=== dendrobates is now known as dendro-afk [00:34] Q" After i installed ubuntu 10 on laptop Dell Vostro 1400 then done but bugs is working freeze? Why ? how can fix it ? === dendro-afk is now known as dendrobates [05:05] is it possible for the alternate installer to start an SSH daemon so one can do a remote installation? [06:03] Hi, is it possible to install some kind of "minimal" gnome in a ubuntu server in order to have a lighter desktop than xubuntu? [06:04] pish: why do you want a gui on a server? [06:05] MTecknology, I'd like a "light" desktop server for development purposes in an old machine (P3/512MB) [06:06] pish: oh.. [06:06] pish: xfce is lighter than gnome - If you want light then check out openbox [06:07] MTecknology, ok thanks. Which one is the distro for ubuntu+openbox? is it lubuntu(lxde)? [06:08] pish: get the ubuntu alternate cd - install just the Command Line system - I think you press F4 at the boot screen to get that - then aptitude install openbox [06:08] pish: warning though - light does not mean easy to learn :) [06:10] MTecknology, ok, I'll consider that. Thanks for your tips. [06:10] np - enjoy [06:10] pish: btw - openbox is what I use every day [06:10] you might also want openbox-themes, obmenu, and obconf [06:13] MTecknology, in an existing Ubuntu Desktop installation, could I install openbox packages just for testing and then remove them and be back to gnome? [06:15] pish: yup [06:15] pish: aptitude install openbox :) [06:16] MTecknology, great. I'll try it right now. Thanks a lot again. [06:32] does anyone know how to limit incomming connections speed based on IP address? e.g. limit each ip max to 1MB/s. Would linux tc do the job, if so how? === rmk` is now known as rmk [08:32] Anybody else in Prague? [08:34] SpamapS: Oh, I'm sure they didn't evacuate the city because you came. :) [10:10] Hi, is there a way to make snapshot on a md volume? and is there a way to run fsck on a mounted filesystem? [10:11] ivo_, depends on the filesystem, but generally no, and md doesn't support snapshots afaik (you need lvm, or a filesystem that has it built in for that) [10:11] :( [10:12] damn I should have sticked with opensolaris [10:12] how else can I be sure my data is ok [10:12] if I have to take my system down [10:12] only to fsck the filesystems [10:12] this is a lot of downtime [10:18] btrfs has checksumming, and will probably be suitable for cautious production use soon [10:19] fsck generally isn't concerned with data corruption, only metadata corruption though [12:41] anyone using mysql with a SA-SCSI 10K drive on Ubuntu ? Any performance issues, driver problems, etc ? [13:12] SandGorgon: I wouldn't expect those problems, actually. [13:14] hiall! Has anyone got an experience run func/certmaster/opensymbolic on Ubuntu Server? [13:16] tried google and lots of other stuff, but nothin seemed to be helpful [13:16] has anyone got hte clue? [13:20] VeeCount: What's that? [13:23] func stands for Fedora Unified Network Controller [13:24] it's some kinda handy tool to manage lots of nodes on yur lan [13:25] VeeCount: hmm, never heard of it [13:26] actually the idea was as follows, i've got about a dozen old machines and wanted to make them work as a cluster [13:26] looked into some solutions likkerrighed, cfengine and so on to automate the processes [13:27] recently i've come upon opensymbolic and it seemed to be the solution, but the manuals are very poor [13:27] VeeCount, have you looked at UEC? [13:27] nope [13:27] what's that? [13:28] Ubuntu Enterprise Cloud [13:28] see if it meets your requirements, since it's going to be way easier to install on ubuntu than anything else would be [13:28] oh yes, sure, but the community frightened me saying that this solution is rather heavy [13:29] well, it depends on what you want to run [13:29] the other issue that i have to keep in mind that the nodes are diskless/// [13:29] UEC works well for VMs from what I gather, but if you just want to run a bunch of threads, might want to go with something more like a beowulf cluster [13:32] thanks qman__ i'll look into that [13:33] I don't know about any of the tools you listed, but it would certainly be possible to set up a PXE boot image for beowulf cluster nodes [13:33] I'm guessing