[00:35] RoAkSoAx: hrm.. " * debian/control: Switch to python-support; Add missing Dependencies." .. did you mean switch to dh_python2 ? [00:35] I thought dh_python picked either, and defaulted to -support [00:36] Yes unfortunately the default is support, but there is a migration underway to eliminte everything except dh_python2 [00:37] Actually in debian/control I don't think anything has to be done.. dh_python2 is in the python package [00:46] SpamapS: right but still it depends on python-support [00:47] hggdh_: is there a way someone can generate test data for http://iso.qa.ubuntu.com/qatracker/dllist [00:48] RoAkSoAx: I dont know -- stgraber or ara would be the best bets [00:48] hggdh_: ok cool :) [00:48] stgraber: still around? [00:50] RoAkSoAx: hrm? What from python-support are you depending on? [00:52] SpamapS: build-depends [00:52] SpamapS: is dh_python2 the recommended way by the debian-python policy? [00:53] twb: yes [00:53] Okey dokey. [00:53] dh --with python2 .. or dh_python2 if you're old school [00:53] I don't package python stuff for the main archive, but I'll try to remember that in future. [00:53] Its a transition that only started a few months ago [00:54] post-squeeze? [00:54] But its now policy in the DPMT that if you touch a package, you migrate it to dh_python2 [00:54] for squeeze it was optional [00:54] Shiny [00:54] wheezy I believe they plan to not have pysupport or pycentral anymore [00:54] SpamapS: that's why the changelog says switch to python support :P [00:54] And the win is that we end up with a single python support framework? [00:55] RoAkSoAx: but the point is, python-support will be *gone*. [00:55] twb: a single one, with all of the lessons learned over the last few years, yes. ;) [00:55] ;) [00:55] just in time for python 3 [00:56] SpamapS: it doesn't really matter in my package cause im installing in debian/tmp/usr/lib/python*/*/ [00:56] Yeah, I remember what it was like when debian-haskell were trying to get a working support structure. [00:57] While arch had a completely automated build whenever upstream made a new release :-P [01:00] RoAkSoAx: why doesn't that matter? you're breaking w/ policy by putting files directly there. They're supposed to go into /usr/share/pyshared [01:00] RoAkSoAx: which, dh_python2 handles [01:01] essentially as long as setup.py lists your modules.. you shouldn't need to do anything [01:02] SpamapS: do ytou hve a link to the policy for dh_python2 [01:04] Heh.. it would appear that this is all "mailing list policy" from the wiki's and web pages. How frustrating. Let me dig a little deeper. [01:04] SpamapS: cause dh_pysupport is "dh_pysupport is a debhelper program that will scan your package, detect public modules in /usr/lib/pythonX.Y/site-packages," [01:05] obviosly the manpage is outdated [01:05] but that's why I'm installing there [01:05] how embarassing. ;) [01:05] for me I mean. ;) [01:06] lol [01:06] SpamapS: anyways, first time trying testdrive? [01:07] RoAkSoAx: No I was looking at the changelog. I won't have time to try testdrive for a while.. the work is piling up [01:08] RoAkSoAx: http://wiki.debian.org/Python/Packaging is all I have.. a newbie guide which suggests only dh_python2 [01:08] k thanks for the link [01:08] RoAkSoAx: I'm quoting the mailing list for debian python modules team that they're working to replace all pysupport/pycentral w/ dh_python2 .. it seems they haven't quite made it canon law yet [01:09] SpamapS: and I personally use TestDrive for everything that involves downloading ISO's or quick tests as it is quick and simple [01:09] SpamapS: indeed [01:09] anyways, I gotta go check the mail [01:11] RoAkSoAx: cool. :) I know it will be useful for some things I need to do soon. :) [01:25] hey, how can I get into a root shell automatically on login? I tried putting sudo -s in my .bashrc but that made it impossible to exit [01:30] SoulPropagation1: uh, don't do that [01:31] SpamapS: why? I only log in for administrative stuff [01:32] yes, but linux has exploits from time to time [01:32] you don't care about security? ;) [01:32] No, not really [01:33] why not? [01:33] SoulPropagation1: you can do administrative stuff without root [01:33] I don't feel like spamming sudo [01:33] eh [01:33] because there's nothing on the box worth securing [01:33] sudo lovesto be spammed [01:33] SoulPropagation1: there's the other boxes it can connect to. ;) [01:34] nothing of value on there either [01:34] sudo is like a fat hawaiian surfer.. eats spam all day! [01:34] SoulPropagation1: so its not on the internet? [01:34] SoulPropagation1: well, hackers can still exploit your server to use it as a spamserver, botnet or even ddos server [01:35] SoulPropagation1: sudo's whole purpose of existence is to be spammed ;) [01:35] and if thats at home, your isp might disconnect you :p [01:35] right and if they could access it I don't think it would be that hard to forge my password twice [01:37] well, like i said, there are exploits too [01:40] yeah I'm pretty sure there are easier better targets [01:40] true [01:43] I mean.. I understand that most servers are relatively juicy targets, what with all sorts of backups on them and stuff and it's probably a good idea that you promote that level of security [01:45] but I don't exactly have much in the liquid assets department, nor do I have anything particularly useful worth taking from me. SO [01:45] can someone please tell me how to automatically go into interactive sudo when I log in without breaking bash? [01:46] sudo su - [01:46] in .bashrc? [01:48] SoulPropagation1: the most valuable asset of your server would be to install a spam mail server or a proxy on it behind your back, not whatever data is on it... [01:50] JanC: I'm sorry but that's just not a big concern of mine. if I notice any devious shenanigans then I'll reconsider. [01:50] juicy targets are easy targets [01:50] and doing something like that would make you an easy and therefore juicy target [01:51] SoulPropagation1: if you don't care, others will care (blacklist your and ask your ISP to disconnect you) [01:51] yep [01:51] Some spammers are smart too. You wouldn't even necessarily notice the missing bandwidth. [01:52] right, you'll only notice after being disconnected ;) [01:52] Yup, seen it happen before. [01:52] at which point fixing things gets difficult :P [01:52] heh indeed. [01:53] so you're saying that right now I can log into any server I want without needing to know a username or password? [01:53] if you knew a new 0day exploit [01:53] and they happen [01:54] see the not-so-random SSH key fiasco a few years back [01:54] May I pm you SoulPropagation1? [01:54] nandemonai: sure [02:14] nandemonai has teh hax D: [02:14] huh? [02:15] "so you're saying that right now I can log into any server I want without needing to know a username or password?" where you say a couple of lines further "may i pm you" xD [02:15] lol [02:15] Not quite. [02:15] i am jacks twicthing elbow [02:16] i know, it was a joke [02:16] i'm so funny hurrr [02:16] I'll say. [02:17] anyways, sorry =) [02:18] Na buddy all good. [02:18] k :p [02:41] humm - does anyone have a maverick system they can test something for me on (what does "invoke-rc.d plymouth status" output?) .. it *should* give output but i'm not getting any :/ === Kiall is now known as Kiall|AFK === Kiall|AFK is now known as Kiall [03:03] i just noticed [03:03] that my ubuntu-server is hacked [03:03] via ipv6 [03:03] i found this [03:03] root@Maverick:~# ls [03:03] f replay_arp-0119-010828.cap tmp [03:03] is it possible to be hacked ? [03:07] how did you find it ? [03:08] did this "server" have a complete hard drive format before being installed with maverick ? [03:11] yes [03:11] i typed w [03:11] so my username donvito [03:11] was loged in from ipv6 ip that i dont own [03:11] i saw in bash history the commands are wroted [03:11] i found sudo su - [03:11] so than i foudn this ls [03:11] f replay_arp-0119-010828.cap tmp [03:15] isn't that an ettercap capture file ? [03:15] or an aircrack capture [03:17] does your server have a wifi interface ? [03:20] yes [03:21] yes, google for 'ARP Request Replay Attack' [03:21] yeah its an aircrack replay capture [03:21] check if you have aircrack installed [03:21] and change the root password, *now* [03:23] all ports [03:23] are