=== zz_sailerboy is now known as sailerboy === _Techie_ is now known as _TechAway_ [01:38] yeshuah: The ubuntu server guide (see /topic) tells how to do this. [02:41] * e_t_ is away: Gone away for now [03:34] hello ppl [03:35] whats the command if i want to install packages LAMP, i forgot it [03:35] that which brings you same screen as on server install [03:35] its not ap-get [03:38] that blue screen === JanC_ is now known as JanC === _TechAway_ is now known as _Techie_ [04:00] also posted in #ubuntu: anyone know of a way w/ libvirt and kvm to have separate users that can only access their own virtual machines? [04:01] trying to set up a vm server for our LUG so that new users can have a machine to learn on [04:21] Is there a way to see descriptions for updates using aptitude or other? [04:41] New bug: #644074 in ec2-api-tools (multiverse) "upgrade ec2-api-tools to 1.3-57419 (api version 2010-08-31)" [Undecided,New] https://launchpad.net/bugs/644074 [05:03] Trying to get virtual users to work with VSFTPD. this http://howto.gumph.org/content/setup-virtual-users-and-directories-in-vsftpd/ hasn't worked. Error 530 Login incorrect when trying to login. [05:04] ?? [06:16] New bug: #644097 in clamav (main) "clamAV's scan stop on /sys dir." [Undecided,New] https://launchpad.net/bugs/644097 [06:23] New bug: #644095 in dbconfig-common (main) "dbconfig shouldn't attempt to configure a database when no database server is available" [Wishlist,Triaged] https://launchpad.net/bugs/644095 [07:56] hi when inserting a usb key in a machine, it triggers some kernel uevent. Have I a way to recreate those uvent via command line. I hoped udevadm test --force --action=add would have done the trick but apparently --force is not an option anymore. [08:07] hi, i have question about php fast cgi (fcgid) and php.ini [08:07] can i use a master php.ini and extend it with a .ini per user? [08:08] my setup it with suexec and having a /var/www/username dir per site [08:16] Hello. [08:17] testing 1 2 3 [08:19] hi moon [08:19] *b [08:19] Hi [08:19] You wouldn't happen to know any thing about Amazon EC2? [08:20] no, only s3, never worked with EC2 [08:20] darn [08:27] anyone knows how to use a 2nd php.ini? [08:27] (fastcgi) [08:40] zaas: i've used it === MenZa is now known as lhavelund [08:46] New bug: #644156 in clamav (main) "package clamav-base 0.96.1+dfsg-0ubuntu0.10.04.2 failed to install/upgrade: Unterprozess installiertes post-installation-Skript gab den Fehlerwert 10 zurück" [Undecided,New] https://launchpad.net/bugs/644156 === _Techie_ is now known as _TechAway_ === _TechAway_ is now known as _Techie_ === _Techie_ is now known as _TechAway_ === bigjools is now known as bigjools-afk === _TechAway_ is now known as _Techie_ === _Techie_ is now known as _TechAway_ === mr5v3n is now known as ChanZ === _TechAway_ is now known as _Techie_ === Guest58816 is now known as lau === zenmower_ is now known as zenmower === _Techie_ is now known as _TechAway_ [12:06] mathiaz: I'd fix it for maverick release. This has security implications and is realtively self-contained, I suppose === sailerboy is now known as zz_sailerboy === zz_sailerboy is now known as sailerboy [12:41] New bug: #644288 in mysql-dfsg-5.0 (universe) "mysql-server-5.0 on 08.04 LTS leaks filedescriptors" [Undecided,New] https://launchpad.net/bugs/644288 [12:45] heh [12:45] my bug [12:45] (http://bugs.mysql.com/bug.php?id=35477) [12:56] <_ruben> nice [12:56] I certainly don't understand why people use distro mysql packages [12:57] distros should treat 5.x.* as patch-levels, not as separate versions [12:57] only bugfixes go there usually [12:57] now distro maintainers think they will do better job at bugfixing than hundred-people-teams [12:57] How do I set a default printer? Man lpadmin tells me it's # lpadmin -d [destination], but lpstat -d still reports the wrong printer as default [13:01] Do I have to restart cupsys for this? === jjohansen is now known as jj-afk [13:06] <_ruben> domas: easy of maintenance (have the distro maintainer keep track of the versions/issues/etc) .. but as always, its a double-sided knife ;) [13:07] ok, I tried lpadmin -d on different systems with multiple printers. It doesn't work anywhere. [13:07] _ruben: hehe, true. but still, I