those tools are supposed to simplify the process [13:34] yup, i have no trouble setting the pxe-boot [13:35] using nfs and tftp [13:37] that area is pretty tough to search on, you have to cut through all the buzzwords and advertisements for real info [13:37] the machines are running fine, the thing i want to do is to consume their resources atogether to run a single VM [13:37] but yeah, it basically comes down to exactly what you want to do [13:38] a single VM? that's not really possible with anything I know of [13:38] google showed me such an exaple using xen [13:38] but it didn't seem to be very handy [13:39] i've tried out mpich, but the procecces im using are not very good parallelized [13:39] you can distribute threads pretty easily, but running a VM across multiple machines has a lot of issues, especially regarding network latency [13:40] could you plaese clarify about distributing threads? [13:41] in a beowulf style cluster, you set up all the nodes to share threads [13:41] they share an NFS drive and balance processes across the hardware on your cluster [13:41] okay then, what tools should i look into for balancing processes? [13:42] but any one process can only exist on one machine at a time [13:42] so if your application doesn't have multiple processes, it won't really give you any advantage [13:42] it behaves much like a machine with lots of processors in it [13:42] but with more lag because of the network [13:44] im still working on threading my app, but behaving like a machine with lots of processor is quite suitable [13:44] actually perhaps is the decision im looking for [13:44] still int search for the tools [13:45] I've been looking but I can't seem to find one that's very ubuntu centric [13:45] I did a few years ago, but that's not much use now [13:46] if you're still interested, i can point on Kerrighed, but unfortunately it's not very stable [13:47] and the common problem i face is that when a single node crashes the whole process goes down, and no batch query tools are helpful [13:53] everything that looks promising turns up missing [13:55] this is the most useful thing I've found so far: http://www.linuxquestions.org/questions/linux-newbie-8/beowulf-cluster-on-ubuntu-10-04-a-818999/#post4029508 [14:00] thanks again qman__, i've looked into it, but still it's mpich, at the moment i'm about to try UEC, greatful for your advise [14:00] bye [14:00] good luck [15:02] hi! [15:03] qman__: I looked into UEC, and they say that Eucalyptus nodes are disk-intensive; I/O wait will likely be the performance bottleneck [15:04] so what do you think will it work on a diskless node? [15:05] hi [15:06] hi, discordianfish [15:06] New bug: #606898 in postfix (main) "package postfix 2.7.0-1 failed to install/upgrade: subprocess installed post-installation script returned error exit status 75" [Undecided,New] https://launchpad.net/bugs/606898 [15:06] i'm glad that i found a ubuntu server install cd but now i wonder: can i use it for network install too? looks like i can't change the package source to network or something like that :/ [15:07] i bet this it possible somehow [15:07] during install of lucid server, the virtual michine task installs KVM, right? [15:07] not Xen host (dom0)? [15:08] discordianfish: i've come upon http://ubuntu-tutorials.com/2007/10/11/how-to-configure-pxe-network-booting-on-ubuntu-for-network-based-installations/ will it be helpful? [15:09] VeeCount: i just want to use the network as package source. that has nothing to do with pxe booting [15:10] guess than i'll just use debootstrap again [15:12] okay, so has anyone a clue on setting up UEC on diskless nodes [16:30] growing my raid +1TB :) oh how I love ubuntu server and mdadm [16:48] hi to all [16:48] im newbie in ubuntu server [16:48] please help [16:49] when i try to browse my website is says access denied [16:49] how to troubleshoot it? [16:50] anyone pls [16:55] anyone pls help me [17:07] buzzzzzzzzzzzzzzzzzzz === dendrobates is now known as dendro-afk [17:53] how much performance do i get running a game server at ubuntu server instead running at ubuntu client??? [17:54] because, ubuntu client has to run graphic stuff right? === XyliosPC1 is now known as Xylios [18:54] Hi,i have a machine with ubuntu 10.04 