[00:00] Nice is process scheduling. [00:01] no sense of humor. [00:02] soulresin: I have a sense of humor [00:02] i noticed that. [00:02] maxbaldwin: i like the cut of your jib. [00:02] thank you. [00:39] Does anybody know how to change the mirror from the command line (apt-config something?)? [01:15] how stable is ubuntu-server comparing to Debian stable ? OK, stable has very old packages [01:20] It's hard to quantify that sort of thing, really. but given I happily run Debian *testing* on a server, I'd wouldn't have concerns about using Ubuntu on a server [01:46] maxb: I have used sarge for years when it was in testing.. really great ! [01:47] maxb: I have some issues with the sarge packages... too old (again) [01:50] MatBoy: I run 8.04 on a VPS - it hasn't had any stability problems since I first booted [01:51] MatBoy: but it is hard to quantify; use cases vary so dramatically [01:51] hence why "works for me!" is usually not a good answer :) [02:04] p_quarles: I have a server running about 2 years now with Ubuntu and without any problems [02:04] so [02:04] Ubuntu is well documented [02:04] Debian isn't [02:04] and old [02:04] and politics [02:05] and no support when you barely need it [02:05] so [02:05] ubuntu all the way ;) [02:05] (but I need to change the root-pass ;)) [02:05] whatever, Debian's great too [02:06] and Lenny is on the verge of release, and I understand has fewer outstanding bugs at this point than Etch does [02:06] perfectly reasonable to use that for a high availability esrver [02:06] server, even [02:17] p_quarles: yes true [02:17] but the question if if they will make it with the release [03:27] hi. i am looking for the drive to my ZTE modem, model MF622 HSDPA USB MODEM. [03:28] where can i find it? [06:26] New bug: #318649 in mysql-dfsg-5.0 (main) "not sure was loading update and system reported a bug" [Undecided,New] https://launchpad.net/bugs/318649 === erichammond1 is now known as erichammond [07:21] anyone in here set up mailgraph? [07:21] i'm just getting broken images on the cgi webpage [08:53] moin [09:27] <[gnubie]> i don't have a gui and i am running ubuntu server 8.04 lts.. is there a detailed documentation about the upstart since sysv initscript is gone? === sleepingyoyo__ is now known as sleepingyoyo [10:45] New bug: #318703 in nagios-plugins (universe) "nagios check_smtp expects integer instead of double" [Undecided,New] https://launchpad.net/bugs/318703 [10:48] hello, i have running an opvpn server, and now i am wondering why there are 2 ip's in the tunnel, my server is 10.23.0.1, but what is 10.23.0.2 good for? [10:48] tun0 Link encap:UNSPEC HWaddr 00-00-00-00-00-00-00-00-00-00-00-00-00-00-00-00 [10:48] inet addr:10.23.0.1 P-t-P:10.23.0.2 Mask:255.255.255.255 [10:52] endpoint [10:54] Nafallo: there is no connection, and the endpoint gets an ip address from a pool like 10.23.0.2 - 10.23.0.254, and also, there could be much more connections as just one [11:09] c64zottel: because you're using a point to point tunnel, which needs each endpoint defined [11:10] if it makes you uncomfortable or you experience any oddities, you can switch to using tap devices instead [11:10] Deeps: i guess i am reading it right now in opvpn FAQ: ifconfig-pool option use a /30 subnet, cause of some window-stuff [11:10] i am right? [11:11] on point to point tunnels, yep [11:11] so, it is a connection between linux and the real endpoint [11:12] Deeps: thank you [11:14] all i did was expand on what Nafallo said [11:19] Deeps: no offense, but just endpoint gives me nothing, ok, i didn't ask that much details, but i expected a bit more. [11:25] Q: If I use ubuntu-vm-builder, how do I log in? [11:33] Jeeves_ > you can specify a login and password [11:34] How> [11:34] ? [11:34] Where? [11:34] ah its in the arguments somewhere :) [11:34] !canonical-- # No manual entry for ubuntu-vm-builder [11:34] Error: I am only a bot, please don't think I'm intelligent :) [11:34] just run it without arguments [11:35] You cannot enter a password [11:35] well I use --user and --ssh-user-key [11:35] but I