[00:06] guys do you know any usb wireless devices i cant go wrong with, with ubuntu-server [00:13] Could someone give me the name of the command line version of ubiquity? I know i've seen it meantaioned on the ubuntu-server list, but I cant find anything about it on the net [00:14] Kamping_Kaiser: there isn't one. [00:14] Kamping_Kaiser: however you may be thinking of debian-installer (d-i), which is completely different software that is both more powerful and more flexible. [00:14] twb, i'm pretty sure there is. it wasnt an ubuntu project, but was meantioned on the list. [00:14] and no, its not d-i [00:15] OK, perhaps I am not possessing all the facts. [00:16] I'll lurk a while, and if no one knows, I'll have to grave dig my maildir :l === Deevz_ is now known as Deevz === mib_zau58pfd is now known as tommy_nmw [04:07] hi [04:07] let me ask about ubuntu [04:07] hello [04:08] who knows how to set up active directory like Ubuntu domain controller ? [04:25] tommy_nmw: that would be OpenLDAP and Samba, I guess [04:25] I don't know much about it [04:29] so who knows well? [04:29] good evening all [04:41] yes [04:58] I want sshd to check the ssh_known_hosts file for all connections and no the .ssh/known_hosts file that is created for each user... === llragsll is now known as rags [05:32] can I have two gateways configured...with 2 different metrics?? I want it to switch whenever the other one drops..Is this possible? [05:37] yes, and theres tools packaged for that purpose [05:37] cant give personal experiance though === Noah0504|away is now known as Noah0504 [05:51] Anyone around? [07:22] <_ruben> hmm .. im starting to experience more and more "issues" with cronjobs lately .. some work as expected, others down run at all, or dont do anything .. guess it's time to do some more investigating [07:24] _ruben: have you tried to start those jobs manually? [07:24] <_ruben> manually they work fine [07:25] <_ruben> could be an $ENV problem i guess [07:25] maybe, try exporting needed variables in the job script [07:26] also, did you run those jobs as the user specified in the crontab or as yourself? [07:30] <_ruben> both as root, which does have some effects on the $ENV (sudo versus cronjob) [07:34] _ruben: and you use full paths in the cronjobs? [07:39] Hi, All, [07:39] In Ubuntu is .HTaccess access enabled by default? The release I am using is 8.04 Hardy. Please [07:39] <_ruben> rst-uanic: yes [07:40] Odd, because I can't install Drupal from source for some reason [07:40] _ruben: and what $ENV variables do you think you use? [07:40] <_ruben> Alex_21: the allowoverride setting configures what stuff you can put in .htaccess sites [07:40] Does anyone have experiance with instaling Drupal from source? [07:40] <_ruben> rst-uanic: none specifically .. i havent researched it very much yet .. was about to do so [07:53] Can I safely remove console kit somehow? It wants to remove dbus as well [07:53] Why do I need that stuff on a server that will never do gnome sessions? [07:55] <_ruben> dbus depends on consolekit, for whatever reason [08:20] <_ruben> it was a $PATH issue .. my cronjob calls /sbin/ipvsadm .. which turns out to be a shell script which then calls ipvsadm-1.24 and apparently /sbin/ isnt in the $PATH for (root) cronjobs [08:42] _ruben: I know dbus depends on it, do I need dbus? [08:43] <_ruben> Skaag: how should I know? :) [08:44] is it required for proper functioning of an ubuntu server? [08:44] ;-) [08:44] <_ruben> Skaag: I dont know what apps you got installed which might need it [08:44] Apache, MySQL... [08:44] <_ruben> my servers dont have dbus [08:44] ok great [08:44] I assume if an app will need dbus it will say so [08:45] I was just afraid it's required for the basic operation of the OS itself [08:45] <_ruben> it should yeah [08:46] ok removed that stuff... [08:46] it's just that sometimes console kit takes 100% CPU for no reason [08:46] and I noticed it's a known bug [08:46] thanks :) [08:47] <_ruben> nice [08:49] _ruben: is it working now? :) [08:49] <_ruben> rst-uanic: this specific one, yes .. i added a "PATH=/sbin:$PATH" to my script :) [08:50] that's good [08:51] New bug: #345470 in bind9 (main) "Please sync bind9 1:9.5.1.dfsg.P1-3 (main) from Debian unstable (main)." [Wishlist,Fix released] https://launchpad.net/bugs/345470 [08:52] <_ruben> i wonder if i should file a bug wrt the ipvsadm behaviour [08:53] _ruben: my $PATH has /sbin [08:54] <_ruben> rst-uanic: but also in a cronjob? [08:54] don't know [08:56] I think cron scripts declare their own parameters or just use full paths :) [08:59] <_ruben> rst-uanic: well .. i did use a full path to ipvsadm, but that doesnt work, because /sbin/ipvsadm is a wrapper script, which doesnt use full pathts [09:02] _ruben: maybe :) [09:05] <_ruben> filed a bug for completeness sake .. at the very least it is now "documented" behaviour :) [09:06] :) [09:07] rst-uanic: Crontabs can have a PATH declared at the top. [09:09] infinity: yes they can [09:10] 02:56 < rst-uanic> I think cron scripts declare their own parameters or just use full paths :) [09:11] My point is that the crontab itself (not what you run from it) can have PATH and SHELL and other such things declared before the periodic entries. [09:11] man 5 crontab [09:12] infinity: ok, and my point is that it is better to use full paths and have no problems [09:12] Generally not... [09:13] infinity: why not [09:13] ? [09:13] Scripts that use full paths (and worse, compiled code with full paths compiled in) aren't very resilient to the whole reason the UNIX filesystem hierarchy exists. [09:14] /usr/local/bin:/usr/bin:/bin <-- That's arranged like that for a reason. So you can override what's in /usr/bin with /usr/local/bin ... And if you don't want that behaviour, you remove local from your path, not recompile or hardcode. [09:17] infinity: And if I want to write a small scripts for my system? I should think of how this will work on other systems too? [09:18] rst-uanic: Why would you bother hardcoding paths on your own system? You know what you have installed, surely? [09:19] rst-uanic: But even small projects, I tend to write to the same standards as something I'd write for public release, based on the realisation that a good 75% of my "small projects" end up deployed to dozens of machines, at the least, and publically released to thousands/millions, at the worst. [09:20] infinity: I am not saying that declaring $PATH is wrong. I saying that if I want to write some quick script using full path is quite ok. [09:20] rst-uanic: Well, sure, whatever you want to do is "okay for you", by definition. It's your system. I'm just arguing it's not actually "best practice" to hardcode full paths to binaries. [09:21] What if we move something from /bin to /usr/bin or from /usr/sbin to /usr/bin between releases, and your script breaks because of it? [09:21] No hardcoded binary locations, no problem. [09:21] infinity: for sure :) [09:24] <_ruben> good point [09:25] infinity: thanks for your explanations :) === cjwatson_ is now known as cjwatson === CaTeYe is now known as cateye [11:06] New bug: #347206 in bacula (universe) "bacula in dapper (2.2.8) catalog upgrade from hardy (1.3.6)" [Undecided,New] https://launchpad.net/bugs/347206 [12:25] I'm trying to mount my back-up disk to upload on to my server but it can't mount it using "sudo mount -t hfsplus /dev/sdf1" it kicks back saying "wrong fs type, bad option, bad superblock on dev/sdf1. Any help? [12:30] sparky_: Anything in dmesg? [12:31] [ 511.957477] hfs: unable to find HFS+ superblock [12:31] Well, there you go. [12:32] i don't know anything about superblock.... [12:33] The superblock is where the filesystem puts information about itself. [12:33] It's created by mkfs. [12:33] is there a way to fix without lossing my data [12:34] oh so just run the mkfs and it will create the superblock? [12:34] Nono! [12:34] i dont think so [12:34] no! [12:34] sparky_: No! [12:34] ok [12:34] mkfs is what you use to *create* the filesystem to begin with. [12:35] sparky_: are you 100% sure that your device name is