think distros should be more aggressive about getting patchlevel releases [13:07] I'm sure more people get bitten by old bugs than new ones :) [13:08] <_ruben> domas: hard to tell, as even known-to-be-bleeding-edge distros (like ubuntu is often categorized) still lag behind quite a bit ;) [13:09] well, ubuntu hardy never had anything else than 5.0.51 [13:09] I guess lucid is now lagging with 5.1 ones [13:09] <_ruben> yeah, should be 6.x ;) [13:12] there's no 6.x [13:13] it is 5.1.41 in ll [13:14] setting default printer reported as 644304. Would this qualify as a server papercut? [13:15] bug 644304 [13:15] Launchpad bug 644304 in cups "lpadmin -d does not set default printer" [Undecided,New] https://launchpad.net/bugs/644304 [13:16] <_ruben> ah, 6.0 has been abandoned [13:18] ye [13:18] 5.5 is next [13:26] New bug: #644308 in unixodbc (main) "package odbcinst1debian1 (not installed) failed to install/upgrade: impossível acessar arquivo: Arquivo ou diretório não encontrado" [Undecided,New] https://launchpad.net/bugs/644308 [13:29] <_ruben> hrm, needa brush up my perl/xml skills stat .. the code i've been producing lately is ugly as fuck :p [13:30] !language | _ruben [13:30] _ruben: Please watch your language and topic to help keep this channel family friendly. [13:31] New bug: #644310 in samba (main) "package samba-common 2:3.4.7~dfsg-1ubuntu3.2 failed to install/upgrade: subprocess installed post-installation script returned error exit status 1" [Undecided,New] https://launchpad.net/bugs/644310 [13:38] _ruben, isn't that a feature of perl, to create self-obfuscating code? :D === sailerboy is now known as zz_sailerboy [13:41] Hi. Im trying to set up some virtual hosts, using libvirt, kvm and qemu and following all guides on the ubuntu website. I used ubuntu-vm-builder to create the guest and everything seemed to work just fine. However i cannot connect to this new guest, and it seems that its not getting an IP thru the default NAT setup. Also I cannot connect to the console via virsh [13:45] bobslaede: "virsh console" is only for text consoles. Use virt-viewer instead. [13:47] ttx, arond ? [13:47] smokie: o/ [13:50] smoser: o/ [13:50] (sigh) [13:50] sigh to you too [13:50] ! [13:50] :) [13:50] now I have to type 4 characters to tabcomplete you [13:50] i was going to ask what i should need to do to get https://bugs.launchpad.net/ubuntu/+source/ec2-api-tools/+bug/644074 in [13:50] Launchpad bug 644074 in ec2-api-tools "upgrade ec2-api-tools to 1.3-57419 (api version 2010-08-31)" [Undecided,New] [13:52] Is there a way I can keep chown the same on ftp uploads using pureftpd ? [13:52] smoser: discussion of benefit vs. risk with the release team. Note that this is technically multiverse, so the rules are relaxed. [13:53] yeah. [13:53] I'd say... given that it's binary-only, the risk is contained [13:54] smoser: when were those released ? [13:54] yesterday. [13:54] ah. [13:54] well, at least announced yesterday. [13:54] so you can't really say they had field testing yet. === bigjools-afk is now known as bigjools [13:54] i can't say for certain that the binaries were not available without mention piror to that. [13:56] that's my only gripe about it. It's a young release. [13:56] smoser: if my vote is asked for, I'd +1. [13:56] benefits outweigh risks. [13:57] well, yes. the only thing that i have to alleviate that is a.) its multiverse b.) we've never (and they've never) released a bug fix release soon after. [13:58] smoser: sounds like you have a good case. [13:58] <_ruben> pmatulis: my apologies, i actually had to reread my own text a few times in order to notice the profanities, what's this world coming to :( [13:59] so i should right up a justification for this and then upload ? [14:00] soren: sorry for the delay. This is a headless server [14:07] bobslaede: So? [14:18] soren: no X [14:19] bobslaede: Is it headless or is it X-less? [14:22] soren: both [14:23] bobslaede: So you're not going to be at its console when you're doing this anyway. [14:23] bobslaede: again: virt-viewer is your friend. Run it on your desktop. [14:23] soren: i dont have any access at all to the guests [14:24] soren: also, my desktop is a windows machine [14:24] bobslaede: Then you're screwed. [14:24] out-of-the-box