=== cpg|away is now known as cpg [03:02] I created a VM with ubuntu-vm-builder, and now I want to move it to another server [03:02] is there anything special? where are the files stored? [03:03] dunno [03:03] it all depends how you built it :) [03:03] generally just move the disk image, cow2? then the config, libvirt export? [03:03] sudo ubuntu-vm-builder kvm lucid --arch 'i386' --mem '512' --rootsize '10240' --swapsize '512' --kernel-flavour 'server' --hostname 'omegarelay' --domain 'home.hilltop.local' --mirror 'http://archive.ubuntu.com/ubuntu' --components 'main,universe' --addpkg 'openssh-server' --name 'ryan' --user 'ryan' --pass 'ubuntu' --ip '192.168.1.4' --mask '255.255.255.0' --net '192.168.1.0' --bcast '192.168.1.255' --gw '192 [03:03] .168.1.1' --dns '192.168.1.1' --bridge 'br0' --libvirt 'qemu:///system' [03:04] that's how I built it [03:15] ah, I see [03:15] it made a directory called ubuntu-kvm in the current directory [03:16] and a run.sh === hggdh_ is now known as hggdh === cpg is now known as cpg|away === cpg|away is now known as cpg [06:22] i want to setup a php/apache server but no mysql === ibiris|afk is now known as ibiris [07:06] can 32bit ubuntu see more than 4Go of ram? [07:08] anepanal1ptos: no you should use 64bit ubuntu [07:08] well, i RTFM'd it, and the answer is yes and no. [07:09] gotta see if i have PAE [07:09] yeah my problem is someone has compiled a shitty driver and it doesnt work with linux x64 [07:09] and woohoo i have pae === ibiris is now known as ibiris|afk === ibiris|afk is now known as ibiris === Err404NotFound is now known as Error404NotFound [09:42] i have two lun /volume for iscsi 1st is regular iscsi (mpio) and another is sanboot while insallting ubuntu on sanboot volume/lun1 ,every thing was good but while installing grub it give error so and cant proceed to install grub ,how i should fix this issuse [09:45] samba35: did you have both luns connected during install? the installer can sometimes be a bit confused and install grub in the wrong place. I've seen that happen. if that's what happens, try to disconnect all but one lun and reinstall [09:46] ok [09:47] when i was trying to install 2nd ubuntu installtion that time my 1st ubuntu was off [09:47] but... was both iscsi luns connected? [09:48] i am trying to installed this on vmware [09:48] and btw, aren't you using vmware for this? wouldn't a vmfs make things a bit easier? [09:48] point [09:49] how do i make vmfs to get installed grub ? [09:49] vmfs doesn't care, but it's probably easier for the ubuntu installer to install on a virtual disk instead of handling iscsi itself [09:50] ic [09:50] how to fix this? [09:51] I'd try to install as normal on vmfs [09:51] single virtual drive [09:51] expand as needed later [09:51] you'd probably want to use lvm on top of that [09:52] sorry no idea of lvm [09:52] well, never mind [09:52] it's not needed, but makes things a bit easier to resize / reallocate space in linux later [09:52] google it [09:52] learning lvm is complex ? [09:53] no, it's easy [09:53] ok [09:53] ok thanks will get back to you [09:53] do you have any idea on dhcp options [09:54] maybe... [09:54] i was trying tftp boot from dhcp server which option i have to use 66 and 67 ? [09:55] how do i test tftp server is running ,when i check the serice and netstat it say its working/running but when i tryed to telnet localhost port/service of tftpd it did not work [09:56] telnet won't work, since tftp is udp [09:56] tftp localhost [09:56] get /somefile [09:57] that file must exist, obviously [09:57] ok [09:57] under the tftp root [09:57] telnet doesnt work on udp ports ? [09:57] telnet uses tcp [09:58] ahh ok thank you adding for my knowledge [09:58] you can use netcat or nmap to probe udp ports [09:58] nmap is usually easiest [09:58] ok [10:14] brb [11:40] how do I setup my bind server to be a slave with no upstream replication to my windows servers? [11:46] jeeves_moss: slave or cache-only? [11:48] veq, I think it'll be a cache only, but in the future, I'd like to be able to dynamically update the Windows boxes with external dynamic clients. [11:48] vezq, is it simple, or a huge pain in the butt to