[00:00] piercedwater: not sure, but Ubuntu works ok on these machines. I have a 2450. booting off the internal SCSI, but have a SATA PCI card for a pair of SATA drives for my actual server directories. [02:23] where does ubuntu store the uuid for the /dev/mapper for cryptodrives in the initramfs? [03:03] i went to edit /etc/initramfs-tools/conf.d/cryptroot only to find that i do not have one. [03:18] I attempted to schrink some logical volumes and now /home and /var will not mount [03:19] Can somone help me fix that [03:27] how come im on an encrypted system yet /etc/initramfs-tools/conf.d/cryptroot does not exist? how do i modify my kernels cryptroot file? [06:06] hi, i'm on an ubuntu server, and i'm finding that my python scripts are randomly terminating. i don't have any apparent internal or external signals that would cause this, so i'm thinking that it is a memory issue. does anyone know where i should look for logs to confirm this? [07:09] boxbeatsy: If you can make the issue happen at will, you could try running one of the scripts inside strace and check the resulting (long!) output to see what killed the script? [07:20] jmarsden: aah thanks for the suggestion. i acutally think i found the problem.. i was running 100 instances of polipo not knowing that it has an inmemory cache that over time builds up to 25MB [07:21] OK. 100 * 25MB = ~2.5GB of RAM... shouldn't be a major problem on a server, though, should it? [07:23] i believe my server only has 1.7GB of RAM [07:23] :\ [07:24] is that unusual? [07:24] i thought it was on the small end, but didnt know (pretty new to workign with external servers) [07:24] it's an EC2 medium sized instance [07:26] New bug: #710038 in mysql-dfsg-5.1 (main) "package mysql-server-5.1 5.1.41-3ubuntu12.8 failed to install/upgrade: subprocess installed post-installation script returned error exit status 20" [Undecided,New] https://launchpad.net/bugs/710038 [07:26] Well, it all depends. At work, real physical servers tend to have 4GB to... well, 128GB of RAM :) [07:27] I'm not sure what the common sizes are for virtual server machines in an EC2 cloud, that's a whole different world. [07:28] holy crap [07:28] yea...i just checked..it's a measly 1.7gb [07:31] Sounds like you could either find a different program to do what you want, or hack polipo to set a smaller max cache size, maybe 4Mb instead of 25MB? [07:37] jmarsden: yea im setting it to <1mb now cause i'm not even using the caching capabilities. i'm just using it to assist in proxy authentication [07:37] OK. Sounds like that should work for you. [07:38] thanks for the help :) [07:39] You're welcome. [08:10] Can I get ubuntu server tech support here? [08:28] thisIsNash: Maybe... it's a little late at night, not many people are around... ask your specific question and see who answers :) === fenris- is now known as ejat === windkids is now known as windGone [10:36] New bug: #710084 in mysql-5.1 (main) "package mysql-client-core-5.1 (not installed) failed to install/upgrade: trying to overwrite '/usr/bin/mysql', which is also in package mysql-client 5.5.8-1.linux2.6" [Undecided,New] https://launchpad.net/bugs/710084 === WaVeR` is now known as WaVeR [13:34] Dustin Kirkland in? [13:41] Woutje123: he's been idle for 21h... not likely. [13:41] aah [13:41] Thanks for replly... [13:41] reply === windkids is now known as windGone [15:18] this is so cool. I am actually chatting from my blackbwrry === Chat8152 is now known as evon [15:20] hope youre not driving [15:20] lol no [15:21] now I can get help with my ubuntu box if it is broken [15:25] evon: It would be cool indeed when you're chatting from your blackberry while being in cairo [16:01] how can I check to see if my NIC is running at gigabit speeds? [16:03] could copy large files, or look at the lights on your switch [16:03] many have a different color or some way to indicate what speed [16:05] dmesg |grep eth0 [16:08] ethtool [16:10] Mr. Kirkland? [16:41] hello? [16:42] howdy [16:43] hey compdoc [16:43] I'm trying to set up 'new mail' emails from my server to another email account [16:43] I'm not sure how to [16:43] any ideas? [16:43] I don't want to just forward the mails [16:44] I'm using Postfix on Ubuntu [16:44] if that helps [16:45] hang out and see if someone helps. Im not a Postfix guru by any means [16:45] you can also google postfix issues - doesnt have to be on an ubuntu system [16:46] I'd use mailbox_command and insert echo "New Mail" | mail whoever@wherever && before it, except I disabled it because it stops execution, and messed up my Maildir config [16:46] thanks anyway :) [16:46] usually, one account can have several email addresses, if thats what youre doing [16:46] yeah, I've had a look, it doesn't seem to be a common problem [16:47] yeah, I've set up some aliases, so for example webmaster -> root [16:47] you know more than I thought :) [16:47] and if I wanted the whole email to get sent, I'd just put other email addresses in .forward [16:47] thanks :) [16:48] but I want at most the subject in the notification, if that [16:48] Hi [16:48] hey [16:49] Dragonshadow: do you know Postfix well? [16:49] I've got a kind've odd postfix problem and was directed to this irc to ask for help xD [16:49] daisy: lol ^ [16:49] lol [16:49] well, lets see if we can help each other [16:49] what's your problem? [16:49] I'm trying to set up 'new mail' emails from my server to another email account, without just forwarding the email [16:50] and without using the mailbox_command variable [16:50] yours? [16:50] Recipient address rejected: gmail.com when trying to send mail [16:50] receiving works just fine though [16:50] is your isp blocking port 25? [16:51] this is local testing with telnet [16:51] try telnetting p25 from outside [16:52] it could be that your isp requires you to relay outgoing messages through their servers, to stop spam [16:52] this is on a dedicated server so... [16:52] ok, so not a dynamic ip then? :) [16:53] nope [16:53] and telnetting from outside works [16:53] can you send to non-local, non-gmail accounts? [16:54] gimme one to try [16:54] lol [16:54] root@nightcoat.info [16:54] be nice :) [16:55] I won't spam :p [16:55] Recipient address rejected: nightcoat.info [16:56] hmm [16:57] Btw, I don't really understand what you want to do [16:57] only forward new mail? [16:58] or rewriting the headers so it looks like its supposed to go to the new mail [16:58] well, I get emails to nightcoat, and I want my gmail address to be notified that there's a new message, without just forwarding the message [16:58] if that makes sense? [16:58] that could get rather spammy if you could do that [16:59] each time you get an email on nightcoat you'd get an email notif that you have new email on your gmail [16:59] yeah, I know [16:59] hopefully I'll be able to set some rules [16:59] like max daily or whatever [16:59] http://serverfault.com/questions/228828/postfix-new-mail-notification-to-gmail is this you? xD [16:59] XD [16:59] yup [17:02] can you post your main.cf? [17:04] its kind've a mess [17:05] embarrassed? I won't laugh [17:05] I honestly dunno if anything in there is a security flaw so I just query'd it to you :p [17:08] I'm wondering if it might be reject_unknown_recipient_domain [17:12] brb [17:15] Dragonshadow: Mine has reject_unknown_recipient_domain too, I don't think it's that [17:17] you've not got anything for mydestination, any reason? or any myorigin. [17:19] hello [17:19] i installed ubuntu 10.10 server [17:20] and also installed ubuntu-desktop [17:20] now my question is how to run gnome ? [17:20] why do you want to run gnome on the server? [17:20] I do :) [17:21] sometime it is helpfull to troubleshoot problem [17:21] to install no machine [17:21] no machine ns [17:21] no machine nx [17:21] azertyu: it'd be easier to just install desktop ubuntu, and apt-get apache etc [17:22] I did what daisy