at least. [14:24] I do have access to the servers console [14:25] You can set up a serial console in the guests or whatnot. [14:25] bobslaede: I thought you said it was headless? [14:25] soren: thru virsh console? [14:25] soren: well, i can connect a screen to it [14:25] soren: its just not at my current location [14:26] it's really quite simple. qemu (and hence kvm) emulates a framebuffer. To look at a framebuffer, you need something that can show graphics. [14:27] Like, say, a monitor. [14:27] soren: which should work thru ssh no? [14:27] monitors? No, they require VGA cables. [14:27] Or similar. [14:28] My knowledge may be slightly outdated. [14:28] soren: does it matter if im connected to the server thru ssh, or at the console? [14:28] All I'm saying is that the solution to your problem is virt-viewer. You can use other VNC clients, but you need to do your own connection forwarding and whatnot. [14:30] bobslaede: Depends on what you want to do. If you want to look at the framebuffer of one of your VM's, it certainly helps to in an environment where such a thing can be easily represented. This could be a modern Linux desktop, like Ubuntu, or maybe Windows if you're happy to do your own ssh+vnc connection plumbing (and generally endure using it). [14:30] soren: so in order for me to access a guest system, i need something with direct access to the guest, X, and virt-viewer (or whichever vnc client) [14:30] bobslaede: What do you mean "direct access to the guest"? === dendrobates is now known as dendro-afk [14:31] soren: since the guest apparently doesnt have an ip address [14:31] bobslaede: Yes, but what does "direct access to the guest" mean? [14:32] soren: on the host system i suppose, since i wouldnt have access to any guests from outside of the host [14:33] bobslaede: You're not really answering the question. You're asking me to verify this: "so in order for me to access a guest system, i need something with direct access to the guest, X, and virt-viewer (or whichever vnc client)". I can't do that when I don't understand what you mean by "direct access". [14:33] bobslaede: You don't have to be logged into the host machine. [14:34] bobslaede: You need a) a vnc client, and b) way to connect to the guest's vnc port. [14:34] bobslaede: The guest's vnc port is on the host. [14:34] soren, but the guest doesnt have an IP, or anyway to connect to it? [14:34] 13:34 < soren> bobslaede: The guest's vnc port is on the host. [14:35] bobslaede: The guest doens't run a vnc server. The hypervisor does. [14:36] but shouldnt the hypervisor then be able to connect to the console? Im asking here because i cannot find any info about libvirt or any of this on google, theres only the default guides, none answer my questions [14:36] Let's start over. [14:37] alright :) [14:37] You want to log into the guest. [14:37] Right? [14:37] Yes, which Ive created with ubuntu-vm-builder for libvirt [14:37] Doesn't matter. [14:37] Its network is screwed, so you can't use ssh. [14:37] So you want to connect to its console. [14:37] Seems like it [14:37] Exactly [14:37] You (wrongly) assume you should use "virsh console". [14:38] I tell you: 12:45 < soren> bobslaede: "virsh console" is only for text consoles. Use virt-viewer instead. [14:38] but is a text console not enough? [14:38] there is no X on the guest aswell [14:38] bobslaede: What happened when you tried? [14:39] 'error: no valid connection' [14:39] And no, a text console is not enough, because 13:26 < soren> it's really quite simple. qemu (and hence kvm) emulates a framebuffer. To look at a framebuffer, you need something that can show graphics. [14:40] alright, sorry, wrong error before [14:40] 'Failed to connect to the hypervisor' [14:42] That sounds like a different problem. [14:42] But it really doesn't change the fact that you need a vnc client. [14:42] QEmu emulates a framebuffer. [14:43] Yeah [14:43] On real hardware, even if you decide to just show text, the monitor still shows it as pixels. Same here. [14:44] soren: so then i would still need X on the host, or a vnc viewer on my windows desktop? === dendro-afk is now known as dendrobates [14:45] hi, amb looking for a landscape(canonical), spacewalk(red hat), xCat (not-debian) suitable