make work? [11:49] vezq: cache-only is easy: https://help.ubuntu.com/community/BIND9ServerHowto#Caching_Server_configuration [11:50] and allowing bidirectional dynamic updates? [11:53] jeeves_moss: http://support.microsoft.com/kb/275866 (funny place to find instructions) :) [11:55] vezq, lol thanks. I have a RADIUS server that will be taking care of some WiFi access points, and I would like to make sure that they dynamically update the DNS IP entries [11:56] jeeves_moss: ok, haven't done dynamic setup myself [11:57] vezq, neither have I, but I think it'll be a million times easier for administration when I have to deal with issues instead of looking up IPs, etc [11:57] jeeves_moss: yep sounds wise [11:58] vezq, I just need to make sure that things work. But........ [11:59] vezq, I'm just waiting for parts to arrive from ebay [11:59] vezq: what WiFi APs you are using? [12:00] meant jeeves_moss [12:00] vezq, it's a toss up right now if I should buy MikroTik APs or build up some Raspberry Pis with all the parts. It's all the same price, but with the Raspberry Pis, I can do a lot more [12:01] jeeves_moss: will there be many APs? [12:02] vezq, if my marketing plan goes well, that's the hope [12:02] recommend to check out also this http://www.ubnt.com/unifi [12:03] vezq, free? [12:03] nope, just cheap with good features [12:04] vezq, thanks! I'll have a look. I was thinking using the MikroTik hardware, and a RADIUS backend [12:05] it has a nice management software which supports Ubuntu too [12:05] also supports RADIUS [12:06] vezq, I'll have a look. My biggest issue right now is making sure I subnet the APs out correctly [12:07] vezq, and making sure that the DNS servers dynamically updates correctly === cpg is now known as cpg|away [14:16] New bug: #1094271 in autofs (main) "autofs missing directory" [Undecided,New] https://launchpad.net/bugs/1094271 === ibiris is now known as ibiris|afk === LordOfTime is now known as TheLordOfTime [16:53] samba35, the troubled iscsi master? ;) [16:54] no [16:54] ;) [16:54] was playing till now [16:55] now googleing [16:55] thanks [17:12] I can't seem to search a folder server-side with the find command, even though the folder is visible on my samba and I can search it in windows explore. $ find /share/pictures -name dscn [17:13] does the find command not work within subfolders? [17:13] I feel like it obviously has before [17:15] nevermind there must have been a syntax error somewhere [17:16] New bug: #1094310 in nova (main) "package nova-compute-kvm 2012.1.3+stable-20120827-4d2a4afe-0ubuntu1 failed to install/upgrade: ErrorMessage: subprocess installed post-installation script returned error exit status 1" [Undecided,New] https://launchpad.net/bugs/1094310 [17:18] There does not appear to be a minidlna channel; so, is there a limit on the # of files minidlna can store in the database? I was trying to 'force-reload' a music folder with 65,000 & it keeps stopping at 9443, and it doesn't even bother to get the image files at all. [17:21] How do I increase the Inotify max_user_watches for minidlna? [17:23] lvmer: it's probably in /proc/sys/fs/inotify/max_user_watches [17:24] royk: and I can change that permanently in /etc/sysctl.conf ? [17:25] by just adding fs.inotify.max_user_watches = # [17:25] yes [17:25] royk: default seems to be 8192, what exactly does this change? folder update notifications? [17:44] I can't seem to get minidlna to get picture files. There is no mention of an error in the log file for the pictures directory. It just seems to ignore it [17:44] as far as I can tell everything is spelled correctly [17:48] is it possible to have more than 1 media directory? [17:48] it seems anything after #1 is ignored [17:48] media_dir=A,/share/music media_dir=P,/share/pictures media_dir=V,/share/pictures media_dir=V,/share/movies media_dir=V,/share/tv === aarcane_ is now known as aarcane [18:20] I'm having an issue with security groups in eucalyptus can anyone help? === Ursinha is now known as Ursinha-afk [18:52] Hello, I have an issue with my server and I'm afraid my SSD is broken. Anyone care to hear me out? :) [18:52] nope. not individually, just ask your real questions