heap: http://www.abc.net.au/news/stories/2008/03/14/2189966.htm === joerlend_ is now known as XiXaQ [08:40] moin [08:43] Jeeves_, ping? (just got home) [08:45] Jeeves_, i have an encrypted email from you - i need to talk to you about it :) [08:46] <_ruben> hehe [08:46] :P [08:47] owh, sad :( [08:47] <_ruben> Jeeves_ probably fell asleep again or so ;) [08:47] hehe [08:49] Jeeves_ is at the office right now [08:49] just not at his desk [08:50] Kamping_Kaiser: What's sad? [08:50] owh, chucking out history [08:50] Kamping_Kaiser: maybe i can help you? [08:51] Kamping_Kaiser: Ah, now I get it. Yes. [08:51] henkjan, i needed to ask him about his gpg key, but i might have just been pebkacing [08:51] woot. my fault :) [08:51] henkjan, thanks anyway [08:51] okay, problem fixed :) [08:52] :) [09:00] <_ruben> grrr.. stupid windows backup server .. its on gbit .. but most of the time im getting 30Mbps to it, and occasionally 300Mbps .. no fun copying 200GB at 30Mbps [09:01] Try 60Gb at 5k [09:01] afk a while, having a liedown [09:02] <_ruben> owh: rather not ;) [09:05] _ruben: Chicken :) [09:53] Kamping_Kaiser: I was at the dentist [09:55] <_ruben> scaru [09:55] <_ruben> scary too [10:38] it stupid that we don't allow degraded raid array to boot :/ [10:39] <_ruben> that'd be rather stupid indeed [10:40] where should i mount root partition in busybox? to /root? [10:50] ivoks: Who doesn't allow degraded array's to boot? [10:53] ubuntu :) [10:55] ivoks: file a bug, I find it sad as well... [10:56] https://bugs.edge.launchpad.net/ubuntu/+source/mdadm/+bug/125471 [10:56] Launchpad bug 125471 in mdadm "Booting from a degraded array could be improved" [Undecided,New] [10:57] i'm fighting with degraded / over the phone for 2 hours [11:01] ivoks: hmm, an old one. I'm going to see if we can do something about it in hardy [11:15] argh... now i have to go to the city 100km away :/ === \sh_away is now known as \sh [11:57] soren, nijaba: is there a particular reason why you are disabling acpi in ubuntu-vm-builder when using --libvirt? [11:57] morning [11:57] hi zul! [11:57] jdstrand_: no idea [11:57] hi jdstrand_! [11:58] nijaba, soren: cause on hardy 386 in a vm, it is needed to use the keyboard [11:58] I don't know all the combinations of releases/archs need it though [11:59] * jdstrand_ ponders 5 releases, 2 archs, and (at least) 3 kernel flavors [11:59] and acpi on/off [11:59] so 60 combinations for ubuntu only [12:06] jdstrand: Oh, that's because I'm an idiot. [12:06] soren -vv [12:07] (when -v just isn't enough;) [12:08] It's quite possible that the libvirt xml I used as a template was for running WinXP. [12:14] jdstrand: The last ubuntu-vm-builder works as expected, right? [12:15] soren: I don't know about the keyboard cause I would always get the grub Error 15 [12:16] Still? [12:16] What's the command line you use again? [12:16] soren: I don't think u-v-b ever created the === cjwatson_ is now known as cjwatson [12:17] for 0.3: [12:17] sudo ubuntu-vm-builder -d ./hardy_386 --domain strandboge.com --hostname hardy-386 --mirror http://192.168.122.1/ubuntu --addpkg openssh-server -s gutsy --vm kvm --user james --part ./partfile --kernel-flavour 386 --components main,restricted,universe,multiverse --ip 192.168.122.10 --libvirt qemu:///system -a i386 [12:17] soren: note I used '-s gutsy' here, cause of the grub error [12:17] soren: I then upgraded [12:17] I'd like to see the cmdline that causes grub error 15. [12:20] soren: do you mean the u-v-b invocation, or the output from grub when running u-v-b? [12:20] The former. [12:20] soren: let me downgrade u-v-b and try again [12:21] jdstrand: Nono, use the latest. [12:21] soren: that isn't where I saw Error 15, but ok (I haven't tried the new one yet) [12:21] <_ruben> hmm .. afaik there's a way to do the installation of ubuntu-server over ssh .. cant find how that was done .. google keeps showing me howtos on how to install ssh :p [12:24] _ruben: google for network-console-udeb, IIRC. [12:24] <_ruben> soren: lets give that a try [12:24] soren: I am running this now: sudo ubuntu-vm-builder kvm hardy -d ./hardy_386b --domain strandboge.com --hostname hardy-386b --mirror http://192.168.122.1/ubuntu --addpkg openssh-server --user james --part ./partfile --kernel-flavour 386 --components main,restricted,universe,multiverse --ip 192.168.122.11 --libvirt qemu:///system -a i386 [12:24] soren: and got this error (but it continued on): [12:24] /usr/bin/ubuntu-vm-builder: line 193: validite_arch: command not found [12:24] soren: I'll let you know when I boot into it [12:25] Ah, maybe I need to spell validate properly? [12:25] oh! [12:25] I happen to know the arch is valid ;) [12:29] soren: crud, apt didn't finish upgrading to ppa4 [12:29] soren: trying again [12:34] soren: rm: cannot remove directory `/tmp/vm-builder-QZXMYX8296/root/lib/modules/2.6.24-12-386/volatile': Device or resource busy [12:34] soren: *but* no grub error [12:35] \o/ [12:35] soren: and no [12:36] soren: but it won't boot [12:37] ACPI: Unable to load the System Destription Tables [12:37] PCI:... [12:37] Loading, please wait... [12:37] and hang [12:38] bleh. i felll asleep [12:39] <_ruben> hehe [12:39] soren: I added the bits and rebooted and it still won't boot... [12:41] soren: interesting [12:42] soren: well, maybe not [12:42] nm [12:42] soren: well possibly :) [12:44] soren: ok, I disabled splash and quiet and it hangs at 'Uniform CD-ROM driver Revision: 3.20' [12:45] I do have a warning about sda and write cache disabled... doesn't support DPO or FUA [12:46] eh heh, it dropped me to initramfs [12:47] /dev/hda1 does not exist [12:47] soren: ^^ [12:47] is this the uuid stuff? [13:03] jdstrand: Sorry, I was on a call with Rick. [13:03] np-- I will be now too :) [13:03] Oh, I thought mathiaz was next :) [13:04] I think I know what at least part of the problem is. [13:05] <_ruben> hmm .. only gettin 1MB/s from the nl mirror .. sucks downloading isos as such low speeds :p [13:05] *shrug* asked for in 5, I said 'ok' [13:11] jdstrand: The problem seems to be that the virtual kernel is not currently installable. [13:12] _ruben: that must be your connection [13:14] <_ruben> henkjan: very well possible .. was getting only 500KB/s to utwente (first dutch mirror in the list) [13:15] <_ruben> even tho im on 100Mbps fiber [13:16] _ruben: bug your provider [13:16] * _ruben fires up a speedtest.net [13:18] <_ruben> hmm .. throughput seems darn crappy at the moment .. strange [13:19] <_ruben> heh .. only 1Mbps to Madrid :p [13:19] <_ruben> ow well .. already completed burning the iso .. no complains for co-workers .. lets hope its just another friday-afternoon-hickup [13:19] get a decent ISP like BIT :) [13:21] <_ruben> henkjan: i doubt you can compete with our current setup .. pricewise .. not sure what we pay, but its no much for a fiber line that's supposed to be 50mbit but in reality gives us 100mbit [13:21] <_ruben> the 100mbit is shared by the complete complex, but the complex is stuffed with low-usage customers [13:22] not today :) [13:22] <_ruben> true :p [13:22] *radiates dislike* [13:22] <_ruben> too lazy to try to figure out where the problem might be [13:23] our fastest internet link at work is 2mb, which is via the usa (and as such has massive latency) [13:24] <_ruben> and contacting the isp isnt that easy .. unless i bypass our very-hard-to-contact direct contact and go straight to our colo provider which happens to be the actual provider of the fiber as well (wedare) [13:24] Kamping_Kaiser: i've got 2x 10G :) [13:24] <_ruben> Kamping_Kaiser: our backup dsl line is already 20mbit i think .. never done any real speedtests on it [13:25] <_ruben> biggest pipe i have avail is the 1gbit uplink we have at our other colo [13:25] _ruben, the fastest link i have access to (itshare sa's) is 8mb~ [13:25] henkjan, :( sounds like a mate of mine at one of the unis in adelaide. his slowest net link is the gigabit to his desktop [13:26] s/net link/link to the net [13:26] Kamping_Kaiser: actually, i'm working at an ISP :) [13:26] henkjan, i gathered :) [13:27] * _ruben should consider a jobswitch :p [13:27] hehe [13:27] _ruben: http://bit.nl/vacatures [13:27] visit australia and see what the 'first world' finds fast (or come to PNG and find out what the 3rd world finds fast...) [13:27] * _ruben clicks on it for the fun of it [13:28] <_ruben> unix engineer might suit me [13:29] _ruben: wel, come join a nice club with smart people [13:29] <_ruben> tho my skills are probably too limited .. they're sufficient for my current place tho [13:29] "Medewerker" - something worker... whats