closed now [03:23] box is under NAT [03:29] donvito2: you should seriously consider re-installing [03:30] yep [03:30] doing that now [03:42] and install that thing that auto creates iptable drop rules on ip addresses which are the source of failed X number of logins within a certain time period [04:06] RoAkSoAx: I can't add any data there without sending an e-mail to everyone who's subscribed to these builds, sorry === _TechAway_ is now known as _Techie_ [04:25] SpamapS: around? [04:36] New bug: #716248 in sheepdog (universe) "lsb functions not being source in init script" [Undecided,New] https://launchpad.net/bugs/716248 [04:44] how can i know what root typed [04:44] what actually user loged as root did to my server [04:44] what kind of commands etc [04:46] donvito2: history [04:46] donvito2: but there are ways to clear that as well [04:48] just found [04:48] what he did [04:50] what for is command less [04:51] usually less is pipped to file so you scroll the file in the terminal example cat /var/log/auth.log | less [04:53] * twb hits thesheff17 with the UUOC bat [04:53] twb: what does that mean? [04:53] how can i get complete log [04:53] http://en.wikipedia.org/wiki/Useless_use_of_cat [04:53] what he did [04:54] i get only screen log [04:54] nothing more [04:54] i need more [04:57] twb: thx...I have used that command for years: less < /var/log/auth.log :) [04:58] donvito2: all I know is history...and what do you mean screen log? [04:58] well when i type history [04:58] i get just some commands [04:59] usually it is limited to 1000 commands...maybe he executed more commands under a different user? [04:59] thesheff17: UUO < [04:59] thesheff17: less can open a file [04:59] It can even open >1 file, and use :n and :p to move between them [04:59] twb: lol didn't know that either [05:35] i'm having problems getting nfs-kernel-server to start : http://pastebin.com/ZsLdVg4P [05:36] gah nvm === airtonix_ is now known as airtonix [05:38] actually... it still doesn't want to start, citing : Cannot register service: RPC: Unable to receive; errno = Connection refused [05:50] hi i'm trying to set up a samba share on my ubuntu server [05:51] does anybody have a recommended guide they can point me to? i've seen a few posts on the forums but they can't seem to agree on what parameters to set [05:51] Ender: the Ubuntu Server Guide covers it [05:52] kk [05:57] awesome! nfs-kernel-server won't start [06:01] airtonix: is it a VM or container? [06:01] twb: this is not a virtual-machine [06:01] Dunno then [06:03] http://pastebin.com/LLVZ6GsG [06:03] airtonix: Does your kernel have the nfsd.ko module available to it? "Won't start" is a bit generic... check logs for details? [06:03] jmarsden: i did : sudo apt-get install nfs-kernel-server [06:04] i expect a working nfs-server [06:04] airtonix: That does not answer the questions I asked, though. [06:04] A novice was trying to fix a broken Lisp machine by turning the power off and on. Knight, seeing what the student was doing, spoke sternly: “You cannot fix a machine by just power-cycling it with no understanding of what is going wrong.” Knight turned the machine off and on. The machine worked. [06:05] jmarsden: because i don't know how to work out if it does im going to say "don't know" [06:06] jmarsden: http://pastebin.com/xkJAqT9U [06:08] He went away? [06:08] airtonix?? [06:12] ok looks like something else is running on port 2049 [06:14] airtonix: sudo netstat -ntlp |grep :2049 # might tell you what that something is [06:14] jmarsden: rpcinfo -p [06:14] Or is that just for clients? [06:14] Obviously he'll also need an entry in exports(5) [06:15] Not sure... but yes, i was wondering if the issue he has is lack of RPC myself. [06:15] nfsd[14745]: nfssvc: Address already in use [06:15] nfsd[1707]: nfssvc: Setting version failed: errno 16 (Device or resource busy) [06:15] * airtonix rages [06:15] Ah, so you are running the userland nfs server and now trying to install the kernel one at the same time? This may not be wise. [06:16] Hear, hear [06:18] would i be better off, performance wise, running a virus-scanned nat through a VM on a windows vista computer with a core 2 duo processor or directly on a p4 system? [06:19] just anecdotally based on your experience or guesstimates is fine [06:20] i'm sorry, i meant virus-scanned NAS, not NAT [06:21] Ender: I wouldn't run VMs on Vista to start with :) Having said that: If you expect the bottleneck to be CPU, and the VM has enough RAM, then the modern dualcore would probably win. But if the bottleneck for your scanning is disk I/O... it'd be more even, I'd guess. Can you install on both and then do some performance measurements? :) [06:23] yes actually i can [06:23] and i will [06:23] eventually [06:23] haha [06:23] but right now i'm still learning the background on the task i have at hand [06:23] started with the ubuntu server samba guide, but i didn't understand some stuff - so now i'm all over wikipedia, learning about ldap [06:32] does anybody in here have any experience with or knowledge of openERP [06:39] jmarsden: ok assuming a fresh start, i removed all traces of nfs from my server that i was able to with apt-get === airtonix_ is now known as airtonix [06:45] if i'm preparing to create a samba nas on my school's network so that my co-workers can have a local networked storage location, i should NOT configure Samba to be a WinNT Domain Controller right? [06:53] Ender: Correct. Also, for the sake of your own reputation, test SAMBA in your home or on a test LAN first, and get familiar with it, before deployment in a production setting such as a school. [06:54] yeah that's what i'm trying to do [06:54] but naturally the brilliant researchers in my lab bought a 215k piece of equipment to analyze their data without thinking about how they're going to transfer files to their workstations [06:55] the euqipment is up and running but they all have to crowd around the damn thing like kids in a lunch line to use it for analysis [06:55] so i kinda need this running asap [06:55] but your point is very, very well taken. and i'm currently working on it at home and i'm using a vm at work to test. [06:57] The reason i ask about domain controllers is that the ubuntu server guide assumes you want to set it up as a domain controller and dives into setting up LDAP [06:57] since i'm doing this test on my home server, is it a good idea to go ahead and do that just so i know that procedure too [06:57] ? [06:58] Ender: if you've not played with ldap before then you're in for a learning curve [06:58] Ender: as a start, google for ubuntu bloke ldap samba [06:59] ok i'm on that. one of the pre-requisites is "An NFS server exporting the users' home direcories" [07:00] I don't think i have that. [07:00] i think the ubuntu bloke blog has a post on that too [07:00] but hey, i wouldn't know since i can't get nfs server to even run [07:01] lolll [07:01] blind leading blind [07:01] yeah so awesome hey [07:01] after removing nfs-kernel-server and restarting, there is still nfs-kernel-server files all over the place [07:01] so in order to set up a windows-readable network share i have to run an nfs server, ldap server, and samba server? [07:02] no [07:02] you just need a samba server [07:02] you only need ldap if you want centralised authentication [07:02] well i need the share to only be accessible to poeple in my lab [07:02] and you only need nfs if you want a network file share system that has less network traffic overhead than samba [07:03] but all the workstations are independent, they don't authenticate over the network at all [07:03] you only need samba then [07:03] why does the ubuntu server guide assume you need ldap? [07:04] it says ldap is a means of managing users; without it you need custom scripts or some other method fo rmanaging users. [07:04] so if not ldap, what then? [07:04] no [07:04] what are you reading by the way ? [07:04] https://help.ubuntu.com/10.10/serverguide/C/samba-ldap.html [07:05] https://help.ubuntu.com/10.10/serverguide/C/windows-networking.html [07:05] Ender: So you are reading a guide section that starts out "This section covers configuring Samba to use LDAP for user, group, and machine account information and authentication." -- and now asking why it needs LDAP ??? :) [07:06] Ender: https://help.ubuntu.com/10.10/serverguide/C/samba-fileserver.html [07:06] * Ender feels like an idiot [07:06] :) [07:06] reading slowly is required [07:07] well at least i learned a lot about ldap from wikipedia (: [07:07] Ender: if you do bother to do something with ldap, i recommend this first : http://tuxnetworks.blogspot.com/2010/07/howto-samba-ldap-on-1004-lucid-short.html [07:07] Ender: If possible, read all of Chapter 17 of the Server Guide, and then use the parts that are actually relevant to what you are trying to do. === pan1nx is now known as info [07:08] jmarsden: ok apparently removing nfs-kernel-server doesn't actually remove it === info is now known as pan1nx [07:09] airtonix: You can try sudo apt-get purge nfs-kernel-server [07:10] airtonix: Or you can rmmod the loaded kernel module, if that is your issue? [07:12] sudo /etc/init.d/samba stop is not working. how do i stop samba so i can alter the smb.conf file [07:12] oh nvm i figured it out [07:12] Ender: sudo service smbd stop [07:12] had to sotp the smbd instead [07:12] yeh thx [07:13] oh, service? [07:13] i just used sudo stop smbd [07:13] its a shortcut [07:13] doesn't work with all services [07:13] but sudo service stop will work with all services? [07:13] pretty much [07:13] Rather, service is a backwards compatbility layer for RHEL refugees [07:14] so the real way uto do it is use /etc/init.d/ stop [07:14] "real" * [07:14] ? [07:14] the no-i'm-not-a-RHEL-refugee way [07:15] sudo apt-get remove nfs-kernel-server && reboot, (some time later) , sudo apt-get update && sudo apt-get upgrade && sudo reboot, (some time later), sudo apt-get install nfs-kernel-server : http://pastebin.com/4MkrUfEd [07:15] yay for fail [07:15] :( [07:16] did you try the purge [07:16] yes [07:16] did you try to remove nfs-common [07:17] purge* nfs common [07:17] yes [07:17] try to just install nfs-common [07:18] how do you trigger the nfs-common configuration === oubiwann_ is now known as oubiwann [07:18] "dpkg --reconfigure nfs-common"?? [07:19] ok nfs-common isntalled and no errors yet [07:19] http://pastebin.com/xJTRzJh1 [07:20] kayyyyyy [07:20] and nfs-server??! [07:20] what's the vi command to start editing at the end of the current line [07:21] a(ppend)? [07:21] no dice, that's to add at the current cursor location [07:22] how's "a" different from "i" [07:22] i prefer pico or nano [07:22] less fluffing around [07:22] then you actually use the home and end keys for what they were intended for [07:22] my bad...not really comfortable with vi [07:22] yeah but all the cool kids use vi [07:22] lol [07:22] only the ones that like to cut themselves [07:22] =) [07:23] oh, it's A not a [07:23] on the right track to xampart [07:23] thx [07:23] \o [07:23] tho* [07:26] so installing nfs-kernel-server on ubuntu lucid is impossible [07:26] that makes me so happy [07:26] lol [07:26] what happened? it was fine after nfs-common wasn't it? [07:27] no [07:27] i still have the problem with rpc [07:28] impossible? [07:28] apparently [07:28] !impossible [07:28] what is that? [07:28] Sorry, I don't know anything about impossible [07:28] it's where it can't be done [07:28] * Ender experiences a segfault [07:28] does not compute [07:28] nah man it can be done [07:28] maybe not tonight [07:29] but it can and will be done [07:29] airtonix_: (a) What makes you think you need the nfs-kernel-server in the first place? What is so wrong about using a userspace nfsd ? [07:30] pardon ? [07:30] Either way should work fine... both at once probably won't. [07:30] i'm pretty sure i haven't install any kind of nfs prior to this [07:30] so why should somethin you call "user space nfs" be installed ? [07:31] I've set up NFS on Ubuntu 10.04 before, it worked then... but you are claiming it is "impossible"? [07:31] apparently it is [07:31] http://dpaste.com/401214/ [07:31] refuses to start [07:31] or rather refuses to install [07:31] jmarsden: the main thing wrong about it, is the speed [07:31] OK, I'll create a Ubuntu server VM and install it... just to check... [07:31] FWIW nfs-kernel-server works on lucid just fine [07:31] Also hardy. [07:32] That's what I thought too. [07:32] But if airtonix_ says it is impossible for him, he can use the userspace one... he call. [07:32] s/he/his/ [07:32] well why won't it install the lucid machine i have here ? [07:33] ok so you think i have the user space nfsd installed. how do i remove it [07:33] I'm not sure what you have and have not got installed, at this point :) [07:33] Do you have a working portmapper? [07:33] it restarts with out errors [07:35] airtonix_: if you haven't actually used this host for anything yet, it might be easiest to just reinstall from scratch, rather than us trying to work out how you screwed it [07:35] is a mask like some kind of permissions facade or is it just a term to descibe the process of assigning permissions? [07:35] twb: not an option [07:35] twb: and btw i didn't "screw it" [07:35] i just tried to install nfs-kernel-server [07:36] airtonix_: well it's screwed, and *I* didn't do it [07:36] so then apparently nfs-kernel-server is broken on lucid at the moment [07:36] Ender: in what context? [07:36] airtonix_: plonk. [07:37] smb.conf's [share] section parameters [07:37] i fear to install anything else from ubuntu now [07:37] Ender: do you mean a umask? [07:37] airtonix_: Failure to take responsiblity for your own actions => unlikely to get help here. [07:37] the serverguide says set the "create mask" parameter to 0755 [07:37] jmarsden: ? i admit i ran " sudo apt-get install nfs-kernel-server" [07:38] Ender: OK, that's not a umask. [07:38] Ender: what that's saying is, if you don't ask for specific permissions, by default they will be 0755, i.w. rwxr-xr-x. [07:39] A umask is basically the same except the bits are reversed, e.g. in your case they would be 022. [07:39] twb, right, so all files moved onto the smb share will be rwx by owner, rx by group, rx by everybody [07:39] Ender: unless the thing creating them further restricts it, yes [07:39] twb, where is a umask useful [07:39] Ender: umasks are how unix does the same thing [07:39] Ender: e.g. if you run "umask" in your shell, it'll tell you that shell's current umask [07:40] Ender: it defines the (maximum) default permissions of newly created files [07:40] twb, files created by the current user, i presume [07:40] Ender: in that shell, yes [07:41] umask is process-specific, and inherited from the parent process. Typically it's set from some default when you initially log in, e.g. in /etc/profile or ~/.bash_profile [07:41] why's the umask 0022 in a root-owned direcory like /etc the same as in my home directory [07:42] Ender: er, umask is PER PROCESS, not per directory. [07:43] process specific. [07:43] so if i run umask and it tells me 0022, that means what [07:44] "any process you run here will operate with 0022 permissions"? [07:44] ermmmm lemme do some reading on t his first [07:44] Any new file you create here will not have group write or other write permission bits set [07:44] Ender: I was about to suggest that [07:44] Ender: http://en.wikipedia.org/wiki/Umask [07:45] already there (: [07:45] Ender: if you don't already understand how binary and octal work, go learn those first. [07:45] well i generally understand what binary is [07:50] oh isee [07:51] yes actually i did know that, but i've never used binary to specify file permissions [07:51] airtonix_: http://paste.ubuntu.com/565266/ [07:52] Ugh [07:52] "# do not edit this file, it will be overwritten on update" [07:52] So how am I supposed to disable you, you darn udev persistent-net-generator.rules? [07:52] * twb tries a dpkg-divert [07:52] jmarsden: thats good [07:54] airtonix_: Exactly. So... you messed with your server such that it breaks, and then refused to take responsibility for doing so. Remember: " so then apparently nfs-kernel-server is broken on lucid at the moment" ? [07:54] jmarsden: um no? [07:54] jmarsden: i just tried to install an nfs server. [07:56] is it better to use upstart than init.d to manage services that have been converted to upstart jobs? [07:56] it keeps complaining when i use /etc/init.d to start/stop smbd [07:57] Hi [07:58] Uh.. not quite sure whats ahppened but my proxy "squid" service seems to have stopped and I cant figure out why or.. how to restart