right ? [12:35] So if you run it on a block device that already has a filesystem, you'll delete the current contents. [12:35] yes, it is the only one that is 1Tb in size [12:35] sparky_: Try "sudo vol_id /dev/sdf1" [12:35] ...and tell me what it says. [12:37] http://www.pastebin.ca/1369065 [12:38] Why do you think it's an hfs+ filesystem? [12:39] I thought that is what i created it as on my mac, but it looks like its fat32 correct? [12:40] Yes. [12:41] so i would us -t fat32 [12:41] use* [12:41] Just mount it. [12:43] it's not showing anything in the dir [12:54] sparky_: What did you expect to see? Did it mount correctly? [12:54] hello [12:54] i expected to see all my back-up'd files over 600 Gb worth [12:55] ive got a spare server laying around and im thinking of putting ubuntu server on there, im curious to know what i could use a box which has 32gb HDD space, P4 1.66 and 512mb ram for? im currently running a mail server, web server, file server and mssql. but this is all windows based so i couldnt transfer it to linux. any ideas? [12:59] ivoks, hi thanx for helping me last night i got working it looks decent [13:04] Giddion: Not sure. Perhaps you could make it a mail, web, file and mysql server? :) [13:06] soren: have you seend apache directory studio? [13:06] seen [13:07] ivoks: Nope. [13:07] soren: it's ldap browser [13:07] soren: it even has scheme editor and ldif editor [13:08] it's a perfect tool for managing ldap [13:08] the only bad thing - it's for eclipse [13:08] * soren really wishes we didn't have to deal with JAva. [13:09] soren: it's a workstation-side app [13:09] soren: on server, we wouldn't have java [13:09] ivoks: Right, right, I was just looking at the other Apache Directory stuff. [13:09] soren: ah, apache directory - ldap implenetation [13:10] grr... keyboard [13:10] hey, they have triggers :) [13:11] ahasenack: I'm almost sure that doesn't outweigh the Java thing. [13:11] sommer: correct :) [13:30] soren, thx for the help I got it going [13:31] Cool. [13:36] New bug: #347098 in samba (main) "Samba crashes Windows XP Home" [Undecided,New] https://launchpad.net/bugs/347098 [13:38] g'morning [13:40] we've just sent a server off to a data centre, however we've ran into quite a problem that was not noticed before we sent it off. The partition that mounts as / is 250mb in size! Is there a way to somehow move the data that is current on the / partition somewhere else, re-mount that temporarely, increase the size of the partition, then move it back? [13:42] you could create new partition [13:42] of the same size [13:42] dd data to it [13:43] fix boot loader and fstab to boot from the other partition [13:43] fix the first partition [13:43] copy data back [13:43] fix boot and fstab and reboot [13:44] of course, be carefull and don't come looking for me if it doesn't work :D [13:44] :P [13:46] AlexC_: Is it the only partition on the system? [13:47] ok, so we've 5 partitions - one each for /home /tmp /usr /var and /. Trouble is we have no more space left to make a new partition, so I'm going to have to kind of do this 'on-the-fly' /home is 213GB, so maybe I could move all partition data into there and only mount that, then delete all others, fix sizes, copy data back over? [13:47] AlexC_: Can I see you partition layout, please? [13:47] well, then copy everything to /home [13:47] soren, sure [13:48] oh LVM [13:48] why we love the [13:48] soren, http://paste2.org/p/169698 [13:48] thee* [13:48] AlexC_: Where's sda1 and sda2? [13:49] and sda4 [13:49] swap? [13:49] sda4 would be extended [13:49] when i create a new user in ubuntu server auto complete and bash history doesn't seem to work out of the box as opposed to the first created user [13:49] and sda7 [13:49] giovani3: There doesn't have to be a sda4. [13:49] soren, they do exist, just not mounted. They are some recovery partitions that ideally we could get rid of (I want to), though boss wants them to be kept there [13:49] anything special i should look at in /etc/profile or something like that [13:49] why