think there is a password argument too [11:35] It's really great that canonical writes all these scripts [11:35] and my version isn't the default one in hardy [11:35] But *SO* annoying that there is no documentation [11:35] I'm running Jaunty [11:36] then it should be fine [11:36] give #ubuntu-virt a try you may get better support [11:36] if you run it without argument is should display the documentation though [11:36] Thanks [11:36] didn't know that was around [11:36] yann2: 'man' should always work :) [11:37] I didn't say there wasn't any problem :) [11:53] I'm looking for a suitable FTP-server along with a web interface to administer FTP-users [11:53] any suggestions? [13:05] cant remember :x [13:05] oups wrong windows :) === MohammadBoozary is now known as Mohammad[B] === ball_ is now known as ball [14:53] hmm [14:54] upgraded my 8.04.1 server to .2, new kernel etc [14:54] now on reboot it hangs configuring network interfaces [15:02] <_ruben> .2 is out? must've missed that announcement [15:04] i didnt realise it was either til the machine didnt come back from reboot and i saw it in grub [15:04] appears to be sendmail hanging [15:11] indeed, at boot time, it brings up eth0, which in turn tries to bring up sendmail, which attempts to do a host on eth0's ip, which hangs [15:12] bringing eth0 up, brings sendmail? [15:12] what did you do to your /etc/network/interfaces? [15:12] indeed, it's in if-up.d [15:12] no idea why [15:12] why? [15:12] # This script is called when a network device comes up. [15:12] # [15:12] # Here is where we'll start sendmail if needed. [15:12] # [15:12] # Written By Richard Nelson [15:12] hehe [15:12] you are wrong, ubuntu doesn't provide that in that file [15:13] that was a custom made change [15:13] i certainly didnt! [15:13] oh, sendmail [15:13] it's in if-down.d too [15:13] post-down even [15:13] ah, directory? [15:13] hmm? [15:13] i thought it's in your /etc/network/interfaces file [15:14] nope, certainly not [15:14] so, sendmail fails to start? [15:14] root@router:/etc/network# find /etc/network| grep sendmail [15:14] /etc/network/if-down.d/sendmail [15:14] /etc/network/if-up.d/sendmail [15:14] /etc/network/if-post-down.d/sendmail [15:15] eth0 ip 192.168.1.254, at boot time it tries to bring up sendmail, which in turn tries to host 192.168.1.254, and it hangs there [15:15] well, it tries to birng up eth0, wihch in turn prompts the sendmail in if-up.d [15:15] what's in /etc/hosts? [15:15] 127.0.0.1 localhost [15:15] 127.0.1.1 router.truman.lan router [15:15] 192.168.1.254 router.truman.lan router [15:15] ? [15:16] that doesn't look good [15:16] remove the middle line [15:16] i added the 3rd line just before reboot attempt number 12 [15:16] remove the second line [15:16] leave the 3rd [15:16] removed [15:16] killed bind, host still hangs [15:16] now reboot [15:17] brb then, this machine doubles as my adsl router [15:17] :) [15:20] nope, still hangs [15:21] commented out auto eth0 from interfaces, rebooted, got a shell prompt, tried to birng it up and it works [15:21] take it down, kill bind, do it again, it hangs [15:21] watching ps ax indicates that when host times out, it tries again [15:21] so, network starts before bind [15:21] and sendmail script tries to find a hostname of that ip [15:22] yep [15:22] that shouldn't hang if everything is ok in /etc/nsswitch.conf [15:22] * Deeps looks [15:22] hello ivoks ! [15:22] mathiaz: hi [15:22] Deeps: grep ^hosts /etc/nsswitch.conf [15:22] hosts: files dns [15:22] egrep even :) [15:23] line above you [15:23] Deeps: and your /etc/resolv.conf? [15:23] search truman.lan [15:23] nameserver 127.0.0.1 [15:23] nameserver 192.168.1.254 [15:23] middle i added recently [15:23] in an attempt to fix (which clearly didnt work) [15:24] hey. anyone thats good with routing between two subnets? [15:24] i figured lo would be up by then, and connect refuse host, and cause it to carry on, didnt account for sendmail to loop host til it succeeded [15:24] Deeps: how about removing those nameservers? [15:24] Deeps: that's a bug in sendmail script [15:24] pretty sure this behaviour is new, just dist-upgraded to .2 [15:24] Deeps: sendmail didn't change in 8.04 [15:25] :> /etc/resolv.conf; host 192.168.1.254; times out [15:26] I got two subnets connected. 