alternative for my .deb based infrastructure. Does anyone know any server-management solution with web-based frontend which supports .deb systems? thanks [14:46] bobslaede: You can put the VNC client whereever you want. [14:46] permalac_: umm isn't landscape such a thing? Maybe ebox? [14:47] !away > dendrobates [14:47] dendrobates, please see my private message [14:47] bobslaede: I can't really say putting anything on a Windows desktop is a good idea, but a vnc client on a server also doesn't sound good. [14:49] soren: i will try a viewer on windows [14:51] jrib: it is the thing, but it cost like 30k a year. And there is no way to link unsupported machines. And most of my machines will have no support cause we have some libraries done by ourselfs. :( [14:53] permalac_: does ebox do what you want? Don't really know much about either [14:57] jrib: ebox , as I see , and if nobody else replays, it's a suite to manage services, almost like webmin. What am I looking for is a system manager, like puppet, cfengine or chef, but with web-based frontend. Something to be able to delegate some administrative tasks to others on the corporation. [14:57] permalac_: I see [15:18] kirkland, you around ? [15:21] smoser: barely [15:21] why would one use iptables "downrules" when the interfaces are going down? [15:21] can i call? [15:22] pmatulis: erm ... to stop the firewall so that people can get into the now "down" interface? :þ [15:22] smoser: sure === dendrobates is now known as dendro-afk [15:23] RoyK: how can you access an interface that is down? [15:23] pmatulis: ever heard of something called irony? :) [15:24] RoyK: ok [15:24] I imagine if you had rules that NAT'd to that interface, you'd want to replace them when that if goes down === dendro-afk is now known as dendrobates [15:26] pmatulis: I guess it's just how ubuntu works - you want to get rid of anything not related to the current setup once you don't need it [15:27] shauno: replace them with what? [15:33] Hi [15:33] is this a good channel to get help with setting up a IPSEC tunnel using racoon and ipsec-tools? [15:34] well, just shoot - someone might know [15:34] * RoyK doesn't [15:36] <_ruben> racoon *shiver* [15:37] hehe [15:38] <_ruben> messed with it ages ago, its quite flexible, in fact, its flexibility makes it near useless for the average task [15:39] should i like pastebin my error? [15:39] <_ruben> that'd be a decent start (though i doubt i'd be of any help personally) [15:39] hehe [15:40] well it's like 2-3 lines of error [15:40] will i get beaten for pasting that here? [15:40] <_ruben> pastebin allows for easier referencing in case you need to restate your issue [15:41] well i have problems with my phase 1 negotiation due to send error [15:43] it also says IPsec-SA request for xx.xx.xx.xx queued due to no phase1 found. [15:44] <_ruben> dont bother with debuggin ipsec-sa issues as long as your phase1 doesnt work ;) [15:44] so if someone here has setup ipsec using racoon before, please let me know and i will expalin further and show my configuration file [15:44] Well, the thing is, it did work.. I think.. [15:44] <_ruben> just out of curiosity, why use racoon and not strongswan/openswan/etc? they tend to be much more user/admin friendly [15:45] <_ruben> if it stopped working without you changing anything, perhaps the remote end changed something :) [15:46] ohh [15:46] well i choose racoon, cause i thought it was the easier method [15:47] <_ruben> racoon is very low-level [15:47] so which is best, strongswan or openswan [15:47] <_ruben> need to define every detail "by hand" [15:47] <_ruben> matter of personal preference really .. i prefer openswan, but must also admit i never used strongswan ;) [15:48] <_ruben> both are forks of the old frees/wan [15:48] are they both up-to-date? [15:48] ok [15:48] <_ruben> openswan is actively maintained, dunno about strongswan, but i think it is as well [15:48] ohh yea, it was freeswan that was the obsolete one [15:49] ipsec gives me headache [15:50] <_ruben> ipsec sure is a complex matter [15:50] i think starcraft 2 is the solution to my headache [15:50] <_ruben> heheh [15:50] is that like cussing in here? [15:50] a game, that requires windows or OS X [15:51] <_ruben> as far as im concerned: no :) [15:51] <_ruben> each os has its purposes [15:51] <_ruben> we dont expect you to run ubuntu-server on your desktop ;) [15:51] Daviey: another go at euca lucid? [15:51] You don't? :-) [15:51] <_ruben> correction: *i* don't ;) [15:52] hehe [15:52] <_ruben> i cant speak for others obviously [15:52] well i will have a look at the *SWAN [15:52] se which one has easiest documentation [15:52] <\sh> _ruben: my laptop is running -server kernel on my laptop ;) but that's me [15:53] <_ruben> for openswan be sure to check #openswan (not that active) and its mailinglists, the website is kinda outdated (most resources go into the code, not the web) [15:53] <\sh> oh my...now I repeat already words ... too much server deployment today error [15:53] <_ruben> \sh: still, running -server on a desktop isnt the same as expecting others to do so ;) [15:54] <_ruben> desktop in the broadest sense, thus including laotops :) [15:54] <_ruben> woah, laptops :) [15:55] <_ruben> anyway .. time for me to head home .. bbl [16:19] kirkland, ami-7238cd1b [16:19] is ebs/ubuntu-maverick-daily-i386-server-20100921 [16:19] us-east-1 [16:19] use that [16:21] smoser: would prefer 64 bit [16:21] price goes up [16:21] us-east-1 ami-7638cd1f canonical ebs/ubuntu-maverick-daily-amd64-server-20100921 [16:37] kirkland: do you want my "maverick" script that automatically spawns the cheapest possible instance given the size (defaults to c1.medium because I find myself doing builds more often than anything else) [16:47] Is there a way to see descriptions for updates using aptitude or other? [16:48] Datz: aptitude changelog xyz [16:49] hggdh: thank you :) [16:49] SpamapS, i want ot see it [16:51] hggdh, i386 is fail on my local :( [16:51] hggdh, testing amd64 lucid -> maverick is more useful at this stage, i think [16:51] hggdh: I'm tring somethign like this with no result:" aptitude changlog linux-headers-2.6.32-24-server [16:51] Daviey: pretty much all the failures I got are related to the apparmour bug ttx raised [16:52] hggdh: I realize my error, working, thanks [16:52] hggdh: which bug? [16:53] <\sh> _ruben: that's right :) [16:53] sbeattie: looking [16:53] Datz: try linux-headers-generic [16:54] hggdh: but you got instances running ? [16:54] hggdh: or you had to disable apparmor to get them running ? [16:54] ttx: yes, about 40% [16:54] ttx: no, I did not disable apparmour. Which is weird, but ah well [16:56] hggdh: having a look, thanks === ivoks-afk is now known as ivoks [16:56] hggdh: hrm [16:56] hggdh: please comment on that bug, that will prove I'm not insane [16:56] ttx: what is the bug#? I cannot find it [16:57] ttx: I will do more, I will add a link to the logs [16:57] hggdh: lastest -generic and -server changelog is the same FYI ;) [16:58] Datz: darn! === jj-afk is now known as jjohansen [16:58] Datz: well, if it is installed you can always look at /usr/share/doc/ [16:59] hggdh: Bug 628055 [16:59] Launchpad bug 628055 in libvirt "Instances don't start correctly: Security Labeling error running aa_change_profile()" [High,Incomplete] https://launchpad.net/bugs/628055 [16:59] hggdh: maybe there is a eucalyptus bug in there, after all. [16:59] ttx: thank you; sbeattie ^ [16:59] hggdh: tis not, just wondering if I needed to update to latest kernel right away. :) [17:01] sbeattie: our latest suggestion is that it's not an apparmor bug, rather a libvirt bug that leaves a strange error message pointing to apparmor [17:01] and potentially a eucalyptus bug triggering a weird libvirt behavior triggering an apparmor-related error message. [17:02] the whole thing being a bit clouded by bug report hijacking. [17:02] * ttx will be back in 2 hours [17:03] ttx, sbeatie: I will be uploading full logs (that is, the whole of /var/log) plus the test programme, and will add a link on the bug [17:06] hggdh: okay. It *is* possible that there's a bug in the apparmor kernel code where it might not be logging a rejection of something when it should. [17:06] ttx, sbeattie: rephrasing: on Lucid, I had 63% success; on Maverick, all failed [17:07] sbeattie: so it is consistently failing on maverick, after all [17:11] New bug: #644481 in multipath-tools (main) "95-kpartx.rules uses invalid -Q argument to modprobe(8)" [Undecided,New] https://launchpad.net/bugs/644481 === mathiaz_ is now known as mathiaz [17:31] When I log into my server I am getting this error message [17:31] "/usr/bin/xauth: /home/matt/.Xauthority not writable, [17:31] changes will be ignored" This was not there yesterday and [17:32] I am not sure what it means....any help? [17:34] When I log into my server I am getting this error message "/usr/bin/xauth:/home/matt/.Xauthority not writable,changes will be ignored" This was not there yesterday and I am not sure what it means....any help? [17:50] with virt-manager, is it normal that i cannot remove guests image files when deleting guests? [17:51] (there is a yellow triangle with and exclamation mark) [17:51] filesystem permissions look ok [18:01] New bug: #644489 in multipath-tools (main) "constantly changes /dev/disk/by-id/{scsi,wwn}-* LUN symlinks with multipathing" [Undecided,New] https://launchpad.net/bugs/644489 [18:02] ttx / hggdh: My local i386 Maverick... 2 instances ran, one went from pending -> running -> terminated... the other seemed to go from pending -> terminated [18:02] Daviey: were you able to copy the console.log and .mxl from them? [18:02] .xml [18:03] * hggdh is, meanwhile, netbooting to Maverick [18:03] hggdh, didn't attempt to :( [18:03] but i'm pretty sure it's the same failure [18:03] hggdh, Give me some good news :) [18:04] I am falling from the tenth floor, right now at the third, and all looks good so far ;-) [18:04] Daviey: I should have them in about 30min [18:14] SpamapS: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-announce === Al2 is now known as AlekseyEnergo [18:36] New bug: #644538 in postfix (main) "package postfix 2.7.0-1 failed to install/upgrade: subprocess installed post-installation script returned error exit status 75" [Undecided,New] https://launchpad.net/bugs/644538 [18:39] mathiaz: could you adjust the flowcharts on https://wiki.ubuntu.com/ServerTeam/SRUPolicy -- where is says 'verification-succeeded' to 'verification-done' [18:40] mathiaz: alternatively, if you give me the sources, I can do it [18:45] hggdh: I've attached the origianl dia files [18:45] mathiaz: thank you [18:46] New bug: #644549 in dovecot (main) "package dovecot-postfix 1:1.2.9-1ubuntu6.1 failed to install/upgrade: subprocess installed post-removal script returned error exit status 2" [Undecided,New] https://launchpad.net/bugs/644549 [18:56] smoser: there? [18:56] here [18:56] what happened with --use-loader from uec-publish-tarball? [18:57] what do you mean ? [18:58] I just tried it, from a netbooted install of Maverick -- it is not there anymore [18:58] hggdh, [18:58] cloud-utils https://wiki.ubuntu.com/ServerTeam/SRUPolicy [18:58] dammit [18:59] cloud-utils 0.11-0ubuntu1 [18:59] $ apt-cache policy cloud-utils [18:59] cloud-utils: [18:59] Installed: 0.15ubuntu1 [18:59] Candidate: 0.15ubuntu1 [19:00] oh [19:00] I know what I did :-( [19:00] installed bloody lucid instead of Maverick :-( [19:01] ah well. since it is there, I will test upgrading [19:01] sorry smoser... [19:02] New bug: #644562 in drbd8 (main) "package drbd8-source 2:8.3.7-1ubuntu2.1 failed to install/upgrade: drbd8 kernel module failed to build" [Undecided,New] https://launchpad.net/bugs/644562 [19:07] New bug: #644566 in drbd8 (main) "package drbd8-source 2:8.3.7-1ubuntu2.1 failed to install/upgrade: drbd8 kernel module failed to build" [Undecided,New] https://launchpad.net/bugs/644566 [19:08] hello everyone. I'm running samba and i have installed webmin, and i am noticing multiple entry's displaying in my Network that are from the same box: http://www.upload3r.com/serve/210910/1285092388.jpg I have checked the smb.config and it does not show any other entries [19:08] how can i fix this ? [19:11] As far as I know, webmin is not supported anymore in Ubuntu. Maybe others can verify? [19:11] !webmin [19:11] webmin is no longer supported in Debian and Ubuntu. It is not compatible with the way that Ubuntu packages handle configuration files, and is likely to cause unexpected issues with your system. See !ebox instead. [19:14] ah crap...lol [19:14] need to remove it then [19:14] thnx [19:21] <_ruben> what tend ppl to use nowaydays as fs for a fileserver? [19:23] _ruben: come again please [19:23] _ruben ext4 [19:23] <_ruben> woah, kinda