suggests - works better [17:22] I don't think you want to be messing with X/gnome manually [17:23] i got ubuntu-desktop and apache allready present on my sys [17:23] doing an ssh session -X how to run ubuntu-desktop ? [17:24] startx doesnt work? [17:26] back [17:26] daisy: you setup to use amavis aswell? [17:26] it's on my to-do list :) [17:26] hrm. [17:27] I don't think it [17:27] nah [17:27] content filter wouldn't touch domains [17:27] Can I see your main.cf? [17:27] it could be a banned from: to: address combination [17:27] yeah, ok [17:28] Hi. I need to login ubuntu server remotely. I am in different place. But that server is in LAN environment. So how could i login through ssh or any? [17:28] anybody here with power management expertise? [17:28] pm-suspend and that stuff? [17:28] no startx not working also daisy [17:28] awanti: you'd need to connect to a machine on it's network [17:28] and then ssh from it [17:28] azrtyu is gdm installed? [17:29] awanti: if you're behind a router you'll need to set up port-forwarding [17:29] or you could do that.. [17:29] lol [17:29] by going to 192.168.1.1 or wherever, it'll probably have it's own config page [17:30] you can use traceroute to find it's address, it'll prob be the first entry [17:30] plz. How? Can u give steps or any ...? [17:30] yes gdm allready present [17:30] ok, type 'traceroute google.com' and paste the first few results here [17:30] yes gdm allready present Woutje123 [17:31] what does "sudo /etc/init.d/gdm restart" do? [17:32] restarts gdm :P [17:32] Dragonshadow, I've query'd it to you [17:32] I see it, thnaks [17:32] thanks even [17:32] lol [17:32] thanks daisy.... ;-) [17:33] Mr. Kirkland not online yet? [17:34] restart gdm but there is no display on my screen [17:34] less pedantically, it'll incorporate changes to config files [17:34] aah [17:34] sudo dpkg-reconfigure xserver-xorg [17:35] anybody know why my eth0 wol setting are reset just before suspend? [17:35] done Woutje123 [17:36] Thanks Dragonshadow & daisy i found the answer thanks once again [17:36] yw [17:36] no worries [17:36] now restart gdm again [17:37] W123: maybe something to do with powersaving [17:37] when i do:"sudo ethtool -s eth0 wol pug" [17:37] ethtool eth0 says: "Wake-on: pug" [17:37] then i issue "sudo pm-suspend" [17:38] it doesn;t wake from unicast [17:38] restart but there is no display , don't forgot i connect to my server with ssh -X [17:38] when I wake the server with "wakeonlan [mac-address]" [17:38] it wakes [17:38] does your motherboard support wol? often they don't, or they default turn off certain features [17:38] then ethtool eth0 says:"Wake-on: g" [17:39] I can wake it using magic packet [17:39] hmm, waking from off, and waking from suspend are totally different :) [17:39] I used to be able to wake it with unicast (ping) [17:39] wake from suspend doesn't use the bios [17:39] I think with either 9.10 or 10.04 [17:40] also make sure it's not set to turn off the nic, on suspend [17:40] azerty, sorry bout that. [17:40] all that stuff i let you do wont work..... [17:40] patdk-lap: how would I go about that? [17:41] :D [17:41] I wouldn't know, I never suspend linux machines :) only windows [17:41] patdk-lap: remember I can wake the system with magic packet (from suspend) [17:41] above you said it wouldn't? [17:41] so what exactly will and won't it do? [17:42] patdk-lap: I set ethtool to wake on unicast, physical and magic packet [17:42] when I issue a suspend it will only wake from magic packet [17:42] oh, so wakeonlan works [17:42] when I check the settings after suspend and wake it's reset to only wake on magic packet [17:43] W123: you say the same machine used to wake from unicast in a previous ubuntu? [17:43] that is correct daisy [17:43] the setting doesn't survive reboots/suspend [17:43] azerty: what happens if you do a "gnome-session"? [17:43] add that setting into /etc/network/interfaces in post-up [17:44] havent tested