mede? [13:29] _ruben: mwoah, if you are motivated to learn [13:30] * Kamping_Kaiser is amused at english words scattered through the site [13:30] <_ruben> henkjan: motivated i am .. i'd feel sorry for my current company tho, they'd be near dead in the water without me .. ok .. a bit exagurated (sp?) [13:34] <_ruben> Kamping_Kaiser: coworker [13:35] ah [13:36] does anyone has any experience with setting up an WPA Enterprise + Freeradius environment? [13:38] soren: I am not specifying the virtual kernel [13:38] (I was on the phone) [13:38] soren: is it required even with the 386 flavor? [13:39] soren: but anyway-- you did see that it booted and dropped to initramfs, right? [13:42] sergevn, i've not used WPA, but i have used Freeradius (a little) [13:43] Kamping_Kaiser: In what context did you use freeradius? [13:43] <_ruben> henkjan: and ede is a bit far me ;-) (not sure if the pun was intended or not) [13:43] <_ruben> for me even [13:44] _ruben: ah, that would be te biggest issue :) [13:44] to authenticate users against a remote system (no group data/login data, just username/password validity) [13:44] <_ruben> henkjan: currently working in rotterdam and living in zoetermeer .. zoetermeer - ede is an 1 hr drive :p [13:45] <_ruben> without traffic ;) [13:45] _ruben: thats what i'm driving daily (rijssen - ede) [13:46] <_ruben> jikes :p [13:49] * _ruben takes a stroll down to the basement to reinstall one of his sles9 vmware hosts with ubuntu 7.10 [13:57] jdstrand: Gutsy and onwards on i386 defaults to virtual. [13:57] jdstrand: Anything else defaults to server (or amd64-server on dapper/amd64). [13:57] soren: but I specified the flavor [13:57] jdstrand: Oh.. Er. [13:57] jdstrand: Good point. [13:58] --kernel-flavour 386 [13:58] :) [14:00] jdstrand: Yeah, I removed a stack of your options as they didn't apply here (network and mirror stuff). [14:00] jdstrand: And accidentally removed the kernel-flavour, too. [14:00] jdstrand: I'll try again. [14:00] soren: too big of a stack apparently [14:00] :) [14:00] np [14:03] jdstrand: Ok, it works for me now. I'll push to launchpad and the ppa. [14:03] cjwatson: hi there--I have a message pending moderator approval on ubuntu-devel@lists.ubuntu.com... mathiaz said you were a moderator on that list, and could probably push it through? [14:04] Kamping_Kaiser: Hi there! [14:05] Jeeves_, hello! [14:06] soren: thanks [14:07] <_ruben> sweet .. install over ssh seems to work .. thanks for the pointer, soren [14:09] _ruben: Er... Which pointer was that? :) [14:10] <_ruben> the .. uhm .. what was it called again :p [14:10] <_ruben> the network console thingie [14:16] kirkland: done. (It would be nice if the supporting function in lsb-base went to Debian so that we don't have to maintain deltas against every single package with an init script!) [14:16] _ruben: Oh! I somehow assumed we were talking about virtual installs or something. :) [14:18] <_ruben> ah :) [14:19] cjwatson: absolutely, you bet. [14:19] cjwatson: any idea if there's a historical, political aversion to LSB in Debian? (something I should be aware of before embarking on that trip?) [14:25] Jeeves_, i've successfully ssh'd in :) [14:27] <_ruben> crap .. unless im blind then the installer doesnt allow the striping option of logical lvm volumes [14:29] kirkland: Debian adopted lsb-base fairly happily, although I think there has been a bit of controversy over 'status' specifically; I'd search debian-policy archives [14:30] soren: ppa5 is a go! \o/ [14:30] cjwatson: very good, will do. [14:30] soren: hardy/i386 with 386 kernel flavor works great now [14:30] soren: so you are calling grub commands on your own now rather than update-grub? [14:32] Hi, so what's with ubuntu security updates when using ebox packages? On the ebox trac page, it says that the ebox team needs to verify every security update. How is this going to work in Hardy? [14:32] TrioTorus: first I've heard of that from them [14:33] TrioTorus: ebox will be in universe, so it will be community supported, so I guess they will provide the updates for us to upload? [14:34] so, generally speaking, there will be no need to activate an extra repository? [14:34] I would hope not [14:34] TrioTorus: can you give a link? [14:35] jdstrand, http://trac.ebox-platform.com/wiki/Document/Development/DebianSecurityUpdates [14:35] this could well be out of date though... [14:38] kirkland: you should search the debian bug track system in the debian-policy package. [14:39] Kamping_Kaiser: I see! [14:39] zul: ^^ [14:39] kirkland: the wiki page on wiki.debian.org about LSB init script has a link to a bug in debian bts about adding status to the policy. [14:39] zul: do you have a comment on this? I really think the ebox folks should just provide debdiffs in the normal way for universe security [14:40] Jeeves_, :) now i have to brush up on my ALOM and tftp skills [14:40] * kirkland going there now, mathiaz [14:40] _ruben: are you using preseeds ? [14:40] the ebox guys provide us debdiffs and we have a look at them and intergrate them [14:41] <_ruben> mathiaz: no, running d-i in expert mode [14:41] so for ebox we are just the packaging guys [14:41] Kamping_Kaiser: You have experience with that stuff? [14:42] _ruben: you get a message similar to: no root disk found ? [14:42] Jeeves_, a little, from mesing with our sun at work before the dev team got their hands on it [14:42] <_ruben> mathiaz: im still in the d-i completing the installation ;) [14:42] mathiaz: from Wed, 19 Jan 2005 01:30:55 +0100 :-) [14:42] Kamping_Kaiser: I would like to do the second one myself and make nice screenshots and stuff, for our weblog [14:43] Jeeves_, sure. i'll try and doc it as i go to help you when you do yours [14:43] _ruben: there is a bug in the debian installer related to pv and lv - it's not able to wipe physical volumes [14:44] _ruben: hum - I misread your comment [14:44] <_ruben> mathiaz: hmm .. d-i doesnt offer me grub either, only lilo .. perhaps i should go down a diff road [14:44] _ruben: you were looking for striping option in lv [14:45] _ruben: that's probably because you've installed /boot on an lv [14:45] _ruben: grub doesn't support LV IIRC [14:45] <_ruben> mathiaz: ah crap!! .. i created a sw raid for /boot and forgot d-i to use it for /boot .. damnit :p [14:46] * _ruben writes a mental note to self: dont do installs on friday afternoons [14:49] * _ruben goes back a few steps to the partitioner part [14:54] d-i in expert mode> you mean "I am an installer developer and want lots of noise" mode? :-) [14:56] Kamping_Kaiser: Coolio [14:57] cjwatson: are bug 154086 and bug 107205 the same ? [14:57] Launchpad bug 154086 in partman-auto-lvm "Installing to HDD with previous ubuntu fails to create fresh LVM claiming group already in use" [Medium,Confirmed] https://launchpad.net/bugs/154086 [14:57] Launchpad bug 107205 in partman-auto-lvm "LVM install crashed - lvm metadata not removed when creating a new partition layout." [Medium,Triaged] https://launchpad.net/bugs/107205 [14:58] cjwatson: when I do my test, the preseed install still fails if there are previous lv/vg/pv on the disk [14:58] <_ruben> cjwatson: i more or less misinterpreted the expert boot option .. i kinda expected it to take me to the d-i menu, which it did, but additional asks me tons of extra questions as well ;) [14:58] <_ruben> "luckily" my lvm is on top of sw raid .. so with clean sw raid, there's no old lvm :p === \sh is now known as \sh_away [14:59] mathiaz: no, I don't think they are. 107205 depends on whether the group is activated or not; your bug is 154086. [14:59] which depends on whether the group exists on the disk or not [15:00] anyway, there's a preseed for that [15:00] I think [15:00] cjwatson: I think we talked about that - and the preseed doesn't work [15:00] well, that depends on which preseed you were using [15:00] it changed in hardy [15:01] cjwatson: I'm using "d-i partman-auto/purge_lvm_from_device boolean true" [15:02] mathiaz: that's the <=gutsy one; use partman-lvm/device_remove_lvm in hardy [15:02] cjwatson: awesome - I'll try that ! [15:03] cjwatson: that should fix a very annoying problem when I reuse hardware to test installs [15:03] night all [15:04] <_ruben> install complete .. now i gotta walk down to the basement again to remove the cdrom to have it boot properly :/ [15:05] zul: so that will be the case after hardy is released? ie they will provide backported security updates? [15:05] jdstrand: that would be my understanding [15:06] mathiaz: please let me know whether that works; however, I'm surprised that it didn't already ask you the