10.0.0.0 net is where the gateway for internet is located. 10.0.1.0 net has internet access but cant connect to other comps on the network. is there a way to route traffic between pcs on the two nets? [15:27] Deeps: does 'host localhost' works? [15:27] ah [15:27] found the problem [15:27] i'm using bind9-host [15:28] switched to normal host and it works properly [15:28] well, bind9-host is default [15:28] and that shouldn't be a problem [15:28] it wasn't on here [15:29] i dont think it was, anyway [15:29] might have been the ovh box it wasn't [15:30] killing bind, using bind9-host, it times out looking up 192.168.1.254 [15:30] killing bind, using host, it successfully looks up [15:30] dnsutils depends on bind9-host | host [15:30] bind9-host works on my machine without bind with no problems [15:31] without any reachable nameservers? [15:31] for resolution of ips in your hosts file? [15:31] right, it doesn't [15:31] there's the problem then i think [15:31] not even for hosts in /etc/hosts [15:32] hm [15:32] Deeps: you shouldn't use localhost|local ip as a DNS, anyway [15:32] :) [15:32] but that doesn't change a thing [15:33] well it's the local caching nameserver for the lan [15:34] host should read /etc/hosts first [15:34] bind9-host appears to ignore it [15:34] it doesn't work as expected, it ignores /etc/nsswitch.conf [15:34] that's a bug [15:34] so i've found two bugs in one go? nice [15:34] only one :) [15:35] sendmail looping host isnt a bug? [15:35] no, it relays on a working functionality [15:35] could be a whishlist bug [15:36] wont let me file a bug against it [15:36] https://launchpad.net/ubuntu/hardy/+package/bind9-host [15:36] it's a bind9, not bind9-host [15:36] alrighty [15:37] https://bugs.edge.launchpad.net/ubuntu/hardy/+source/bind9 [15:41] Deeps: good catch [15:42] zul: ping [15:42] ivoks: yo homie [15:43] ivoks: can you confirm? https://bugs.launchpad.net/ubuntu/+source/bind9/+bug/318828 [15:43] Launchpad bug 318828 in bind9 "bind9-host does not respect /etc/nsswitch.conf" [Undecided,New] [15:43] zul: bug #227410 [15:43] Launchpad bug 227410 in bacula "[SRU] catalog Backup fails because .my.cnf is not read" [Medium,In progress] https://launchpad.net/bugs/227410 [15:43] host has alwasy just asked dns, "getent hosts" is a much better way of going through ns [15:43] zul: that's fixed in hardy proposed, and it's fixed on native upload in intrepid and jaunty [15:43] ivoks: ok [15:44] zul: SRU is requested, and people reported that proposed version works for them [15:44] zul: what else do we need to get this into updates? :) [15:44] I frequently use host vs getent hosts to determine differences between dns and hosts in case of trouble.. [15:44] ivoks: people reporting it in the bug report saying that it works? ;) [15:44] zul: it worked for one guy and me [15:44] zul: no one else said anything :) [15:45] ivoks: please update the bug repot then [15:45] maswan: either host needs to respect nsswitch.conf, or the sendmail startup script needs to use getent hosts or whatever instead; there's a bug somewhere though [15:46] Deeps: I'd say sendmail fix is needed, documentation for host say that it is a DNS lookup utility, not a generic name lookup utility [15:47] maswan: please comment on bug 318828 then! [15:47] Launchpad bug 318828 in bind9 "bind9-host does not respect /etc/nsswitch.conf" [Undecided,New] https://launchpad.net/bugs/318828 [15:47] More of a user friendly dig than a good way of determining how hosts resolve [15:47] Sure [15:48] what purposes does the hosts line in nsswitch serve then? [15:49] Deeps: For everything that resolves through libc, those rules apply. [15:49] ok [15:49] * Deeps reboots now with host instead of bind9 host in the hopes it works [15:49] since host is for diagnosing dns etc, it has it's own resolving logic [15:50] i would expect it to read /etc/nsswitch.conf, or, at least, my /etc/hosts [15:51] back to normal [15:51] * Deeps files a bug against sendmail too then [15:52] ivoks: Well, it'd still differ in some corner cases I'd expect, since it doesn't use the libc resolver. [15:54] i wonder if apt and v6 will play nice again now too [15:54] ivoks: hosts in ldap, nis, etc are very unlikely to be queried and parsed by host [15:54] true [15:54] this is a sendmail bug [15:54] :P [15:54] Deeps: sorry for misleading you [15:54] that was my first instinct too [15:55] Deeps: i'll close your bug :) [15:55] ta [15:56] "sendmail uses 'host' instead of 'getent hosts' to determine local ip resolution"? [15:56] yeah [15:56] Deeps: right; you really love sendmail? [15:56] not really, it was installed by default [15:56] by default? [15:56] sendmail? [15:56] I'll not comment on the bug then [15:57] maswan: you can, sure [15:57] i assume so, i cant see any reason why i'd have installe... oh [15:57] probably installed it for a website i was developing for a friend [15:57] Actually, I just had it all typed out. So "comment", then head off for dinner. :) [15:57] Deeps: we install postfix by default, and there could be an error where exim could get installed, but no, not sendmail [15:57] we love our users :D [15:58] postfix isn't installed, nor was it by default (mail-server task wasn't requested) [15:58] Deeps: if you install any app that requires MTA, you'll get postfix or, in some strange cases, exim [15:58] wasn't a packaged app, i just needed an MTA for the oscommerce site i was customising [15:59] sendmail was what i'd used in the past without fuss [15:59] ok [15:59] i've just heard people complaining on debian's packaging of sendmail, which we include in ubuntu too (in universe) [16:00] * Deeps removes [16:05] New bug: #318828 in bind9 (main) "bind9-host does not respect /etc/nsswitch.conf" [Undecided,Won't fix] https://launchpad.net/bugs/318828 [16:33] One question about routing: I got two subnets connected. 10.0.0.0 net is where the gateway for internet is located. 10.0.1.0 net has internet access but cant connect to other comps on the network. is there a way to route traffic between pcs on the two nets? [16:33] not just internet from the router on 10.0.0.1 [16:36] alaz: do you mean that 10.0.0.0/24 is on eth0 and 10.0.1.0/24 is behind eth1. and you wat acces from that computer to both subnets or what [16:48] alaz: the default router for clients on the 10.0.0.0/24 subnet needs to know to route 10.0.1.0/24 to the ip address of the machine in hte middle of both subnets [16:51] alaz: likewise the default router of the machines in 10.0.1.0/24 needs to know to route 10.0.0.0/24 to that same machine [17:10] Deeps: the router is connected to 10.0.0.3. so i should create a route 10.0.1.0 - 10.0.0.3? [17:12] alaz: i dont understand your network layout, if you could draw a diagram it'd be easier to explain to you [17:13] Deeps: og give me a few minutes [17:19] I'm looking for a suitable FTP-server along with a web interface to administer FTP-users, any suggestions? [17:19] ebox with whatever ftpd's included in main? [17:19] Deeps: right, forgot about ebox. will check it out! thanks [17:20] hello! [17:21] any idea how to fix this? http://pastebin.com/m3516281a [17:26] anyone? [17:26] hello there [17:26] looks painful spiekey [17:27] I had the same issue once.. very painful [17:27] can't remember how I fixed it [17:27] doesnt work with aptitude neither? [17:27] Can anyone tell me how can I setup ubuntu's vnc to start running since the logon screen? [17:27] whats the command line command? i dont like the gui [17:29] Deeps: http://jabrahamsen.com/Drawing4.jpg [17:29] Please, give me some link for quick gid. (for dummer):-D [17:30] alaz: is 10.0.1.1 also on the 10.0.0.0/24 network? [17:30] yep [17:30] alaz: what's its 10.0.0.0/24 ip? [17:31] eth0: 10.0.0.3 eth1: 10.0.1.1 [17:31] has its own dhcpd [17:31] 10.0.0.3 is assigned to another pc according to your diagram [17:31] Deeps: its just a typing error. [17:31] ok [17:32] on 10.0.0.1, add a route to 10.0.1.0/28 via 