in the channel. [18:54] I have a server running 11.10. My system and all swap partitions are on the SSD but all storage etc is done to a RAID5 volume. I'm trying to transfer a 12MB file to my SSD but it keeps telling me it doesn't want to write and that the disk is full. Although, when I run "df -Th" I can't see the disk in the output [18:55] So I'm afraid that my SSD has somehow broken, and that if I reboot I cant start again [18:55] Is there another command I could use, because I can't seem to check if the disk is full or not [18:57] <_ruben> Plizzo: how about df -i, perhaps you ran out of inodes? [18:57] _ruben: I tried running multiple commands, but neither of them display my system partition [18:58] _ruben: What else can I do, should I start a dd to an image on my RAID? [18:58] <_ruben> oh, i missed the part it missing from the output [18:59] <_ruben> does it show in 'mount' ? [18:59] <_ruben> anything odd in 'dmesg' ? [18:59] Do I just type "mount"? (without quotations= [18:59] <_ruben> yeah [18:59] yep [19:00] And you might want to pastebin the full result of mount, just to provide extra context on your system's filesystem setup. [19:00] !pastebin [19:00] For posting multi-line texts into the channel, please use http://paste.ubuntu.com | To post !screenshots use http://imagebin.org/?page=add | !pastebinit to paste directly from command line | Make sure you give us the URL for your paste - see also the channel topic. [19:00] This is what mount gives me: [19:00] http://paste.ubuntu.com/1473739/ [19:00] My SDD is /dev/sdb, but only sdb1 is mounted, and that's /boot [19:01] <_ruben> line 1 shows / being mounted [19:01] <_ruben> pastbin df -h and df -i as well [19:02] df -h: http://paste.ubuntu.com/1473742/, df -i: http://paste.ubuntu.com/1473744/ [19:03] <_ruben> / is full [19:03] <_ruben> 100% use [19:03] In see that now, I actually misstook the first entry :/ [19:03] Is there a smart way to check where all space is? [19:04] <_ruben> i tend to use ncdu for that, but that's not installed by default [19:04] And my disk is full so I can't install it [19:04] <_ruben> might need to free up some space before being able to install it [19:04] <_ruben> yeah :) [19:04] hehe, yep [19:04] On it! [19:05] <_ruben> check for old kernels .. those tend to pile up and take a fair chunk of space [19:06] <_ruben> and /var/log might be a good start as well, logs can grow fast unexpectedly, i always put them on a seprate lv/partition [19:07] _ruben: I'm checking logs, but the syslog is just about 12MB etc, and the others are way lower [19:07] _ruben: I have the webserver on the raid, so that's not it [19:07] _ruben: Maybe the Plex database has somehow expanded weirdly [19:10] found it! [19:10] The plex Media server log is 41GB.... [19:10] What the hell... [19:12] _ruben: Is there some way to limit file sizes? [19:18] _ruben: You on? [19:19] I've removed the file that was clogging the file system, but df -h still shows 100% used [19:22] Everything works again, thanks guys! :) === BlackDex_ is now known as BlackDex [19:51] <_ruben> Plizzo: plex probably had the file still open :) [19:51] _ruben: Yeah, I noticed that I had a debug setting checked in PMS which is why it was filling up the file so badly [19:52] _ruben: It debug logged all transcoding as well :P [19:52] <_ruben> limiting filesizes isn't a trivial task afaik, one usually uses logrotate software to limit the growing [19:53] Hello [19:53] I was wondering if anybody here is familiar enough with iptables and networking, I am trying to figure out a port forwarding problem. [19:54] I installed 12.04 a while back, then added lxfe and ruby rvm. now some rvm gems won't build, missing dependencies. is there anything weird about ubuntu server which would cause that? or, just broken packages? [19:54] <_ruben> triton_: try asking the actual question(s) :) [19:56] Ok, so I have 3 remote installations, each one uses the same ubuntu OS as a firewall, router, nat. I needed to forward ports from inside the local network to the Internet. On 2 installations out of 3 it worked (same exact configurations) on 1 I can't open the port forwarding though the iptables rules are the