reboot, but suspend doesn't work [17:44] I think it's reset just before suspend [17:45] otherwise it should wake up and have the wrong settings for the next suspend/wake [17:48] W123, perhaps it's just not persistent, and isn't being set at waketime [17:48] annoying, but have you tryed this: http://ubuntuforums.org/showthread.php?t=234588 ? [17:49] the init.d script might help [17:50] bye guys [18:28] from xterm unable to connect to ubuntu-desktop [18:29] unable to connect with how? [18:35] from ssh [18:35] this is what i got if i try startx from my xterm [18:35] http://paste.ubuntu.com/560318/ [18:42] eh? you don't start X from there, you just start programs, like xterm or something else [18:43] so what you are trying to achieve? [18:48] 550 5.1.1 : Recipient address rejected: gmail.com [18:48] azertyu: there is a package called xvfb which is a x frame buffer...simply do apt-get install xvfb then tunnel x through an ssh command example: ssh -X root@192.168.1.5 and then you can run any X program though the frame buffer and it will appear on your machine. I use it all the time to run firefox on servers that don't have GUI remotely. [18:55] ok [19:10] back [19:10] yes me also i use to do same as use thesheff17 [19:10] but the problem i can't able to run startx from xterm [19:11] i got this error : http://paste.ubuntu.com/560318/ [19:11] it looks like the xserver is already started [19:11] did you try to hit ctrl-alt-F7 [19:12] you can also try to do /etc/init.d/gdm restart should restart everything for x. [19:12] or are you trying to do this all remotely? [19:16] yes correct [19:17] azertyu: well I never tried to run the whole xserver remotely....usually people use vnc or freeNX [19:18] i isntalled freenx on server [19:18] i use no machine X from client [19:18] i also installed ubuntu-desktop on my server [19:19] azertyu: you might also ask in the ubuntu irc. I don't know about running an entire xserver remotely. [19:19] when i try to connect to my server from no machine X i just only got xterm opening but can't open ubuntu-desktop [19:19] you shouldn't be able to [19:19] cause you need a way to access it [19:20] be it vnc, nx, or some other remote desktop viewer [19:20] and if i do startx from server i got error that error : http://paste.ubuntu.com/560318/ [19:21] patdk-lap: i installed freenx on my server and nomachine x on client [19:21] can't open ubuntu-desktop that's my problem [19:21] well, you shouldn't be running startx [19:21] cause it's already running [19:21] ok [19:21] well, sounds like you need to configure nx correctly then [19:21] to get access [19:21] then why i don't have display ? [19:21] how did you login to it? [19:21] what you meean " correctly " ? [19:21] ssh? [19:22] no from nomachine x [19:22] and if it doesn't give you a display, then something is messed up [19:22] follow the instructions on how to set it up again [19:25] questions about pgsql ? in unbuntu server [20:02] hello to all. anybody here with some knoweledge about networking problem debuging. i have a ubuntu server and all people say it is heavy slow. i need somehow to analys or better isolate where the problem is located. IS it My ISP with the Fixed IP that is maybe bad Routed, is it my Linksys WRT54GL Router or is it my Ubuntu Server itself. [20:02] anybody here how can help me a little maybe ? [20:07] xperia: wireshark will log everything that is going in/out of the server on the network. top will show you current cpu/ram load on the server. Sounds like your clients are connecting through wireless....if you hook a cable directly to the network is it faster? [20:09] After installing Kubuntu 10.10 my NFS mounts does not work anymore. I just get "mount.nfs: Connection timed out" when trying to mount them. How do I fix this? I am asking here in hope there is someone who knows NFS as I get no reply in #kubuntu. [20:09] thesheff17: thank you a lot for your answer: [20:09] i just maked a ping to