10.0.0.3 [17:33] Please, give me some link for quick gid for server configurations. (for dummer):-D [17:34] z10dej: http://doc.ubuntu.com/ubuntu/serverguide/C/ [17:35] Can anyone tell me how can I make Ubuntu's VNC to work on the logon screen? [17:35] Deeps: thenks. Russian server ha [17:35] Deeps: Rassian server is clear in this qastion:) [17:43] spiekey: not sure why yann2 thinks it's painful...dpkg is the right tool to use, but you need to find out why nagios is complaining...need more detail than what you are showing [17:43] sre there any package disadvanatges on the 64 bits version ? [17:43] *disadvantages [17:44] anyone here knowledgeable with ipv6 and apt? [17:44] zoopster > basically when a package's uninstall script doesnt work (breaks, return 1) the script doesnt get removed [17:45] yann2: understood, but that doesn't mean its painful...there is always a reason for it [17:45] yes but well it took me quite a while to figure it out last time :) [17:45] and last time I think the reason was a bug [17:46] yann2: ah...painful to YOU. Ok. [17:46] zoopster > well I'm all listening :) how do you do when a package refuses to uninstall? but thanks for the nice hint that I'm stupid :) [17:47] yann2: wow...wrong side of the bed today? [17:47] or bad english, I must have misunderstood :) [17:49] yann2: I only said that it's painful to you. I have had the same problem and quite recently...I dug through to figure out what caused it and fixed it...turns out that my problem was a bug as well, but the remedy was rather simple...install again and remove correctly [17:49] how do you reinstall if its already installed? dpkg-reconfigure? [17:51] yann2: sorry...too much to write there...the package was partially removed...or I guess partially installed. [17:54] alaz: what tool did you use to draw your network map? [18:09] Deeps: are you sure that there even exists modules for any kind of ftp-server for ebox? [18:10] frippz: well pro and pureftpd both use system user accounts for their authentication by default [18:10] frippz: as does vsftpd [18:10] frippz: so once you've installed it, all you need to do is add users without shell access [18:10] Deeps: ah well, that's true. I used that method before but with webmin [18:10] i'm sure ebox has user management in it, which is what you originally asked for iirc [18:11] !webmin | frippz [18:11] frippz: 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. [18:11] ^^ that would be why i recommended ebox [18:11] Deeps: Error: "^" is not a valid command. [18:11] ah :D [18:12] damn... I better remove those install that are already present then :) [18:12] *webmin installs [18:12] * frippz purges webmin from five servers at once ^^ [18:15] Deeps: I was actually thinking about using MySQL auth or something like that [18:16] frippz: proftpd has a mysql auth module [18:16] Does this line mean that the httpd process is started by init? It's a line from ps -ef: www-data 13833 1 0 13:56 ? 00:00:01 /usr/sbin/httpd [18:16] frippz: vsftpd might too [18:16] dinsdale07 > try pstree [18:16] but yes [18:16] this will be a server where clients uploads stuff to us, but they should all have their own home dir, so maybe system accounts will be better suited for this [18:20] yann2 - with pstree the same process shows up as perl: " |-perl,13833 ". I don't understand how to find out about how this process was started. [18:22] htop on the other hand shows /sbin/init as the PID [18:30] Deeps: are there any alternatives to ebox? [18:41] frippz: sure, loads, none of wihch are supported by ubuntu though [18:41] hello. i've followed https://help.ubuntu.com/community/Exim4 after that I couldn't use SMTP from my home e-mail client so i've changed dc_relay_domains from '' to '*' and now it seems that my exim is not asking for any authentication. why isn't it asking for authentication? [18:41] Deeps: yeah, I read up a bit on it just now [18:42] Deeps: I find ebox a bit lacking. I'll have to check it a bit more before I pass any real judgment though :) [18:59] zul: i don't understand pitti's comment :) [19:01] ivoks: bug # again? [19:04] sorry, wasn't here :) [20:15] hi [20:15] has anybody tried zfs on ubuntu? [20:15] New bug: #318915 in mysql-dfsg-5.0 (main) "max_connections fixed to 886" [Undecided,Confirmed] https://launchpad.net/bugs/318915 [20:19] albertico: since it doesn't exist, no [20:20] hmm... what about the zfs-fuse? [20:20] did that ever go anywhere? [20:21] p_quarles, I have read about issues on performance [20:21] p_quarles, then, any suggestions on a filesystem with snapshot/copy-on-write capabilites? [20:22] right, but is that proejct still even in active development? haven't heard anything frmo zfs-fuse in a while [20:22] btrfs is in the newest kernels [20:23] p_quarles, everything I have found so far points only to experimental use of btrfs for now [20:24] p_quarles, have you tried it? [20:24] no, I have not, and yes, it is still experimental' [20:24] but unlike zfs-fuse, it is actually still being actively developed [20:25] p_quarles, ok... so I read... [20:25] in any event, I think a need for those features would mean using Solaris at this point [20:25] p_quarles, how about ext3cow... any advice you can give? [20:26] not something I'm familiar with [20:26] but, it sounds like a workaround, and I prefer to avoid those if possible [20:26] particularly since there is a free version of Solaris now [20:27] p_quarles, well, there is someone marketing win 2003 shadow copy on my work... [20:27] I don't follow [20:27] p_quarles, so I am looking for a similar functionality [20:28] Oh, I see what you mean [20:28] p_quarles, it's a windows "feature" that creates snapshots of files [20:28] like I said, OpenSolaris gives you ZFS -- Ubuntu doesn't, and won't have that until btrfs is here [20:30] p_quarles, ok... I will stick to my strategy of using vfs plugins for samba... or maybe switching to freebsd for my file servers... [20:30] albertico: then't you'll move back when brtfs goes stable :D [20:31] ivoks, yep ;) [20:34] p_quarles, thanks for your feedback [20:34] sure [20:51] New bug: #318954 in krb5 (main) "Please sync krb5 1.6.dfsg.4~beta1-5 (main) from Debian unstable (main)." [Wishlist,Confirmed] https://launchpad.net/bugs/318954 === erichammond1 is now known as erichammond [21:54] hey guys [21:55] I am looking for a cheap data center to host a DNS server for me. This will be a tertiary DNS server running TinyDNS simply to make sure all of our DNS are not hosted on the same subnet/AS/datacenter [21:55] New bug: #318975 in kerberos-configs (universe) "Please sync kerberos-configs 1.22 (main) from Debian unstable (main)." [Wishlist,Confirmed] https://launchpad.net/bugs/318975 [21:55] can anyone recommend a trustworthy company that will be fairly cheap? [22:01] New bug: #296719 in kerberos-configs (universe) "kerberos-configs fails to configure if dnsdomainname fails" [Undecided,Fix committed] https://launchpad.net/bugs/296719 === max_ is now known as maxbaldwin [22:35] New bug: #318994 in mysql-dfsg-5.0 (main) "package mysql-server-5.0 5.0.67-0ubuntu6 failed to install/upgrade: subprocess post-installation script returned error exit status 1, server cannot bind to port" [Undecided,New] https://launchpad.net/bugs/318994 [22:43] Sendmail ques re Ubuntu 8.04: Sendmail is running on my sendmail.server box. As seen from there, port 25 is open (nmap -p25 localhost), but closed when viewed from other Ubuntu 8.04 boxes on local network (nmap -p25 sendmail.server). On sendmail.server, firewall is off, TCPwrappers allows all traffic from local network, and there are no errors in mail.log. /etc/pam.d/ holds no sendmail modules. Local network is a subdom [22:43] ain of 192.168, and sendmail's access file set to "192.168 RELAY". Any ideas on what could be causing this weird behavior? [22:45] Additional info: All firewalling takes place at the gateway, not between boxes on my domain. === max_ is now known as maxbaldwin === erichammond2 is now known as erichammond === erichammond2 is now known as erichammond === erichammond2 is now known as erichammond === maxbaldw1n is now known as maxbaldwin