question [15:06] cjwatson: well - I'm not doing manual installs.. Everything is automated. [15:07] mathiaz: in general, I would appreciate it if you could go into more detail about the exact way in which a preseeded install fails, rather than just saying that it fails. Usually a "failure" means that it asks a particular question (which might be something you can respond to, or might be a fatal error) [15:07] mathiaz: but then you should be familiar with using DEBCONF_DEBUG=developer to figure out which question it is asking so that you can answer it! [15:07] it's rarely useful to treat these things in a binary worked/didn't-work way [15:07] does anyone objects to MIR for libqwt5-qt4; it's a simple qt library, having it in main would make it possible to have bacula gui console [15:07] cjwatson: ok - I'll look into that when my preseed fails. [15:07] happy pi day! [15:08] i mean, good morning [15:08] :P [15:08] mathiaz: in particular, if partman-lvm/device_remove_lvm is indeed what you need to set, DEBCONF_DEBUG=developer would have shown that the installer was asking it [15:08] but if it's something else, then the log should be helpful in figuring out what [15:10] cjwatson: thanks for clarifying this. === \sh_away is now known as \sh [15:17] mathiaz: also, happy to help with automation problems on #ubuntu-installer; I pay more attention there than here, in general [15:18] * _ruben should probably go there and "whine" about lack of lvm striping support in the installer ;) [15:20] that's odd, mdcfg has code for raid0 [15:20] oh, lvm striping, not raid striping [15:20] <_ruben> yeah [15:20] <_ruben> nasty way of achieving raid10/raid0+1 (always mix those 2 up) [15:21] you could just use raid5 [15:21] <_ruben> since raid10 is lacking from the installer as well (which would be another solution) [15:21] <_ruben> i need good write performance [15:21] ah [15:21] <_ruben> which basicly means i should get hw raid ;) [15:24] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=384521 though nothing's ever been done with it [15:24] Debian bug 384521 in installation-reports "wishlist: striped logical volume with LVM" [Wishlist,Open] [15:24] I'll reassign it to the right place, at least [15:26] <_ruben> ah, nice [15:26] <_ruben> aug 2006 .. hehe [15:29] <_ruben> time to go home .. weekend [15:30] hi :) [15:33] i've been struggling with updating a server from 6.10 to 7.10. i updated from 6.10 to 7.0.4 and finally got over the problems i had with that (it wouldn't boot) and so yesterday went through the process again to update from 7.0.4 to 7.10 and i again have a server that won't boot. [15:47] this is what happens : http://i25.tinypic.com/25sym8g.jpg [15:48] i booted from the install disc and ran a rescue shell, but their don't appear to be any raid or lvm programs available. what can i do? [16:12] lunaphyte_: the installer's rescue *mode* should give you raid/lvm [16:12] assuming this is an alternate install CD [16:12] (or server install CD) [16:13] at least as of 7.10 [16:14] don't try booting the installer with init=/bin/sh or root=blah or whatever, that won't help much [16:14] rescue/enable=true would be the parameter you want, or the rescue mode on the CD boot menu [16:21] cjwatson: yeah, it's the server install cd. i think i've figured at least one problem out - i have some sort of cd-rom or optical drive issue, although it's not clear what. [16:21] maybe someone here can make sense of it: [16:23] if i boot from the cd on a computer in my office, i can go through the process of getting into rescue mode, etc with no errors. so that says to me, the cd is fine, the iso i downloaded is fine, etc,etc. [16:24] but, if i take that same cd and use it to boot this server i'm working on, it boots, it loads the installer, it reads files off the disc, yet it complains that there is a problem with integrity of the disc, and craps out. [16:25] so on the one hand, i'd conclude that since i proved the disc is good on another computer, it must be the optical drive. but why would the optical drive appear to be able to do everything it needs to do expect for reading this one particular file? [16:25] i guess i'll stop whining and just go find another drive and swap them out. :) [16:25] have you tried cleaning the drive lens? I say this a lot, but it does