was this not LVMed? [13:50] AlexC_: That's why I wanted to see your partition layout. The output of fdisk -l or parted print or something. [13:50] soren, ok [13:50] well, having / on 260MB partition in this case isn't that bad [13:50] when will people learn to use single big partition for system, and then big big partition for service data :) [13:50] I just have 10G / and huge /a ;-) [13:50] soren, http://paste2.org/p/169702 [13:50] /usr is separate, /var is separate... [13:50] domas: Hopefully never :) [13:51] ivoks, it sure is, as I can't do any updates to the server due to it running out of disk space [13:51] i guess kernel is creating problems [13:52] AlexC_: Shame. I was hoping we could just extend your root partition. [13:52] having boot bind-mounted to directory in /var would help :) [13:53] soren, same here. If I was to delete sda1 and 2, that would be possible? [13:53] though they are very small, but it would give us something to play around with [13:53] AlexC_: Technically, yes. I wouldn't, though. [13:53] AlexC_: if i were you, i would delete sda1, sda2, sad3 and sda5 [13:53] AlexC_: sda2 is several GB. [13:53] soren: why? :) [13:54] AlexC_: and then create / on top of all of them [13:54] i'm not sure that's possible, since three of them are primary and one extended [13:54] domas: Why on Earth would you do something like that? [13:54] sda1 is marked type dell utility, sda2 is fat32, so i'm guessing server's recovery partition? [13:55] oh, hmm, was scrolled up [13:55] ignore me [13:55] Deeps, correct [13:55] soren: why on earth would you constrain yourself in /var, /tmp, etc ;-) [13:55] though I see no need for them (recovery partitions) [13:55] soren: i understand ubuntu has specific philosophy (like storing essential database files in /var/run ;-)) [13:55] right... [13:55] that's very stupid :) [13:55] we should fix that [13:56] domas: Eh? [13:56] I can give you 10 things that are idiotic about mysql package on ubuntu [13:56] I filed one of them, and nobody even looked at the bug :) [13:56] domas: pid in /var/lib/mysql? :) [13:56] is there a reason you can't just dump all the data on a drive, and start over? [13:56] ivoks: binlogs in /var/run [13:56] domas: that's right [13:56] ivoks: what is right in that? [13:57] giovani3, server is physically few hundred miles away from us now, though with the RAC we have it *may* be possible, however I would not like to do this [13:57] anyway, mysql writable ~mysql is as bad =) [13:57] nothing, i ack on that bug [13:57] "CHECK TABLES" at restart is _extremely_ annoying on terabyte tables, you know.. ;-) [13:57] well, even on 100GB databases it is quite bad. ;-) [13:58] soren: see, people put arbitrary limits where they don't need them, and don't have proper setup for anything else [13:58] soren, what are the technical reasons for not deleting sda2 and extending sda3 into that space? [13:58] AlexC_: your datacenter doesn't offer kvm? [13:58] you could ask to borrow a usb drive, backup your stuff [13:58] domas: could you summarize all those pitfalls? [13:58] giovani3, I don't know tbh [13:58] domas: so that we can fix them all [13:58] then ask them to put some install media in the cd drive, and start over with LVM! [13:58] AlexC_: well I'd be asking if I were you [13:58] ivoks: as long as someone will listen, sure [13:58] domas: i will listen [13:59] ivoks: lately I find out that people who pay me listen to me way more than whose who don't ;-)) [13:59] ivoks: ok! [13:59] AlexC_: The fact that sda2 is located *before* sda3. [13:59] You can't extend backwards. [13:59] ah [14:00] domas: The problem isn't that people contrain themselves. The problem is that it's too tedious to move stuff around. [14:00] domas: if you file a bug on LP, add me (ivoks) as surbscribed [14:00] domas: or, if you'd preffere to send an email, ivoks at ubuntu com [14:00] ivoks: oki [14:01] soren, would formatting sda2, then copying all data from sda3 to 2, then delete 3, extend 2 into 3 work? [14:01] ivoks: e.g. https://bugs.launchpad.net/ubuntu/+source/mysql-dfsg-5.0/+bug/293258 - is there for quite some time already [14:01] Launchpad bug 293258 in mysql-dfsg-5.0 "mysql user has home directory writable by mysqld" [Undecided,New] [14:01] AlexC_: Yes. [14:02] soren, is that a yes, it is a good idea, or a technically possible yes - but shouldn't really, yes? [14:02] ivoks: it isn't critical, but quite bad practice (used by many.. ;-) [14:02] i know [14:02] AlexC_: Probably a good idea, yes. === scfh is now known as scfh_ [14:02] i have to fix conf on couple of my installs [14:02] and i forgot to fix all that [14:03] so, once you have a list, we'll fix all of them in one go [14:03] oki! [14:03] soren, ok cool, could dd handle the difference in partition sizes if I was to do a direct copy of sda3 to sda2? [14:03] puts some pressure on me now! [14:03] or, should I format sda2, resize that to the same as sda3 - then copy, then extend? [14:03] bbl === scfh_ is now known as scfh [14:04] ivoks: there're also few bugs, mysqld_safe loops at 100% after mysqld restart, etc [14:04] AlexC_: Sort of. [14:04] AlexC_: It won't break. :) [14:04] domas: i've seen that one, too :/ [14:04] but ... ? :P [14:04] ivoks: I just really have to sit down and review that, maybe should ask my managers to approve this to do that on my payroll [14:06] hehe, I have separate big /tmp only on boxes that have LVM snapshot roundrobin configured [14:06] does 50G /tmp sound ok? [14:07] ivoks: kudos though for having apparmor profile [14:07] ivoks: I'm giving 'mysql security' talk soon at MySQL Conference, and it will be one of hilights in it =) [14:07] AlexC_: Sorry, I was on the phone.. [14:07] soren, no worries [14:08] AlexC_: Just dd it over. The filesystem will still be those 200-300 MB, so once you've joined the partitions, you should resize2fs it. [14:08] ...as that will extend it to be the size of the underlying block device. [14:09] domas: well, mysql's apparmor isn't my work :) [14:10] soren, ok, so: 1) Format sda2, 2) dd sda3 to sda2, 3) delete sda3, 4) extend sda2 5) resize2fs ? [14:10] AlexC_: don't bother formatting sda2. [14:11] oh really? [14:11] would I not then end up with half of sda3, and half of sda2 on the sda2 partition? [14:11] dd'ing sda3 onto it will overwrite that anyway. [14:11] You would anyway. [14:11] bbl [14:11] mkfs doesn't go through every bit of the block device zeroing it out. [14:11] It just puts a superblock there. [14:11] ahh ok [14:11] ivoks: hehe, now that hardy will fix my apparmor bug soonish, I'll be able to deploy it widely \o/ [14:12] soren, ok, I'll start on it =) This could be fun, since if I can't boot into the server I'm fecked =3 [14:12] AlexC_: Be careful. Make sure grub know where to look for stuff. [14:12] New bug: #293258 in mysql-dfsg-5.0 (main) "mysql user has home directory writable by mysqld" [Undecided,New] https://launchpad.net/bugs/293258 [14:12] Or lilo or whatever you're using. [14:13] soren, yes, indeed. Just to confirm: 'dd if=/dev/sda3 of=/dev/sda2' is correct? [14:13] That looks about right, yes. [14:13] ok, cool - thanks for all this btw [14:14] and it is ok to do while sda3 is mounted? [14:16] I would probably remount it readonly. [14:17] ...make sure you have a root terminal open ready to remount it rw. [14:21] soren, hum, and the server would still boot up ready for me to SSH in if I was to edit fstab, change to readonly, and restart server? [14:22] Oh, you can remount while running. [14:22] sudo mount -o remount,ro / [14:22] mm, interesting =) [14:24] soren, I'm so hesitant to do this, as it's a long trip to London if I mess up. I am currently SSHed in, so I remount, dd, then do same command once finished (though changing 'ro' to 'rw') ? [14:25] if you've not got time just say and I'll poke someone else [14:26] AlexC_: Again: Make sure you've got a root prompt somewhere else, ready to remount rw. [14:27] ..in case something goes bad. [14:27] soren, I'm not quite sure I understand you, I can only login via SSH - would that do? [14:28] have another ssh session open at a root prompt [14:28] Open up another terminal, ssh into the server, "sudo su -". [14:28] ok [14:28] Which boot loader are you using? [14:29] grub [14:30] Alright. Tell grub that your root partition is /dev/sda2 [14:30] ...since sda2 and sda3 will have identical UUID's, so I'm not sure which one will be chosen. [14:30] ok, I shall do that now. One thing I have just thought, we'll need to make sda2 bootable, or will dd copy that data across? [14:31] dd won't copy it across, no. [14:31] ok [14:31] It's a flag in the partition table. [14:31] Not in the filesystem superblock. [14:32] ok that makes sense, and is something we'll need to do once dd has completed then [14:34] ok, I did the remount - though 'mount' says it is still rw [14:34] mount looks at /etc/mtab [14:34] ahh ok [14:34] time to dd [14:37] right, dd finished [14:38] mount / back as rw, mount sda2 just to see if data has been copied correctly is next step? [14:41] ok that appears to have worked, though I am a bit concerned fdisk still says it is a FAT32 partition - is that because it just hasn't updated, or has everything been copied to fat32 (and more importantly, will that be 'ok')? [14:41] fdisk doesn't know about the contents of partitins. [14:42] It just sees that the partition's type is that. You can just change it to "Linux". [14:42] (type 83) [14:43] ah ok, so that is just a ... 'label'/name for it? [14:43] Sort o, yes. === cateye is now known as CaTeYe [14:46] soren, is it worth changing it to 83? Also, how do I make this partition bootable? Btw I shall send you a donation once I have done all this, as your help is extremely appreciated [14:46] AlexC_: I would probably change it to avoid confusion. [14:47] I don't know if the bootable flag really makes any difference, but it's something you do from fdisk as well. [14:47] I thought bootable was quite essential, other wise grub couldn't boot it? [14:48] I doubt grub cares, to be honest, but seeing as there's very little margin for error, you should probably set it :) [14:52] ok, awesome - that's done [14:53] technically I should now be able to reboot and everything *should* work, on the new partition? Then after that I simply delete sda3 and increase sda2 size? [15:00] That's the idea, yes. [15:30] anyone here familiar with openerp? trying to run the server in jaunty and it fails. comp [15:30] complains about xpath [15:31] ERROR: Import xpath module [15:31] ERROR: Try to install the old python-xml package [15:31] ScottK: ^ any ideas? [15:31] install python-xml package? [15:31] heh [15:31] the instructions are a clue [15:31] ivoks: its installed, obviously. [15:31] jaunty is also ... not stable [15:31] giovani3: Im not a n00b :P [15:32] rgreening: file a bug report? [15:32] rgreening: By design python-xml is not in the default path. [15:32] giovani3: I'd rather he fix it. [15:32] :) [15:32] ScottK: and I am capable, with a hint... [15:32] as you know [15:33] rgreening: I don't recall what you have to add to the path to see the old python-xml stuff, but that's what's needed. The python-xml package should yield sufficient clues. [15:33] So ScottK, does that mean, I need to mod the openerp server script to update the PYTHONPATH? [15:33] /usr/share/pyshared/oldxml/ [15:33] rgreening: I think so. As ivoks says. [15:33] just ignore me :) [15:33] cool. [15:34] ScottK: I'll take a look [15:34] thanks ivoks [15:34] dpkg -L python-xml might help [15:34] Great. IIRC it's in Universe so you can upload it. [15:34] i'm not sure my answer is correct [15:34] cool [15:34] np, I'll look around see if I can work it out. [15:36] soren, still alive? It's booted up into sda2 =) So far so good, what did you say I needed to do to make sda2 as big as I needed? [15:36] just use fdisk to change it? (well, mount sda2 as ro first I would assume?) [15:37] AlexC_: No need. [15:37] AlexC_: Just remove sda2 and sda3, recreate sda2 with the start sector of the old sda2 and end sector of the old sector3. [15:37] Err... s/sector3/sda/3/g [15:38] soren, I can't remove both, if I do that I loose all data