cryptic indeed now i read it back :p [19:24] or reiserFS [19:24] <_ruben> which is the recommended filesystem to use for a fileserver .. was kinda defaulting to ext4, but curious how others might think of it [19:25] _ruben http://en.wikipedia.org/wiki/Comparison_of_file_systems [19:25] I use ext3 for Samba sharing, on a small installation though. [19:25] http://www.linux.org/lessons/advanced/x1254.html [19:26] im using ext4 on ubuntu-server [19:26] ext3 for boot [19:26] <_ruben> for /boot i tend to use ext2 even [19:26] reiser is pretty unmaintained for a long time. [19:27] <_ruben> my previous install was using xfs, but heard quite some horror stories about that one in the mean time ;) [19:27] stick with ext4 [19:28] its made for high performance [19:28] :) [19:28] <_ruben> software raid5 over 6 1tb disks .. performance isnt the key here ;) [19:28] so i get in today and my ubuntu vm has crashed pretty badly [19:28] "task apache2: blocked for more than 120 seconds" [19:28] <_ruben> oh well .. 686mins to go for raid volume initialization ;) [19:29] unresponsive on ssh [19:29] unresponsive on http [19:30] pwnguin, that is a kvm issue [19:30] I have had the same issue before. [19:30] its vmware =/ [19:30] nvm [19:31] vmware console isn't responding to keyboard input [19:31] well not true [19:31] i can switch vts [19:31] but i cant log in [19:31] short of rebooting, is there anything else one can try? [19:32] quentusrex: indeed, a lot of googling suggests kernel bugs [19:35] should i change the default port that ebox uses for better security enhancement. [19:35] How would that improve security? [19:35] pwnguin, I was able to resolve the issue with "sudo /etc/init.d/networking restart" [19:36] quentusrex: i can't log in =/ [19:36] if you can get in any other way [19:37] you can't do it over standard networking [19:37] its a vm [19:37] If spamassasin can filter spam why one need to use SquirrelMail plugins? [19:37] kvm has a recovery console [19:37] ive tried ssh, ive tried the vmware virtual console interface [19:37] and I can get in over that. [19:37] i can switch vts on the virtual console, but the login prompt isn't responding [19:38] it kinda looks like reboot territory. it's not a mission critical system, and it's already as bad as it can get [19:40] i guess the only question is what logs will still be around after a reboot [19:51] Hi [19:52] Jsut notced some new behavior from my Ubuntu-server; its set to forward on email from root to my email account, and the last two three weeks been getting this email about.. twice a ay? [19:52] "start: Job is already running: anacron" [19:52] how do I let another group to install pkgs ? giving them full sudo is the only option ? [19:54] I cant find an up to date package for ClamAV === _TechAway_ is now known as _Techie_ [19:55] u92: What release are you on? [19:55] Im not even sure what Anacron is; I was under the impreeion that you just used.. cron; I checked Crontab but I have no tasks running [19:56] So; Im unsure whats calling anacron; or what its supposed to be doing that its producing the error [19:56] hardy 8.04 [19:56] u92: The version in hardy-backports is ~ up to date. We're working on testing the new release now. [19:57] the 0.96.* release, cause i updated apt, and i keep getting 0.92.* [19:57] u92: Then you don't have -updates enabled. [19:57] 0.95.3 is in hardy-updates and 0.96.1 is in hardy-backports [19:57] New bug: #641771 in samba (main) "caplock not active when the remote to the computer windows" [Undecided,Incomplete] https://launchpad.net/bugs/641771 [20:03] Hi, I don't want to mess this up, so could someone help me write a crontab command that runs 'apt-get update' then 'apt-get safe-upgrade'? === lau is now known as Guest51281 === Guest51281 is now known as lau [20:10] kevinash: Which file are you editing? The user's crontab? [20:10] For y'all's information, I just reported: https://bugs.launchpad.net/ubuntu/+source/libnss-ldap/+bug/644632 [20:10] Launchpad bug 644632 in libnss-ldap "nssldap-update-ignoreusers needs to be configurable to ignore users" [Undecided,New] [20:11] pedahzur: the admin user I've set up (a sudoer) [20:11] kevinash: So you invoked 'crontab -e' as that user? [20:12] pedahzur: From the