same. [19:56] I am looking for some hint on how to analyze a problem where I am trying to do port forwarding and it's not doing the job though iptables have the rules, /proc/sys/net/ipv4/ip_forward is 1, /etc/hosts.allow and hosts.deny only deal with sshd and not these ports (44,55) [19:58] <_ruben> triton_: 1st step, do sudo iptables-save -c and see if those rules ever got matches (as in, are not showing [0:0]) [19:59] Is it against the rules to paste output of such commands in the channel? [19:59] how do I install xfce onto ubuntu server? I have lxde but cannot seem to get xfce. [19:59] pastebin? [19:59] <_ruben> triton_: pastebin indeed [20:01] _ruben: http://pastebin.com/db3AvqQs [20:03] triton_: on the cli, you can say "cat foo.txt | pastebinit" and it will do it for you, automagically :) [20:04] thufir_ : crazy stuff! [20:04] triton_: :) [20:05] triton_: you might have to install something to get that to work, just takes a sec, tho. [20:05] yeah, it's not installed here, wants me to apt-get install. [20:06] <_ruben> triton_: use tcpdump/wireshark/etc to determine where things go wrong, could be the dnat itslef, or the reply traffic for instance [20:07] <_ruben> triton_: and i assume you're not testing from the box itself? [20:07] _ruben: no, I am not testing from the box [20:08] <_ruben> ok, good [20:08] _ruben: but what about that iptables output, it had [73:4384] and [4:228], not 0:0 [20:08] <_ruben> and you do can reach 192.168.10.100:44/55 from that box [20:09] <_ruben> those are packet/byte counters .. when it says 0:0, that rule never matched any packets [20:09] _ruben: yes, the video camera is reachable from the ubuntu box that is the router, I can even vnc and view the camera output while on that machine [20:09] <_ruben> ok [20:10] <_ruben> depending on your network layout, it could be a case of triangle routing, where the return traffic from the camera is going through the box running iptables [20:12] _ruben: well, yes, one thing I cannot confirm is the physical layout of anything in either of those 3 installations. The only thing I know that IS different in this case is that they have some sort of a repeater in the LAN because in that location the cameras are so far away (more than 40 meters I think) and they couldn't get this to work without another powered switch or repeater somewhere... [20:12] ...in the middle [20:14] though I thought that if I can already view the output from camera on the machine itself, the repeater in the middle is probably nothing of consequence? [20:20] ok, thanks [20:22] <_ruben> triton_: doubt that as well .. done any sniffing yet with tcpdump/wireshark/... ? [20:24] _ruben: looking with tcpdump [20:26] _ruben: well, I can see with tcpdump -n -e -ttt -i eth1 that when I hit the box, it is actually forwarding from the external IP to the 192.168.10.100:44 [20:29] _ruben: http://pastebin.com/3AN4Pk79 - this is what tcpdump is telling me [20:30] _ruben: and I replaced my IP with 'my.own.ip.address' though :) [20:32] <_ruben> triton_: is 192.168.10.100 reachable through eth0 by any chance? in that case it's the MASQ rule interfering [20:35] _ruben: that would be strange, the eth0 is the Internet nic. I do have that rule though: iptables -t nat -A POSTROUTING -o eth0 -j MASQUERADE (but in all 3 installations) [20:36] I am going to drop that rule, see what happens [20:37] _ruben: no, the address isn't reachable through eth0 and without the masq rule nothing really changed (for the forwarding) [20:38] <_ruben> wonder why it shows up with external ip then [20:39] <_ruben> triton_: try sudo iptables -t raw -I PREROUTING -p tcp --dport 44:55 -j TRACE .. that should give detailed logging as to which rules are being matched [20:41] _ruben: I guess I missed something, how does it show with external ip? tcpdump shows: my-ip-address.54354 > 192.168.10.100.44, that's all where it shows. But this: 192.168.10.100.34239 > 239.255.255.250.1900 I don't understand. [20:42] I have a logrotate question. Tomcat is configure to rotate catalina logs daily but it does not compress them. I want logrotate to compress these old log files if they are found. I have found that I can do