on them. Currently sda2 is the copy of sda3 [15:38] Yes, you can. [15:38] ScottK: prob this path... /usr/share/pyshared/oldxml [15:38] Removing the partition does not touch the data on it. [15:38] AlexC_: ^ [15:39] rgreening: Since that's the one ivoks suggested, I'd tend to think yes. [15:39] hmm. tried exporting it. no go.. [15:39] soren, ok, that makes sense. And I do this all with fdisk, right? [15:40] AlexC_: Yes. [15:43] ivoks: heh, I see intrepid has a bit better scripts than hardy :) [15:43] domas: still, we should fix it in hardy too [15:43] domas: For what? [15:43] mysql [15:46] New bug: #347361 in net-snmp (main) "libsnmp-python segmentation fault" [Undecided,New] https://launchpad.net/bugs/347361 [15:48] soren, ok, that is done. I know have '/dev/sda2 * 12 307 2377620 83 Linux'. However, 'df -h' is still showing /dev/sda2 as 260mb, what is the next step after this? [15:50] rgreening: sys.path.append('/usr/share/pyshared/oldxml/_xmlplus') will get you started, but then import xpath dies in progress leading to more clues. [15:51] ScottK: yeah. They import boolean (which is part of python 2.6.. and other mess...) [15:51] So I think that leads to more work worth doing. [15:52] hehe [15:52] rgreening: It's actually xpath that imports boolean. [15:52] I wanted to demo this software to someone today... no go for Jaunty... [15:52] Intrepid will be similar probably as python-xml is out of the path there too. [15:54] ivoks: damn you people, quite a few problems have been fixed in intrepid :) [15:54] ivoks: I was looking at all init scripts and saw "oh, they changed this, they changed that" :) [15:55] ivoks: I'm tempted to set up PPA for hardy with intrepid packages [15:55] ScottK: right from the python intrepreter, importing xpath is broken, so it leads me to believe there's either some voodoo needed or the oldxml stuff is just broken.. [15:56] I would tend to suspect an incomplete python-xml update for Python 2.6. [15:57] am I right in thinking, resize2fs /dev/sda2 will resize the filesystem to take up all the partition [15:58] oh my, whats a rats nest i've uncovered.... [15:58] ScottK: Bug 337759 [15:58] Launchpad bug 337759 in openerp-server "openerp-server depends on python-xml and is therefore uninstallable ATM" [Undecided,New] https://launchpad.net/bugs/337759 [15:59] * ScottK was looking for In Progress -> rgreening.... [15:59] ScottK: its an existing bug report [16:00] Yes, and? [16:00] g'bai [16:00] rgreening: You're allowed to fix ones you didn't report. [16:00] I know, I was just indicating I found an open one to work against :) [16:02] ScottK: seems to me, modifying the oldxml to not load boolean may be whats necessary. [16:03] rgreening: From what little I know about this, that seems likely. [16:03] It's dead upstream, so don't feel bad about patching it. [16:03] though what breaks outside of that, is unknown [16:03] ok [16:03] I'll give that a shot [16:09] ScottK: I expect a big patch to remove boolean may fix this. It's in 44 places in the package. [16:09] That's not so bad. [16:09] :) just time consuming.. but hopefully this is all thats required to fix openerp [16:23] hello everyone. noob question: i need to add php-imap support to a current ubuntu server setup. if i run "sudo apt-get install php5-imap" will that leave the current setup intact and just add on imap? or do i need to run that command with all the modules/addons i currently have enabled? [16:24] broheem: it will install the module, it doesn't remove your other modules [16:24] giovani3: perfect, thank you! i wasn't sure if that was replacing php5 with the new module or just adding it on. that answer my question. take care. [16:25] broheem: no problem -- that goes for all modules in debian/ubuntu -- in some cases, though, there will be multiple packages that replace one another such as xxx-mysql versus xxx-pgsql [16:25] because only one kind of support is built in at a time, for example [16:26] giovani3: ah, yes, that makes sense. thanks for the info. much appreciated! [16:26] np [16:27]