command line I do 'sudo crontab-e' [20:12] kevinash: [20:12] oops [20:12] kevinash: OK, what schedule to you want for the update commands? [20:13] pedahzur: daily at about 5am (I'm actually OK with doing the timer bit, though) [20:14] kevinash: 0 5 * * * apt-get update && apt-get safe-upgrade [20:14] kevinash: You can do tabs between columns, instead of spaces. [20:15] pedahzur: Ah, so the && means 'wait until the first command has finished'? [20:15] <_Techie_> just wondering, whats the flag safe-upgrade do? [20:15] kevinash: Basically, execute the next command if the previous command completed successfully (exited with an exit code of zero), but yes, 'wait' is implied. [20:16] _Techie_: Truth be told, I'm using 'aptitude', but I was told that most of you folks use apt-get, so I thought I'd phrase it that way. [20:16] kevinash: any reason you wouldn't just install unattended-upgrades? and use the /etc/cron.daily/apt script to manage it? [20:16] New bug: #644632 in libnss-ldap (main) "nssldap-update-ignoreusers needs to be configurable to ignore users" [Undecided,New] https://launchpad.net/bugs/644632 [20:16] pedahzur: Thank you so much. That's extremely helpful [20:16] kevinash: no problem [20:17] <_Techie_> kevinash, if you were told that the otehr day, it was me who told you [20:17] zoopster: If I knew that was possible I'd have done it that way! Thanks for the tip. [20:17] _Techie_: [20:18] zoopster: Do you know if this is available for aptitude too (Don't answr if it's a RTFM, response...) [20:18] kevinash: this==unattended-upgrades? [20:19] zoopster: yes. sorry. [20:21] kevinash: looks like it uses the python apt:: library? so I'd say no [20:28] zoopster: thanks for looking. I guess I'll use pedahurz cron script [20:28] kevinash: it's a project on launchpad.net so you can dive into the code, just a FE for what's already there for apt [20:30] zoopster: I'll take a look. Thnanks for your help [20:30] Anybody know a reference comparing Postfix and Exim? I'm rebuilding my mail system from scratch and not sure whether I should stick with my old MTA? === _Techie_ is now known as _TechAway_ [20:32] howdy all [20:32] quentusrex: well, alt+sysrq+e seems to have fixed part of it [20:32] theyranos: no reference off the top of my head, but in my (somewhat limited) experience, Postfix has been easier to configure. [20:32] silly question: what log files are deleted on reboot? [20:33] pedahzur: i've heard that too, which is why I'm considering the switch. [20:33] * theyranos will now go read the postfix manual [20:33] theyranos: How complex is your setup? You may be able to configure everything you need for Postfix just from the Debconf install questions. :) [20:33] where is the ubuntu custom error page for apache2 on ubuntu server? Is it just the apache default or...\\ [20:34] <_ruben> theyranos: a rather common response to such a question is: use whichever you are (currently) most comfortable with [20:34] pwnguin: that depends on the logrotate configuration. Also syslog configuration. [20:34] nakhlawi: i know dmesg is gonna dissapear [20:34] cd [20:34] whoops [20:34] javatexan: check /usr/share/apache2/error/ [20:34] <_ruben> theyranos: if you're used to exim and know how to tell it to do what you want it to do, why bother switching to an alternative? [20:34] pedahzur: Not very. It's the primary MX for my domain. Messages get dumped in a dovecot maildir. [20:35] _ruben: While I've been using exim for several years, I still get hopelessly confused every time I try to change even the slightest thing about its configuration. So the idea behind possibly switching is hoping to find something easier to manage. [20:36] <_ruben> theyranos: ah ok, that's a pretty valid point then ;) [20:36] theyranos: Yeah, sounds pretty easy. I'd install postfix, answer the basic questions, and see how far that gets you. You can always 'apt-get remove --purge' if you want to start over. [20:36] pedahzur: Fair enough. [20:36] <_ruben> if you got the resources, you could setup 2 vms, one with exim and one with postfix, try to configure them identically (the way you want), and see which works best for ya ;) [20:37] <_ruben> postfix+dovecot integration is quite simple indeed === lau_ is now known as Guest31883 [20:38]