one of my domains where my server is running and got very high ping times of about nearly 1 Second. [20:09] 64 bytes from zux182-249.adsl.green.ch (80.254.182.249): icmp_req=9 ttl=64 time=0.933 ms [20:09] http://pastebin.com/bFS0VWMG [20:09] Asking me what here the Problem could be [20:13] xperia: high ping times usually means bad internet connections on one of the two sides. Or bad routing. You can use tracepath on linux to see where it is routing through. [20:14] xperia: also maybe slow DNS lookup. See if you get the same ping times just pinging the IP. [20:15] ahh okay yes i have bind running ony my server. could be really the bind server the problem [20:21] your local bind is unlikely to be causing that problem [20:22] xperia: 0.933 ms is almost one millisecond, not almost one second :) [20:23] New bug: #710319 in apache2 (main) "CPU usage is incorrect on server-status page" [Undecided,New] https://launchpad.net/bugs/710319 [20:23] xperia: Those are fast pings not slow ones :) [20:28] xperia: yea if your dns lookup is local then that prob isn't the problem. Try to run speed tests on both sides of the connection. [20:29] hmm okay 0.8 ms is sure fas i thinked it is more like 800ms but how does it looks on your side really outside of the lan ? [20:29] can maybe someone ping my server and say the times with [20:29] ping www.wificom.ch and ping 80.254.182.249 [20:29] thanks in advance for help [20:29] thesheff17: No, I think he just misread the ping output [20:30] fox news ftw! http://gibboni.apcdn.com/full/38135.jpg [20:30] xperia: That is not pingable from here -- do you have it firewalled off? [20:32] jmarsden: hmmm will just look at it but normally at least the router should bepingable. maybe you can try port 80 ip adress 80.254.182.249 [20:32] to ping this for sure must works as it is open over the router to the server [20:32] xperia: You can't "ping" a TCP port :) Ping uses ICMP. [20:33] you can ping a tcp port, tcpping :) [20:34] ohh tcping would be great at port 80 ip adress 80.254.182.249 [20:34] or domain www.wificom.ch port 80 for testing [20:34] patdk-lap: Well... OK, but that's not really a ping according to W. Richard Stevens TCP/IP Illustrated :) [20:35] :) [20:35] ya, it's more of a single port, portscan [20:37] xperia: httping http://www.wificom.ch gives me lines like: connected to www.wificom.ch:80, seq=0 time=513.75 ms [20:38] jmarsden: thank you really a lot. you are great. that is at least some numbers that i can work with it [20:39] can you maybe make same for my isp site www.green.ch to see what he has for a number to see how does it looks on his side ? [20:39] httping http://www.green.ch [20:39] 513.75 ms is quite a lot that is half second [20:40] 64 bytes from webz.agrinet.ch (81.221.254.34): icmp_req=1 ttl=244 time=136 ms [20:40] xperia: httping http://www.green.ch is a little better: connected to www.green.ch:80, seq=2 time=372.49 ms [20:41] xperia: Bear in mind this includes some response from the server, httping http://www.yahoo.com and see what you get. Maybe 280ms or so. [20:41] hmmmm 100 ms faster but still a lot [20:41] and how does it looks for google from your side as reference [20:41] httping http://www.google.ch [20:42] For google I get 64 bytes from iad04s01-in-f99.1e100.net (72.14.204.99): icmp_req=1 ttl=55 time=35.8 ms [20:43] the sheff17 are you from swiss itself like me. you have very low numbers one time 35.8ms and other time 136ms [20:43] I'm in Chicago, IL USA [20:44] I'm getting 70ms or so to Google, from Southern California, USA. [20:44] google though may be able to do some crazy routing to the closet server. [20:45] *nearest server. [20:45] haha yea...if I use www.google.com or www.google.ch it both resolves to the same IP for me: 72.14.204.99 [20:47] thesheff17 yeah they have a lot if geoip stuff [20:47] i just maked this here and got shocked [20:47] httping www.google.ch [20:47] PING www.google.ch:80 (www.google.ch): [20:47] connected