it [07:58] Only thing in the logs is this "Feb 10 07:50:40 beluga init: squid main process (1401) terminated with status 1" [08:01] have you tried /etc/init.d/squid restart [08:01] No such command, nor does service squid restart work (so such service) [08:02] did you recently install anything that might have broken it [08:02] or uninstalled something [08:02] Nothing for several days [08:02] Goddammit, why is my interfaces(5) file filled with mtab entries [08:02] have you restarted your system to see if that fixes it? [08:02] I was using it right up to the moment it crashed it .. just exited [08:03] twb, it's probably because your squid proxy service is broken like Slyboots [08:03] well the service has been running for two months without issue [08:03] Slyboots: What does dpkg -S /etc/init.d/squid output? [08:04] dpkg: /etc/init.d/squid not found. [08:04] Did you install squid from the Ubuntu package? or some other way? [08:04] Ender: no it's because e2fsck is playing silly buggers [08:04] Ahh.. think I installed it from the ubuntu repo [08:05] twb fsck added lines to your interfaces file?!?! [08:05] baaaaaad [08:05] Slyboots: OK, what does dpkg -l squid output? [08:05] why isn't my samba share showing uppppppp [08:06] Ender: no it shuffled inodes around [08:06] ii squid 2.7.STABLE9-2ubu Internet object cache (WWW proxy cache) [08:07] Slyboots: So... you installed it from a package, but something deleted the /etc/init.d/squid file. [08:07] Do you have any idea what might have done that? [08:07] Not in the slightest [08:08] maybe it was just an fsck-shuffled inode [08:08] jmarsden: no, squid is an upstart job in 10.04 [08:08] lrwxrwxrwx 1 root root 21 2011-01-26 14:15 /etc/init.d/squid -> /lib/init/upstart-job [08:08] OK, so that backwards-compatibility link should still be there... [08:08] twb: I just installed it in a VM and dpkg -S /etc/init.d/squid works here ... [08:08] yeah, plus he already tried service squid restart [08:09] which is calling it as an upstart job right [08:09] Throw down debsums or cruft and work out what else is missing [08:09] Ender: yes [08:09] well if it was installed from a package, why not just uninstall the package and install it from the repos? [08:09] won't that re-generate the /init.d file? [08:10] Ender: And we will have no idea what deleted bits of it, so most likely whatever did it will do it again... [08:10] ah, yes, well...there's that, huh [08:11] Mmm [08:12] Slyboots: sudo apt-get install debsums && sudo debsums -g && sudo debsums squid # I think? [08:12] Whats that do? [08:13] Slyboots: We need to find out what other bits of the squid package have been deleted or corrupted, so we install the debsums package, generate sums for packages that have none already, and then test the ones for squid against the installed squid files. [08:14] OKau [08:14] Slyboots: But you can do just sudo apt-get install debsums and then read the man page, since you should not trust random people you meet in IRC, including me :) [08:16] kkjj [08:16] Ender: IRC is not vi :) [08:16] rather, my samba share isn't working! why? [08:16] lol [08:16] seriously [08:17] i should be able to browse the samba share form a linux desktop right [08:17] or, for that matter, from a windows VM in side my ubuntuy desktop right [08:17] Ender: test it with smbclient on the server, then smbclient on another host. Only *then* try a GUI client [08:17] jmarsden: I would be doing "debsums -S" or so, to ask it about ALL packages [08:18] Ender: testparm /etc/samba/smb.conf might also be wise [08:18] Okay; ran the program and created the checkcumns [08:18] But says "No checksums for squid" [08:18] jmarsden: btw, :) i didn't "screw anything up" nfs doesn't like ipv6 [08:19] jmarsden: but hey thanks for your help [08:19] you're using ipv6 on this server? [08:19] no [08:19] it's enabled by default [08:19] so you disabled it and now it's working? [08:19] apparently [08:20] wow cool [08:20] how'd you disable it [08:20] http://www.webupd8.org/2010/05/how-to-disable-ipv6-in-ubuntu-1004.html [08:20] but obviously i'll need ipv6 at some point in the future so, not a foolproof plan [08:21] maybe i can disable it for the WAN interface only [08:21] how do i run a samba client on the server [08:22] Yea for some reason the system does not have the md5 sums for squid [08:22] Ender: smbclient [08:22] Ender: or did you mean something else ? [08:22] no that's right, but i need additional arguments [08:25] how do i use smbclient [08:26] Slyboots: Try sudo debsums -p /var/cache/apt/archives/ -g squid [08:26] Ender: Did you read its man page? [08:26] No md5 sum or deb avaibable heh [08:26] the man page is terrible for smbclient [08:27] well i'm sure it's fine but it's late, i'm tired, and it's very voluminous [08:27] but i found a website with the correct information [08:27] and yes, i can see share on smbclient run from the server itself [08:27] but ? [08:28] but i can't access it form other systems on the network [08:28] Slyboots: sudo apt-get -d squid --reinstall [08:28] Slyboots: And then sudo debsums -p /var/cache/apt/archives/ -g squid [08:29] wtf i can view the share using smbclient from my ubuntu box but i can't open it in the gui [08:29] Ender: There is a GUI on your server??? How did that get there? [08:30] i'm getting tired and being unclear. i apologize. [08:30] i'm running the gui on my desktop [08:30] jmarsden: Okay; it reports.. All OK [08:30] OK. Does the server run any kind of firewall/packet filter? ufw maybe? [08:30] Ender: ^^ [08:31] jmarsden: Aye, I use ufw [08:31] i didn't install it [08:31] but it appears to be installed [08:31] Ender: So, did you makethe relevant holes in it for SMb traffic ? [08:32] i just ran ufw allow samba [08:32] and it worked [08:32] is that all i have to do?!? [08:32] Try accessing your shares from a workstation to answer that question :) [08:32] well i don't think it was a permissions issue, actually [08:32] Ender: 19:17 Ender: test it with smbclient on the server, then smbclient on another host. Only *then* try a GUI client [08:33] twb, i already did that, and smbclient reported the share when run from the server [08:33] in fact, smbclient reports the share when run from my desktop [08:33] That should've tipped you off that the firewall was the problem [08:33] AND i can SEE the share in firefox from my desktop [08:34] wait...the fact that i could see it with smbclient form my desktop indicates it's a firewall problem? [08:34] Note that ufw is *installed* by default, but not *enabled* by default [08:34] oic, then it's disabled [08:34] i haven't enabled it, anyway [08:34] Ender: sorry, I thought "19:32 is that all i have to do?!?" was rhetorical [08:34] Ender: sudo ufw status [08:35] Status: inactive [08:35] so it's not a firewall problem [08:35] OK, it is not that. So... what is the issue? smbclient works from the workstation, you said? [08:36] well i type smbclient -L and it kicks back a line naming my share and giving the description i loaded into smb.conf [08:36] i haven't used smbclient to touch a file or anything [08:36] Ender: OK, so now test that smbclient can actually use the share to read and write files [08:36] well yeah but i don't know how smbclient works yet [08:37] lemme fidn out [08:38] * jmarsden retreats to bed... goodnight all. [08:41] gnight jmarsden [08:44] ok [08:45] so i have now been able to connect to the share on my server from my desktop and place a file from my desktop to my server using smbclient [08:45] why can't i connect using the Network location on gnome? [08:57] Ender: I'd guess that Windows/SAMBA network browsing is not working on your network. Are both workstation and server in the same workgroup ? [08:57] (But I really am going to bed!) === FkCek is now known as FkCek|a [08:59] no, but i told the share to be browseable [08:59] shouldn't i be able to browse shares on other workgroups? [08:59] besides, does ubuntu even have a workgroup? [09:00] * jmarsden thinks you need to read up on widnows network browsing now, and to read /etc/samba/smb.conf to answer the last question. [09:00] lol [09:00] i'm done reading for tonight [09:00] goodnight! [09:00] Goodnight. [09:00] * Ender really IS going to bed, unlike jmarsden [09:00] (: [09:00] :) [09:08] <[diablo]> morning all [09:09] <[diablo]> I am looking to setup an installation