seem to be a fairly common problem, and distressingly hard to figure out [16:26] there could well be a problem positioned such that it only affects data in particular parts of the disk [16:36] kirkland: confcall today? [16:36] jmtt: we were just trying to figure that out.... [16:37] jmtt: none of us had an invite [16:37] kirkland: please join now. we are still on. [16:37] cjwatson: yeah - that was my first thought. [16:38] i figured it out - the drive was connected w/ a 40 conductor cable instead of an 80 conductor cable. [16:38] that's one of those things that, for me, i always forget to consider because i still don't really understand it. [16:39] jmtt: joining.... [16:39] kirkland: same confcall info as last time if you can find it [16:40] kirkland: i can't private msg you for lack of freenode registration [16:40] jmtt: gotcha, we're looking it up [16:40] kirkland: ok [16:40] jmtt: can you add me to the permanent invite? dustin.kirkland@canonical.com [16:51] ok, i was able to get recovery mode to fully boot. it saw all of my md devices, and all of my lvm volumes, and i was able to start a shell using my root volume. how can i troubleshoot this? [17:04] hi all [17:04] i have some problems here with networking [17:04] i have two nics in a 7.10 server box [17:04] ubuntu-desktop installed [17:05] can ping adsl router on one nick [17:05] can't ping past it [17:05] and now it won't boot so never mind, reinstalling [17:06] * AnRkey_eepc breaks down in tears [17:29] if i have 5 md devices (md0-4), should they all show uuids in the output of vol_id? [17:57] can someone help me with [17:57] this [17:57] please [17:57] ? [17:57] http://ubuntuforums.org/showthread.php?t=724213 === \sh is now known as \sh_away === Administrator__ is now known as AnRkey_eeepc [18:10] how do i enable routing on ubuntu? [18:11] i have 2 nics in this box and I want my second server to be able to get internet access through it [18:41] AnRkey: same way you would do it in almost any linux distribution [18:43] sudo sysctl -w net.ipv4.ip_foward="1" [18:44] or uncomment the line about ip forwarding in /etc/sysctl.conf [18:44] and reboot [18:44] err, no, sudo sysctl -p /etc/sysctl.conf iirc [18:44] man sysctl in other words ;) === ScottK3 is now known as ScottK2 [18:54] faulkes-! [18:54] faulkes-: did you get my message re mysql? [18:56] lamont: Let me know if you want to discuss the postfix bug. [18:56] ScottK: later on or maybe this weekend, sure [18:57] OK. No rush. [18:57] there was another one that came in today wrt not honoring smtpd_recipient_restrictions or some such that I suspect is Pebkac. [19:01] soren, nijaba: I am *loving* the ubuntu-vm-builder/libvirt combination. It and you guys are *awesome* [19:03] jdstrand: message re: mysql? [19:03] * faulkes- looks around [19:03] faulkes-: I'll recap [19:03] faulkes-: I recently did a -security update for mysql [19:03] * faulkes- nods [19:04] faulkes-: it had some significant changes, esp on dapper-feisty [19:04] faulkes-: so I uploaded it to -proposed for more testing [19:04] * faulkes- nods again [19:04] faulkes-: mathiaz thought it might be a good idea to announce it somewhere on the forums [19:04] maybe a developer section? [19:04] faulkes-: let me get the bug... [19:05] bug #201009 [19:05] Launchpad bug 201009 in mysql-dfsg-5.0 "[mysql-dfsg-5.0] fix for several open vulnerabilities in -proposed" [High,Fix committed] https://launchpad.net/bugs/201009 [19:06] faulkes-: really all that is needed is for people to enable -proposed, install the packages and go to the bug and say 'works for me on ' [19:06] * faulkes- nods [19:07] faulkes-: I am not sure where the best place would be for that (I am a total forums newbie) [19:07] faulkes-: or if it is appropriate at all for the forums [19:07] faulkes-: thoughts? [19:08] well [19:08] I think it is appropriate [19:08] ubuntu dev link forum seems most appropriate but it's dead [19:09] zul: did you send your diffs to debian from your samba upload ? [19:09] not yet [19:09] jdstrand: given the nature of the bug, I would say security forum is probably the best place [19:09] its on my todo list [19:10] zul: great - are you using submittodebian ? [19:10] mathiaz: will be :) [19:10] zul: hum... actually I'm not sure it would be usefull in that case as it's not a merge [19:10]