to www.google.ch:80, seq=0 time=588.58 ms [20:48] that is 0.5 Second for Pinging Google while you have 0.03 [20:49] what is on my side wrong hmmm [20:49] let ping google itself with normal ping [20:49] try using the ip only and see if you have the same response time. [20:49] ping www.google.ch [20:49] PING www.l.google.com (209.85.143.99) 56(84) bytes of data. [20:49] 64 bytes from dy-in-f99.1e100.net (209.85.143.99): icmp_req=1 ttl=55 time=51.8 ms [20:50] so with normal ping is much better only 51ms [20:50] while with httping it need 500ms or better 10 times more hmmmm [20:51] thesheff17: pinging the ip itself gives same result as pinging the domain [20:51] ping 209.85.143.99 [20:51] PING 209.85.143.99 (209.85.143.99) 56(84) bytes of data. [20:51] 64 bytes from 209.85.143.99: icmp_req=1 ttl=55 time=51.5 ms [20:51] hmmm need to find out why on your side and on my side httping is such high [20:51] mine is around double/triple when using httping gives me 117.76ms and ping gives me 36.5 ms [20:52] httping went down to around ~80.00 ms after a couple pings [20:53] yeah have same effect here. if i repeat the command several times it get very low. looks like some caching [20:55] some network analysis tool would be great which can give information if router is porblem, server is problem or ISP Connection or IP Problem [20:56] xperia: Have you tried mtr? [20:58] jmarsden: unfortunately no. i am total new to this kind of problems but i will just look at it. thanks a lot for the tip ! [20:58] Ok, I need to go, but it might help, it is an enhanced traceroute that shows time and packet loss per hop. [20:59] okay thank you still for your help. saved the numbers for sure that you give me ! [21:04] i think best is that i replace the server with a new one and remove the router and instead put the router software direct on the new Server [21:04] but i will need additional Ethernet Plugs where i can put then the switches for the other Computers [21:12] one last question. does anybody know some server dignossis script/tool that meassure diskacess performance, memory performance and such things and give some information about the server performance itself ? [21:13] sysbench, lmbench, iozone3, bonnie++ ... there are plenty of benchmarking tools around. Try apt-cache search benchmark [21:15] okay thanks jmarsden. allways a pleasure to read your very helpfull posts [21:15] xperia: You're welcome. [21:38] thesheff17, how can my server's ip address which is .107 even be able to connect to my router when iv only allocated ip addresses from .101 to .106? [21:38] it does connect [21:39] but i dont understand y [21:39] heh? [21:44] if by "allocated" do you mean set the DHCP pool? [21:44] in any case, you're probably using a class C subnet, which means any address from .1 to .255 will work [22:14] qman__, im using a bridged connection [22:15] that really has no bearing on subnetting [22:15] hmm [22:16] bridging just means connecting the lines at layer 2 instead of routing [22:16] there's still a router somewhere further down the line [22:17] or nearer [22:17] i was suprised how fast ubuntu server startup after cold boot [22:23] hmm [22:23] on my file server, I've noticed the 'mkdir' command taking a very long time [22:26] creating, moving, and copying file is no problem, just mkdir [22:26] but what i cant figure out is how my guest VM can see 1 network PC even tho the .xml file for the machine has interface type='network' [22:27] not [23:04] finally [23:30] binaryhat: just got home...did it work? [23:31] binaryhat: usually most networks are subnet mask 255.255.255.0 which means anything from 192.168.1.1 to 192.168.1.253 should be able be on the same network. [23:31] binaryhat: and able to talk to each other. [23:32] binaryhat: your DHCP range is usually a subset of that so usually 192.168.1.100-192.168.1.150 is setup for DHCP. [23:51] is there any good reason to upgrade from 10.04 to 10.10 ?