mech. similar to Kickstart ... I have done those before on RHEL... whats the mech. for Ubuntu Server please? [09:11] https://help.ubuntu.com/community/KickstartCompatibility if you must. https://help.ubuntu.com/10.04/installation-guide/i386/appendix-preseed.html for the "Debian way", preseeding [09:11] +1 for preseeding [09:11] I never understood why kickstart was ported to d-i [09:12] <[diablo]> mmmm [09:12] <[diablo]> thanks guys [09:12] You're welcome. [09:16] <[diablo]> is preseed cuter than kickstart? [09:16] <[diablo]> I remember KS was pretty nice [09:17] In the Debian/Ubuntu implementation, preseed is more functional than kickstart. Cuteness is not relevant for server admins anyway :) [09:17] <[diablo]> hahaha [09:18] <[diablo]> will read thru the doc you linked me, and give it a whirl [09:19] [diablo]: preseed is more elegant [09:19] <[diablo]> cute [09:19] [diablo]: in that it doesn't conflate multiple DSLs in a single translation unit [09:19] <[diablo]> sorry, mean cool :P [09:22] Anybody playing with this new elliptical curve stuff in openssh? [09:22] (Just curious, rather than wanting technical support.) [09:26] <[diablo]> twb, wot is it? [09:27] New bug: #716310 in openssh (main) "ssh connections disconnects automatically" [Undecided,New] https://launchpad.net/bugs/716310 [09:28] [diablo]: a new key type, like DSA and RSA, but "more betterer" [09:28] <[diablo]> oh [09:28] http://tools.ietf.org/html/rfc5656 [09:28] <[diablo]> interesting... [09:29] <[diablo]> did not know of that [09:30] You'll need natty; it's new as at 5.7 [09:36] Huh. Did you know that if you use the same passphrase to symmetrically encrypt your private keys, ssh-add will reuse it to decrypt all of the keys you pass it? [09:36] twb: I was curious and read up a bit on it, The SHA2 family consists of four variants -- SHA-224, SHA-256, SHA-384, and SHA-521 -- named after their digest lengths [09:36] " [09:37] don't they mean sha-512? :) [09:37] Er, until it changes. [09:37] I have two passphrases, but due to the ordering, it prompted me four times: http://paste.debian.net/107132/ === _Techie_ is now known as _TechAway_ [10:34] hello there [10:34] what this command do ? [10:34] svn co file:///svnroot/master/ \ ~/repositories/ [10:37] creates an SVN checkout of /svnroot/master/ to ~/repositories/ [10:37] file:// is a URI schema for local files, just remove the "file://" to get the path [10:43] Well.. [10:43] It depends. [10:43] If you write it on one line, it'll probably fail. [10:43] I don't think svn co can take three arguments. [10:44] Oh, it does work. [10:45] Well, it doesn't work, but svn co does accept three arguments. It wants the second one to be a url. " " isn't. [10:46] i can't cd to /svnroot tsimpson [10:47] rtyuio: I don't know where you got the command from, so I can't comment on how to fix it [10:48] i ask the question in general [10:48] soren: I assumed that the paste was 2 lines concatenated to one for the paste [10:48] how work a command like that ? [10:49] you are giving the arguments "co file:///svnroot/master/ \ ~/repositories/" to the svn command [10:49] tsimpson: I've been here for too long to make assumptions like that :) [10:50] rtyuio: That command will checkout from svn at /svnroot/master to ~/repositories/ (if typed correctly). [10:55] checkout what ? [11:07] rtyuio: see http://en.wikipedia.org/wiki/Apache_Subversion for a description of what svn is [12:55] has anybody set up openldap on ubuntu? I'm at a step that's hanging and I'm not sure why. === SquishyNotHere is now known as squishy [13:00] Anything change in 10.04 that'd make the network slower for dual nic setup? [13:10] any reason at all the my network would be slower on 10.04? any setting's I can tweak to make it go faster? === lionel__ is now known as lionel [13:37] Hi all. I'm trying to connect a ubuntu server 10.10 to a Iomega Storecenter NAS with NFS enabled. I was able to connect to the NAS a couple days ago, but after a reboot I am having issues. I'm not sure which version of NFS the storcenter uses, but everything I try gives me an error. NFSv3 says the requested nfs version is not supported... NFSv4 says Protocol Not Supported. Any ideas? [13:41] takamarou: rpcinfo -p? [13:43] any certain line you are looking for? Or should I pastie it? [13:43] The latter [13:44] I'm just triaging man [13:45] http://pastie.org/1548915 [14:07] New bug: #298823 in exim4 (main) "Exim mail failure messages go to local address, even if no local delivery is available" [Undecided,Incomplete] https://launchpad.net/bugs/298823 [14:09] morning all [14:11] takamarou: what is in /etc/default/portmap ? [14:11] airtonix, looks pretty empty [14:12] New bug: #702714 in exim4 (main) "package exim4-base 4.71-3ubuntu1 failed to install/upgrade: subprocess new post-removal script returned error exit status 1" [Undecided,Incomplete] https://launchpad.net/bugs/702714 [14:12] airtonix, OPTIONS="" and a bunch of comments [14:12] takamarou: wait is your ubuntu machine the nfs-kernel-server or the Iomega machine ? [14:13] ubuntu machine is the client connecting to the NFS. The Iomega Storcenter is the one the NFS Server [14:13] that changes things [14:14] because now, you have to seek support from iomega [14:14] damnit [14:15] It's not an error on the Iomega.. I think. I can mount the NFS just fine from my local ubuntu box [14:15] which is version ? [14:15] 10.04 or 10.10 ? [14:15] 10.04 [14:15] ;) [14:16] what reason is there to have 10.10 on the server ? [14:16] +6 [14:16] None.. Actually, I intended to have 10.04 on it, but hit the wrong download. [14:16] But now it's in production.. can't take it down at this point [14:17] airtonix: none at all unless you hit bugs that aren't fixed in 10.04, or if you need newer hardware support or features [14:17] i'm out of ideas [14:17] OK. Thanks for your help :) [14:39] Daviey: this might sound like an odd question but would you expect the 127.0.0.1 address on the loopback device to be disabled in ipv6 only installations? [14:40] JamesPage, i know what bug you are talking about... [14:40] :) [14:40] JamesPage, I only have mixed ipv4 and ipv6 machines, so i need to disable ipv4 and test that bug [14:41] With ipv4 disabled, i wouldn't expect lo to have 127.0.0.1 [14:41] ...and ubuntu (ntp) still work [14:41] Daviey: doing some testing at the moment - the only way I can find to get rid of the ipv4 address is to manually remove it (ip addr del) [14:41] is there a way of globally disabling ipv4 [14:41] yeah [14:42] ::1 <-- ipv6 localhost [14:42] (shortage notation) [14:48] any hints on my my 10.04 install's terminal is very slow when used in kvm? [14:56] hey spdw [14:56] howdy zul [14:56] whats the error again? [14:56] sendmail gives me this when an email does not go out: Feb 10 14:12:28 arkli sm-mta[13934]: p1AECCKH013934: localhost [127.0.0.1] did not issue MAIL/EXPN/VRFY/ETRN during connection to MTA-v4 [14:57] Daviey: seen that before: ^^^ [14:57] it is intermittent, the same message a few minutes later to the same address went through fine [14:57] interesting.. [14:57] spdw: i dunno use postfix ;) [14:58] spdw: actually you can grep p1AECCKH013934 in the same log file and see what it is doing [14:58] * RoyK thought sendmail went out of style some time before y2k [14:58] zul: I know how you lean. Sendmail came for free when rackspace commissioned the server. At least that's what I'm being told. [14:58] spdw, Are you using auth to the relay server? [14:58] Daviey: is there an easy way to find out? I did not setup sendmail [14:59] I don't like sendmail [14:59] spdw, If you have root, i would consider switching to ubuntu favoured mta which is postfix, or debian's which is exim4 [15:00] Daviey: I definitely do and if that's the easiest fix then I'm good with that. I was hoping this was a small misconfiguration problem but it doesn't appear to be that way. [15:00] There's no such thing with sendmail :) [15:00] spdw, many agree that sendmail is a world of pain [15:00] soren: that has been my experience in the past too. Hence my comment about not liking sendmail [15:00] i still have my bat book but its collecting dust [15:00] spdw, i would try sending mail via telnet to port 25... [15:01] zul: you live in the middle of nowhere, I'm not stopping by to pick that thing up :) [15:01] Daviey: trying now [15:01] spdw, How is the mail getting to sendmail... some app? [15:01] spdw: its light reading [15:01] hah [15:03] Daviey: telnet is fine [15:03] spdw, That error seems to happen if you just 'ehlo server' and 'quit' with no payload. [15:03] Daviey: the email is sent through php [15:03] spdw, i think it's your php app to blame TBH. [15:04] spdw, if you run those two commands quoted, do you see the same error in your mail logs? [15:04] checking === oubiwann is now known as oubiwann_ [15:05] Daviey: yep, you're right [15:05] spdw, So, that indicates sendmail IS working [15:05] but the PHP app is failing to send a payload. [15:05] Daviey: cool, I'll take a look from that side then [15:06] spdw, super [15:06] spdw, I'd love for you to feedback what caused it. [15:07] Daviey: sure. I believe, but will verify, that we're just using the mail() method. [15:08] spdw: ill bring the bat book tomorrow for you [15:15] Daviey, zul: Thanks for the help [15:15] Daviey: np [15:17] er...spdw === guampa is now known as guampafk [15:18] spdw, np [15:24] * RoAkSoAx time for rhcs3.1 testing [15:26] hallyn: do you have a doc or a pointer to setup lxc and libvirt? [15:42] zul: there is http://s3hh.wordpress.com/2010/09/07/easier-creation-of-libvirt-lxc-containers/ if that's what you mean [15:43] zul: but that's all [15:43] zul: did you have q, or are you thinking for release notes? [15:43] hallyn: no i was just looking for a general guide [15:43] hi folks... I have an svn repo on a ubuntu-server... and I need to change the password.. but when I do on the server.. I am having trouble logging again from ssh ... how do I update my key locally ? entering the new password doesn't seem to be enough [15:44] hallyn: but if im using you lxc-guest package do i need to do half of those things in the script there? [15:47] zul: checking the scripts, cause i don't recall offhand what all i had to do [15:49] zul: all right, you're starting with a full uec guest image, right? [15:49] zul: the one thing I expect you to have to do is provide a console service on /dev/pts/0, bc that's where libvirt-lxc offers console [15:49] hallyn: yep unmodified as of right now [15:49] gotcha [15:50] thanks [15:50] ill ask if i have any more questions [15:51] great between kvm and python my CPU is 100% [15:52] RoAkSoAx, that sounds great! [15:52] RoAkSoAx, have you tried starting kvm with --don't-abuse-me? [15:52] zul: ok - yeah all the rest looks ok. You can still use my base.xml as a template if you haven'te already devised your own [15:53] hallyn: thanks [15:56] Daviey: lol... seems virt-manager issue [15:58] RoAkSoAx, ahh [15:58] virt-manager is a troubled soul at times. [16:00] Daviey: indeed [16:01] Is a "complete removal" in synaptic = sudo apt-get remove --purge *package* ?? [16:02] Roasted, try #ubuntu ... i imagine most here don't use synaptic. [16:03] e cineva ;)) [16:03] http://problemepc.team-forum.net/ [16:04] Daviey, good point :P [16:11] Daviey: unfortunately, there have been humungous fixes to bind for DNSSEC support between 9.7.0 and 9.7.2P3 [16:11] Daviey: I don't know if it's sanely possible to pick them all out and try to backport them [16:11] if sudo killall -9 freeradius is the wayt o kill the freeradius process, how can I restart it without rebooting? [16:11] Daviey: and it kind of sucks that Lucid doesn't have working DNSSEC support [16:12] hallyn: have you seen KVm using 100% CPU lately? [16:12] Daviey: maybe we should try and get a one-time SRU of 9.7.2P3 into lucid and maverick [16:13] hallyn: or at least more than 50% [16:14] hallyn: yay it works [16:15] RoAkSoAx: on natty? [16:15] RoAkSoAx: it sounds familiar, can you search lp bugs? i think someone else had that with virtio [16:15] though i thought that was maverick [16:16] hallyn: right now with maverick, I'm launching a VM with testdrive and in somepoints doing IO it goes all the way to 100% CPU usage, and for a while was around 50% [16:16] y/win 26 [16:17] hallyn: launching it with virt-manager, it does stay in 100% even though I close virt-manager, kvm stays in 100% [16:17] lemme check my list [16:22] RoAkSoAx: no, i was thinking https://bugs.launchpad.net/ubuntu/+source/kvm/+bug/703811 but this sounds different [16:22] Launchpad bug 703811 in kvm "kvm guest's cpu usage with virtio storage driver goes up to 100% because of flush process" [High,Incomplete] [16:22] RoAkSoAx: is it doing anything in the meantime? [16:22] i.e is the VM moving along? [16:23] hallyn: yeah. let me get you a quick video [16:23] if sudo killall -9 freeradius is the wayt o kill the freeradius process, how can I restart it without rebooting? === airtonix is now known as jadakren [16:31] hallyn: yeah I think that's the one!! I'm crrently installing a new VM and while installing the base system and things that do disk IO apparently, that's when the load goes up [16:33] RoAkSoAx: maybe kill -HUP would reload its config file? I don't have it installed here, what does the man page say? [16:33] ^Roasted [16:33] jmarsden: Error: "Roasted" is not a valid command. [16:33] Bah. [16:34] I got it [16:34] :) [16:34] facing a new problem now [16:34] but I doubt any of you guys have tinkered with freeradius. It's a specific error to freeradius itself. [16:35] hallyn: did you resolve whatever it was you needed last night? [16:37] SpamapS: i sent the email to upstart-devel instead [16:37] SpamapS: no replies yet, and i've had no brilliant ideas [16:37] SpamapS: so i think i'll just end up having to special-case cgred in /etc/init/libvirt-bin [16:37] hallyn: http://me.roaksoax.com/kvm.ogv (ignore the sound :) ) [16:38] lol, now i gotta go watch :) [16:39] RoAkSoAx: but your fflush-N aren't showing high usage? [16:39] hallyn: nope, just the CPU and the graph in vrit-manager is half load [16:40] hallyn: oh I did see that email [16:41] RoAkSoAx: pls file a new bug, mention it MIGHT be a dupe of that one, and upload your .xml's so i can take a look [16:42] RoAkSoAx: you're sure you're using kvmand not plain qemu right? :) [16:42] hallyn: yeah i;m using KVM lol :) [16:47] New bug: #559307 in exim4 (main) "Local delivery hard link vulnerability" [Low,Fix released] https://launchpad.net/bugs/559307 [16:50] SpamapS: were you drafting a reply? :) [16:51] hallyn: will later for sure. [16:56] Roasted: you do know there is a specific channel for freeradius right? #freeradius [16:56] oh believe me, I know [16:56] go ahead and hang out there and see how many people talk in 72 hours time [16:56] ;) [16:56] Roasted: thats a good sign that maybe its time has passed.. ;) [16:56] I haven't used freeradius in atleast 5 years [16:56] SpamapS, you're telling me. [16:57] but isn't freeradius still more secure than standard WPA2? [16:57] hallyn: what other xml's should I upload besides the VM's one? [16:57] Roasted: there is no wifi security. ;) Use VPN and/or SSL at all times. [16:57] SpamapS, well, you know what I mean... [16:58] Is radius still "more" secure than WPA2? [16:59] they have nothing to do with each other [16:59] I thought they did... [16:59] nope [16:59] wpa2 is a wifi encryption thing [16:59] it can use radius auth, or psk auth [16:59] how does radius auth differ from psk? [16:59] RoAkSoAx: the VM ones, and the storage ones [16:59] psk = personal right? passphrase, etc. [17:00] no [17:00] psk means preshared key, one password for everyone [17:00] radius auth you have one per person [17:00] RoAkSoAx: virsh vol-dumpxml and virsh pool-dumpxml [17:00] and you can limit with certificates too [17:00] well from that standpoint alone I would think radius would be far superior. [17:00] so user must have valid cert to access [17:01] ya, but it has nothing to do with how secure wpa2 is [17:01] so it's not how secure it is, it's the way it implements the authorization? [17:01] ya [17:01] but in a way I would think that would still aide in greater security, even though it's using the same level. [17:01] it just goes about it in a more complicated way, it seems. [17:01] yes, over all it would be more *secure*, but it doesn't make the encryption of the channel any better :) [17:02] well, I work in IT at a school district. One of the students last year got some program to reveal the passphrase we were using for wifi. [17:02] as a result, they would get on our network on their phones, etc. [17:02] I think that's