this with the size parameter, but I would rather do it based on a time stamp older than when the log rotate was last run. is this possible? [20:42] <_ruben> triton_: that's traffic originated by the camera itself, probably looking for upnp stuff [20:43] _ruben: I wonder.. maybe upnp should be turned off [20:43] <_ruben> triton_: perhaps, but it shouldnt interfere with the actual probloem [20:44] _ruben: so I added that iptables trace line [20:46] <_ruben> triton_: also, not much sane can be said about this without knowing the exact network layout (interfaces, subnets, etc) [20:48] _ruben: that's true. There are 4 computers, 1 is the ubuntu machine, 3 others are local network, all windows xp. There are 4 cameras (the 10.100 is the master and thus server). The ubuntu machine is the gateway. The ip addresses are all assigned statically. [20:49] it's a small store actually [20:51] <_ruben> triton_: xp machines are in 192.168.10.0/24 as well? and if so, why use port forwarding? :) [20:51] I guess the weird part is only that 2 others are fine, forwarding, translating with the same iptables rules and settings [20:52] _ruben: the xp machines are 192.168.10.2, .11, .12, yes. Why not use port forwarding? What's a better way? [20:53] <_ruben> triton_: if they're in the same subnet, why not connect directly to 192.168.10.100? [20:54] _ruben: oh, but the camera is supposed to be visible to the outside. The port 44 allows viewing what's happening inside the store from a browser (with some activex or java applet) and port 55 allows the same for a smartphoen [20:55] _ruben: you are right, when they connect to the camera on any one of the windows machines inside for example, they just go to the local address of the camera and it works [20:55] <_ruben> triton_: do you want to reach those cams from those xp boxes or from outside that netwrk? [20:56] _ruben: from the outside, the inside is not important at all [20:56] <_ruben> ah ok [20:56] they can see what's happening inside without cameras : ) it's a small store, maybe 100 square meters. [20:57] <_ruben> hehe [20:57] it's like a security service, when you have a few stores in one network, then it makes sense [20:58] <_ruben> triton_: one possibility, is the default gateway configured correctly on the camera? [20:58] I have to check on that. I'll try [21:02] _ruben: I'll have to ask the service guy who installed this for some credentials to get into the admin interface, I only can get as far as using vnc now to open a browser on that ubuntu machine and hit the 192.168.10.100 :44 (this does give me a correct response) but I don't know how to administer the cameras without camera server credentials. [21:03] _ruben: thank you for all the help that you provided really [21:03] I think it's time for me to ask somebody who is physically there to help out [21:03] it's another country :) [21:05] _ruben: I only build and supply store and chain management software, supply chain management, resource planning and such. They asked me to help them with the cameras and I was able to in other stores, but this one is giving me too much headache [21:07] it's also crazy funny (or not) to observe immediately attempts by some hostile machines to break into the vnc session that I opened temporarily via ssh. It's crazy how infested the Internet is [21:11] <_ruben> hehe, yeah [21:16] Ok, have a good day, night, morning or evening! [21:16] bye [21:25] for future reference, if one is using ssh to start a vnc server, one might as well tunnel the vnc connection through ssh as well [21:26] friends don't let friends send keystrokes over the internet unencrypted === cpg|away is now known as cpg [22:28] hi all [22:30] here my partitions: http://pastebin.com/bwtLzf7t when i try access to: cd /var/lib/mysql show me a Error Message: "Permission Deneied" i try with: sudo cd /var/lib/mysql but show me thesee rror: sudo: cd: command not found why? need format or unmount or? thanks [22:49] any way to pin a package from autoupgrading? [22:50] i.e. i have to either dist-upgrade or install [package] to force it to upgrade [22:50] or in other words, is there a way to hold back a package when i do apt-get upgrade