server. it has run well for 1 month, then when i reboot it halt just before booting ubuntu. it is not an hardware problem because with a live cd or another version of ubuntu (8.04) all run well. maybe is grub2 that halt the system? do you have any idea? [18:56] Today i reinstalled ubuntu 10.04, all was right untill i updated the kernel, maybe is thew new kernel? is there a way to know if this is the problem? [19:05] zermann: Boot into the old kernel? [19:06] jpds, it halt before grub2 display anything, is it possible that grub2 halt the system ? [19:06] Unlikely. [19:06] is there a log from grub2? [19:07] No. [19:07] What kind of machinen is it? [19:07] it is an AMD x2 4000 [19:09] chipset nforce 405 [20:22] Hi list I am having 2 webservers A--B on B a website is accessible as http://localhost:9090/portal from internet I want that to be accessible on my public domain name as www.mydomain.com/site so on A I have following vhost configuration http://pastebin.com/c1rb2C0r and on B I have following http://pastebin.com/PfZDxyYT [20:22] some how HTML is broken [20:22] please see if at A ProxyHTMLURlmap I used correctly [20:24] I am using Ubuntu 10.04 with KVM [20:46] \q [20:46] exit [21:25] hi, it is possible that ubuntu 10.04 server do not boot if system clock is set some hours earlier? [21:28] One more question, when i log in i get this stat: Memory usage: 0% but, if i see 'free -m' : -/+ buffers/cache: 380 3517. How is this possible? [21:28] zermann: i hope not, why? [21:29] pmatulis, it is a week that my system halt at boot start [21:29] but sometime is all ok [21:29] zermann: ok, but why do you mention the clock? [21:30] because when it doesn't boot i saw the clock 2 hours in the past [21:30] maybe is a battery problem but, the system do not boot [21:30] zermann: how many times did you see this happening? [21:31] every two or three days [21:32] i notice that because in the past two week there as been storm near my house so i had to halt and unplug the system [21:32] storms [21:32] sorry for my poor english [21:39] zermann: could set the right time and reboot a few times and then put a bad time, reboot, and see what happens? [21:40] zermann: i hope you are not proven right :-) [21:45] pmatulis, i just done that 5 minutes ago, it is not the problem i was wrong. [21:45] I'm having an issue with sending e-mail through my posfix/dovecot system from my WM5 device. I can get my mail over TLS, but I get a failure when I try to send. I've seen losts of posts of the same problem, but no one has seemed to be able to fix it [21:47] Jeeves_Moss: i guess wm5 device does not follow the smtp standards [21:47] pmatulis, is there a work around? I REALLY need to be abe to send from my PPC [21:48] Jeeves_Moss: you will need to increase logging level on smtp listener (master.cf) and post some logs [21:48] Jeeves_Moss: when the device tries to send [21:49] pmatulis, where in the master.cf file do I change the log level? [21:50] Jeeves_Moss: you may actually need to increase verbosity of the tls/sasl component. are you using saslauthd? [21:51] I think I am [21:51] Jeeves_Moss: see if you can start it in debug mode (-d maybe) [21:51] Jeeves_Moss: should say in man page [21:52] Jeeves_Moss: you should also determine if you can send w/o TLS [21:52] ok, one sec. I just have to finish this phone call [21:52] Jeeves_Moss: maybe it's a more basic problem (SMTP) [21:53] Jeeves_Moss: regardless, the logs should tell you what's up [21:58] hey.. have you guys ever used the pure-python Lamson mail server ? I was actually hoping someone who had used it - it seems orders of magnitude simpler than anything else [22:06] pmatulis, ok, sorry, Fiance on the phone. better not hang up on here or my life is over. LOL Bridezilla attacks! === dendro-afk is now known as dendrobates [22:50] hey, does anyone use rtorrent in here? [23:21] hi [23:22] ive got a ubuntu server installation and messed a up dm-crypt experiment, now i cant skip the password question for the drive at boot. its say that i should be able to press S to skip that part, but neither s nor S did work [23:22] its always trying to input a password [23:23] is there a way to boot linux without mounting some partitions? ive tried "single" but the problem still persists