why radius has something to offer us. [17:02] It's not entirely my decision. I'm just the one with the project that was assigned to me. [17:02] if you use cert verification, that would block it [17:03] cert verification with... radius right [17:03] you can only do cert verification with radius [17:03] okay [17:03] I thought so [17:03] just wanted to make sure we were on the same page [17:03] yeah right now I'm trying to generate some certs for my test environment. [17:03] I set that same thing up, years and years ago [17:03] but the guide is telling me to run "make" but that's it, which isn't working for me. [17:03] Is there anything better to use than radius for our scenario, given you now know why we were aiming for radius? [17:07] this make thing is getting old. I have no idea how to work around this. [17:07] their documentation sucks. [17:07] lynxman: ping [17:07] kirkland: pong [17:07] lynxman: hey, i just uploaded a new mcollective fixing the two bugs you filed (thanks for the reminder, zul) [17:07] lynxman: i need you to do one more thing ... [17:08] New bug: #716528 in qemu-kvm (main) "kvm 100% when performing apparently IO operations" [Undecided,New] https://launchpad.net/bugs/716528 [17:08] lynxman: create a .postrm script that removes the user [17:08] kirkland: just updated the bug with the script [17:08] lynxman: rock [17:08] :) [17:08] lynxman: i'll get another upload [17:08] cool \o/ [17:09] lynxman: in the future, if you want credit in the changelog, send a debdiff ;-) [17:09] lynxman: if you don't care, this works too :-) [17:09] roasted, I dunno, I haven't used radius on unix for so long, I have one radius setup I use for everything just about, but it's windows based, cause they are running AD [17:09] kirkland: I'll have it in mind, thanks ^^ [17:09] kirkland: just happy to see the package doing cool stuff :) [17:11] RoAkSoAx: so just to make sure, have you tried just running 'kvm -hda delme.img -cdrom ubuntu-10.04-server.iso -boot d -m 1G" and seeing what it does with your cputime? [17:11] lynxman: mcollective_1.0.0-0ubuntu4 should be ready for your testing shortly [17:12] lynxman: please test package installation and uninistallation for me ;-) [17:12] kirkland: excellent, thank you ;) [17:12] will do so [17:12] Ahh! lol.. Figured ouy why my Squid Proxy kept crashing.. [17:12] out of disk-space :P [17:13] * SpamapS hands Slyboots the Sombrero of Glory [17:15] * Slyboots glows with mighty uh.. mexican power? [17:16] Ole! [17:16] * SpamapS forgets how to do accents [17:16] patdk-wk, yeah, we have MS Radius set up now, but there's a bug with MS Radius that backfires with certain wifi cards. It just so HAPPENS that it's the card we have in 540 something laptops... which drops authentication after 30 days and doesn't regain it unless we hardwire them to re-authenticate. I've heard from numerous people FreeRadius doesn't have this issue, hence our interest. [17:18] hallyn: by running kvm etcetc the load is not so high, so it is something related to libvirt [17:18] or must be [17:19] RoAkSoAx: not necessarily - have you also tried [17:19] kvm -drive file=delme.img,if=virtio,index=0 -cdrom x.iso -boot d ? [17:19] hallyn: im running the same isntalling through TestDrive (which uses the username you are logged in and rtuns a vm with virtio and kvm -etcetc) and the load is not that high [17:20] s/same installing/same ISO installation/ [17:20] s/rtuns/runs [17:20] patdk-wk, I'm assuming that FreeRadius can integrate with MS LDAP... ??? [17:21] hallyn: this is the command used in testdrive: "kvm -m %s -smp %s -cdrom %s -drive file=%s,if=virtio,cache=writeback,index=0,boot=on %s" [17:21] ldap is ldap [17:21] but LDAP is where the users are, and we have nearly 2,000... that's where the user base is stored to authenticate "from".. right? [17:23] hallyn: also note that the high load appears when apparently is doing intesive IO operations [17:24] hallyn: and at the same step of the installation, the load is different between a kvm launched by libvirt than the one launched by testdrive [17:26] RoAkSoAx: pls use 'ps -ef | grep kvm' to get the full commands executed by both [17:26] hallyn: libvirt /usr/bin/kvm -S -M pc-0.13 -enable-kvm -m 512 -smp 2,sockets=2,cores=1,threads=1 -name natty1 -uuid 31631c3c-a914-6365-12b7-9103b4e07a3b -nodefconfig -nodefaults -chardev socket,id=monitor,path=/var/lib/libvirt/qemu/natty1.monitor,server,nowait -mon chardev=monitor,mode=readline -rtc base=utc -no-reboot -boot dc -drive file=/var/lib/libvirt/images/natty1.img,if=none,id=drive-virtio-disk0,boot=on,format=raw -device virtio-blk-pci, [17:27] RoAkSoAx: and paste in the bug. i'll look there in a few mins [17:27] hallyn:yeah that's what I'm doing :) [17:29] RoAkSoAx: thanks :) [17:30] ;) [17:34] i just rebooted and for some reason my root is mounted as a read only file system? [17:36] probably cause it's corrupted [17:37] reboot into recovery mode, and try fsck'ing the fs [17:37] for some reason my grub timeout is zero so i'm booting into a live cd now [17:37] thta works [17:37] using 10.04 by the way [17:38] it seems really hard to change the grub timeout... there is no more menu.lst file, and some script called 00-header or something has timeout stuff all over the place [17:38] not sure what to touch in there [17:38] heh, /etc/default/grub [17:42] RoAkSoAx: /// [17:42] sorry, internet connection fail [17:43] RoAkSoAx: is comment #5 correct? [17:43] RoAkSoAx: bc it shows 'if=none'. it looks garbled [17:45] * hallyn bbl [17:46] hallyn: yeah something's weiord going on here :/ [18:12] panfist, grub has been updated to grub 2 which is different in the way in manages its config [18:21] New bug: #716576 in tftp-hpa (main) "package tftpd-hpa 5.0-11ubuntu2 failed to install/upgrade: ErrorMessage: il pacchetto tftpd-hpa non è pronto per la configurazione impossibile configurarlo (stato corrente "half-installed")" [Undecided,New] https://launchpad.net/bugs/716576 [18:22] we had a little accident at work the other day - a generator was started and pumped 1kV or so into our servers.... some 30% of them died, but all of the supermicro servers survived, some 12 boxes or so. According to the statistics, at least four of them should have died, but they are all running... Bad thing those Sun boxes that cost 4x the price of the supermicros died :P [18:24] :) [18:25] RoyK: testing emergency power supply gone bad? [18:26] * alvin holds 1 minute of silence for those poor Sun boxes [18:26] not even testing - the power was to be shut down for an hour or so [18:27] so we got a generator - we've done it before - but now, during startup, the generator generated a rather ugly surge [18:27] alex_joni: make that a second :P [18:27] or some milliseconds [18:28] * RoyK doesn't like Sun^WOracle hardware [18:28] ya, I don't like *connecting* generators till they are warmed up [18:29] I didn't [18:29] I was going to say: "why not, they're indestructible. But well,...." [18:29] perhaps they should have warmed it up before they fused it? === NG_ is now known as ng_ === _TechAway_ is now known as _Techie_ === hikeonpast___ is now known as hikeonpast [18:55] shouldn't the ups eat the overvoltage? [18:56] ah. timestamp fail [18:59] shauno, depends [19:02] we typically feed both pole & genny into a transfer switch, and then into the ups from there. let the ups handle the phase distribution, and eats the time between the pole going down and the genny coming up [19:03] that way the ups would still do it's job if the genny's dirty [19:06] hmm, could easily be a faulty ups part also [19:06] maybe it detected the large spike, and might of been too large for it to handle [19:06] and that caused it to go into bypass mode? [19:06] heh, electricity is fun, all kinds of unexpected things can happen :) [19:06] that should go to battery [19:07] bypass is typically the ups can't give you juice, so it feeds you the wall and hopes it works [19:09] just curious; I work with/for a particular brand of ups, so I have trouble picturing layouts that aren't what we deploy [19:10] ya :) [19:10] I had a ups on my, just turning itself off cause of the generator [19:11] found the generator was bad [19:33] * patdk-wk wonders if there is a wol issue I missed :) [19:33] seeing thousands of wol requests flooding me from the inet [20:16] New bug: #716654 in postfix (main) "latest ubuntu updates killed mail delivery" [Undecided,New] https://launchpad.net/bugs/716654 === ng_ is now known as NG_ [20:31] New bug: #716659 in multipath-tools (main) "Root filesystem goes "Read only" after installing multipath-tools on Lucid" [Undecided,New] https://launchpad.net/bugs/716659 [20:35] hi there [20:35] i've got samba working properly on my virtual server! yeahhhhh [20:35] now i need to set up some rudimentary security. how? [20:44] for example, i'd like to require all users to have to supply a password to access the sahre [20:44] how do i do that? [20:56] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=612296 [20:56] Ruh roh [20:56] Debian bug 612296 in wnpp "O: bacula -- network backup, recovery and verification - meta-package" [Normal,Open] [20:56] ruh roh [20:56] bug=612296 [20:56] how'd you get that to pop it up [20:57] just the URL? [20:57] the URL [20:57] ic. cool. [20:57] zul: ^^ .. bacula has been orphaned [20:57] why is swat refusing un:root and the root password? [20:58] NDROfTheLine: because it wants to protect you! [20:58] SpamapS: hmmmm... [20:58] NDROfTheLine: you can define a non-root user as an administrator and use that [20:59] SpamapS, well i understand the danger of root logins, i really do, but in order to create a new share and manage my existing shares i need root access don't i [20:59] SpamapS, i've tried logging on with my account and all i can do is view information about my shares. [20:59] zul: does Canonical use it internally? Might be strategic for us to take up maintainership [20:59] SpamapS: already one step ahead of you [20:59] planning open-world domination huh [21:00] SpamapS: im not sure but we should take it over [21:00] redhat cluster debian maintainer is also looking for someone else to maintain it [21:01] why can't i administer my samba shares with my normal login? [21:01] on swat, that is [21:02] SpamapS: just fired off an email to John [21:02] cool [21:02] SpamapS: thanks for pointing that out [21:03] NP.. just by chance saw it w/ the changelog for bug #689327 getting fixed [21:03] Launchpad bug 689327 in dbconfig-common "bacula-dir.conf shows syntax error when mysql dbname is left blank" [Low,Confirmed] https://launchpad.net/bugs/689327 [21:03] asdf [21:04] sorry, i hit the wrong button. on the Ubuntu Community Swat page, it says "Log inas a user with proper priveleges"; how do i ensure my user has proper privelege to administer samba? [21:08] SpamapS: replied to the bug as well [21:16] anybody here know where the ubuntu server guide is [21:16] !serverguide [21:16] The Ubuntu server guide may be found at http://help.ubuntu.com/10.04/serverguide/C/ [21:17] samael6: ^ [21:17] no i just installed it from the ubuntu software center and i cant find it [21:18] samael6: I'd look under/usr/share/doc.... [21:19] does anybody know how to set up a user with proper priveleges to use SWAt to manage my samba shares? [21:19] samael6: and you can always ask dpkg (-l or -L, not sure) [21:19] it's -l [21:20] what's the package name of the guide [21:21] ubuntu server guide [21:21] ubuntu-server-guide [21:22] there's a guide as a package? [21:22] well i can't install it with apt-get [21:22] iuno, that's what samael6 is saying [21:22] its ubuntu-serverguide [21:22] !info ubuntu-server-guide [21:22] Package ubuntu-server-guide does not exist in maverick [21:22] !info ubuntu-serverguide [21:22] Package ubuntu-serverguide does not exist in maverick [21:22] or not [21:22] :P [21:23] well it is actually ubuntu-serverguide [21:23] !ubuntu-serverguide [21:23] !info ubuntu-serverguide [21:23] i followed your advice and found the package bubt it wasent what i expected [21:23] its a log [21:24] yeah me too [21:24] !info ubuntu-server-guide lucid [21:24] !info ubuntu-serverguide lucid [21:24] Package ubuntu-server-guide does not exist in lucid [21:24] ubuntu-serverguide (source: ubuntu-docs): The Ubuntu Server Guide. In component main, is optional. Version 10.04.3 (lucid), package size 371 kB, installed size 2608 kB [21:25] maybe you'll have to install ubuntu-docs [21:25] to get the offline version installed in your docs directory [21:25] !info ubuntu-serverguide lucid [21:25] ubuntu-serverguide (source: ubuntu-docs): The Ubuntu Server Guide. In component main, is optional. Version 10.04.3 (lucid), package size 371 kB, installed size 2608 kB [21:25] but anyway just google filetype:pdf ubuntu server guide and get the pdf [21:25] no, the package exists [21:26] yeah but it doesn't install the server guide in /usr/share/docs [21:26] non sever related but what are you guys using to player your music in on linux? [21:26] (it goes hand in hand with sever administration though) [21:27] server* [21:27] you can ask that question in #ubuntu if you want recommendataions on desktop music players [21:27] i just use rhythmbox [21:28] fair enough [21:29] it is installe in /usr/share/ubuntu-servguide [21:29] *installed [21:29] NDROfTheLine: ^^ [21:30] thanks i got it [21:30] NDROfTheLine: and its dpkg -L which gives you a list [21:31] NDROfTheLine: sorry, I forgot who asked the original question [21:31] but i thought it was a application thats shows the html files [21:36] no prob it wasn't me [21:37] dpkg -l by itself returns the entire list of isntalled packages [21:38] anyway [21:38] i'm still trying to get swat working. how do i configure a user that can administer samba?? [21:40] im not looking for commands im looking for where this can be viewed [21:42] open firefox. [21:42] in the location bar, type /usr/share/ubuntu-serverguide/html/C/index.html [21:44] guntbert, how'd you figure out the storage location of the ubuntu-serverguide? google? dpkg? [22:01] NDROfTheLine hes probbaling dozing [22:01] zul: /win 19 [22:01] arrrrgh [22:01] well i missed his logoff due to a timeout [22:02] zul: quick questions. I have a init script I pass this in rules: dh_installinit -p gfs2-utils --no-start --no-restart-on-upgrade -- start 65 S . start 2 0 6 . [22:02] what languages do you guys under your belt [22:02] have [22:02] zul: in the init script Default-Start: S and Default-Stop 0 6 [22:03] zul: but when installing, this happens: update-rc.d: warning: gfs2-utils start runlevel arguments (2 3 4 5) do not match LSB Default-Start values (S) [22:03] update-rc.d: warning: gfs2-utils stop runlevel arguments (0 1 6) do not match LSB Default-Stop values (0 6) [22:03] ideas?> [22:08] SpamapS: ^^ === jadakren is now known as airtonix [22:45] Welcome to #ubuntu-server, psywiped [22:45] trying to set up a script to twitter my servers ip address but it keeps saying that gt is not a valid command using this page as a guide http://onthefencedevelopment.com/?p=289 [22:46] psywiped: gt isn't defined as a command anywhere there. Pastebin your script, redacting anything sensitive please. [22:48] http://pastebin.ubuntu.com/565591/ [22:48] psywiped: And the exact, copy pasted error you get? :-) [22:49] heh; line 11 [22:49] $CURRENT_IP > $IPFILE [22:49] Ah, tere it is. [22:49] > is the html entity for > [22:49] There we go. :-) [22:49] I wondered why they were backgrounding something. [22:50] http://pastebin.ubuntu.com/565593/ [22:51] o hell [22:51] that would make sense [22:51] so echo $CURRENT_IP >; $IPFILE [22:52] without the ; [22:54] ok now im getting this http://pastebin.ubuntu.com/565597/ [22:55] I think that's just because of the space between -- and user [23:00] use # to comment out in bash? [23:04] correct [23:06] it looks like the space between -- user was causing the problem [23:07] :) -- alone means something very specific to most commands [23:08] now to figure out why its not updating twitter [23:09] my issue is that i dont trust the dynamic dns im using right now to update if my servers ip address changes and i want a backup method [23:10] but the script doesnt look like its working [23:11] http://pastebin.ubuntu.com/565605/ [23:11] but no update in twitter [23:14] this is what i'm going off of http://onthefencedevelopment.com/?p=289 [23:17] forever alone [23:20] RoAkSoAx: re the update-rc.d warning.. that just means that the postinst is override the LSB header. [23:20] overriding [23:37] New bug: #716767 in clamav (main) "clamav-daemon does not start because of "Malformed database" daily.cvd" [Undecided,New] https://launchpad.net/bugs/716767