=== The_Pugilist is now known as The_Pugilist|awa === ppetraki_ is now known as ppetraki === veebull_ is now known as veebull_away [05:04] does anybody here use lxc? [05:04] I imagine a lot of people do. [05:04] heh. i'm a bit curious. how do i assign an ip to a container on the same subnet as the host? [05:05] I imagine you'd need some sort of bridge set up on the host. [05:05] i'm having difficulty finding decent instructions [05:06] oh ok. so it HAS to be a bridge? [05:06] Yep, same as any other virt tech. [05:06] i just figured i could change the veth in the container to the same subnet but that seems like it would be a bit problematic [05:07] hm. i'll keep looking for some good instructions on how to do it. can't seem to find anything that is straightforward [05:15] https://wiki.debian.org/LXC/SimpleBridge <- i'm guessing these similar instructions would work in ubuntu? [05:20] hm. that doesn't make sense. you lose the physical interface on the host and replace it with br0? [05:38] jpds: so let me get this straight.. if you remove eth0 from the host.. you can still access the internet on the host through the bridge with the ip? [05:38] that just seems strange [07:08] hi all. I installed 'standard' a ubuntu 64 bit server , 13.10 version then upgraded to 14.04 version. I installed lamp as 'extra' packets and everythings seems fine. I need ssh or scp root access in order to connect directly for some programming, but i can't login as root in any way. Note that i'm trying to login form local lan and the server is on my local lan. Accessing to the server via ssh wi [07:08] th the 'only one' user that i created during installation works properly and then i can use sudo to act as root. I already readed and searched some documentation about it ma all the suggestions doesn't solve the problem. [07:08] *ma=but [07:08] Good morning. === a1berto_ is now known as a1berto [07:13] gigirock: then make sure you have a root password and root login is enabled in sshd.conf [07:14] Ben64, yes according with all the docs, i use PermitRootLogin yes in sshd_config [07:15] gigirock: Have you given root a password? (as Ben64 says) [07:21] lordievader, mmmh no i assume that the root user has the same password as the default user.... [07:21] and now you know what happens when you assume [07:22] Hihi [07:22] do you mean i have to give "passw root" command ? [07:22] gigirock: you have to sudo passwd root [07:22] do "sudo -i" to give yourself a root shell, then you can use "passwd" [07:23] The root users is disabled by default, that means it doesn't have a password. However it can still be used by logging in with other means (keys, sudo, etc). But you can also set a password for root. [07:23] also.. if it is just a matter of using a root shell you can always just sudo -s [07:23] Damned.... the password can be the same of the only one user that i have on that server ? [07:24] you can set it to whatever you want [07:24] personally i'd use a different password for root but to each their own [07:24] thank you... i will send a box of beer.......next time [07:25] sorry people, i have another problem about that ssh , but on a centos server.... :) [07:27] can't help you there [07:27] i'm still a bit confused about bridging [07:27] Ben64, i have to switch to #centos-server :) ? [07:28] i'm not sure the channel for it [07:28] in ubuntu you disable the eth0 interface.. create br0 and use eth0 as its port ? [07:28] and you should still be able to hit the ubuntu box on the network ? [07:30] Macer, as far ai I know you have to build a bridge in that way, but normally you will use br0 only for 'other' connection than your local connection [07:31] gigirock: yeah i'm trying to put lxc containers on the same subnet as the host [07:31] akin to fbsd jails [07:31] i guess this requires a bridge to be created on the host [07:32] so will the host using br0 with eth0 as its port just work like it normally would ? [07:32] https://wiki.debian.org/LXC/SimpleBridge [07:32] i was just looking at that and noticed that eth0 is disabled in the example. it just seems odd to me [07:32] Macer, i did something in the past but using two physical different eth board... then with routing and bridging [07:33] well i only have 1 physical nic [07:33] but i want to use lxc similar to how i was using jails in freenas [07:33] with the jails being on the same subnet [07:34] i guess by default it uses its own dhcp and nats it [07:38] Macer: My server uses the br0 inteface for its network. The kvm vms tap into that bridge device. [07:38] ah ok. so the br0 simply acts as the net interface? [07:38] instead of eth0 on the host? [07:39] sorry. just never had to manually set up a bridge before that wasn't done automatically for me [07:39] and i guess this is the only way to get lxc containers onto the same subnet as the host [07:39] It does act as eth0 here. [07:39] and all you have is eth0 with no ip and br0 with the host ip? [07:40] or something to that effect? [07:41] i'lll try it out when i'm next to the box [07:41] so i don't have an oops moment heh [07:41] Macer: http://paste.ubuntu.com/7456296/ [07:41] The vnets are the vm's. [07:41] oh ok... nice. thanks :) [07:42] actual VMs.. not lxc containers right? [07:42] i'm sure the setup is exactly the same tho [07:44] oh ok. it looks like it's pretty much by default setting up its own bridge..i'll have to toy with it later. [07:44] Macer: Qemu-kvm vms, yes. [07:44] http://paste.ubuntu.com/7456302/ [07:44] so it makes sense that eth0 would be blank [07:45] thanks tho... that's awesome. appreciate it [07:46] Macer: As you can see from my post, br0 is the master over eth0. [07:48] do you set that up when you make br0 in network/interfaces? [07:48] can i see your interfaces file on your host? [07:49] Macer: http://paste.ubuntu.com/7456307/ === RaptorJesus_ is now known as RaptorJesus [07:49] ah ok. that's pretty straightforward [07:49] i'm guessing it's set by the bridge_ports ? [07:50] Nope, wrote it myself. [07:50] no i mean the "master" part [07:51] bridge_ports eth0 [07:51] Wrote that myself too ;) [07:51] in the lxc example there is no entry for eth0 tho [07:51] https://wiki.debian.org/LXC/SimpleBridge [07:51] it says to comment out the entire section about eth0.. that's where i got a bit confused heh [07:52] in yours you still haveit enabled and set to manual [07:52] I've used this one: https://help.ubuntu.com/community/KVM/Networking [07:54] ah that's awesome. thanks a lot. [07:54] really needed that information :) i'll try it out when i'm in front of the box and see how it goes and set up lxc to be on the same subnet as the host [07:54] thanks again. [07:57] Macer: No problem, have fun. [08:35] lordievader: hm. i managed it set up the bridge [08:35] but i can't seem to hit anything via the internet from teh lxc console [08:35] but i can ping internal ips [08:35] :/ [08:40] Macer: From the host can you ping everything? [08:44] yes [08:44] nameserver is the same as well [08:45] there really isn't anything about the need for routing afaik [08:45] the bridge and the container ip are correct [08:45] and i can ping the local IPs [08:46] Macer: You probably need to allow your firewall to forward those packets: iptables --append FORWARD --in-interface br0 -j ACCEPT [08:46] And possibly sysctl -w net.ipv4.ip_forward=1 [08:46] And possibly "sysctl -w net.ipv4.ip_forward=1"* [08:47] nope [08:47] heh [08:47] that's odd because you'd think [08:47] that this would just act like a typical virtualized nic === stooj_ is now known as stooj [08:56] have to be missing something here [08:56] forwarding didn't work [08:56] i think it may have to do with the lxc default config [08:56] if ican ping inernally i should be able to hit the outside [08:57] Macer: You can ping the host and other machines on your network? [08:58] yes [09:01] how odd [09:03] Can you ping 8.8.8.8 from a vm? [09:06] connect: Network is unreachable [09:07] no. can't ping outside IPs either [09:07] 64 bytes from 192.168.1.1: icmp_seq=2 ttl=64 time=0.755 ms [09:07] but i can ping the internal gateway [09:07] :/ [09:07] How does your kernel routing look? [09:13] although i do keep getting some bridge made by lxc [09:13] not sure where it is coming from [09:19] oh i think i may have found the problem [09:19] although i have no i dea why but service networking restart isn't working [09:19] nor is running the script itself to restart [09:22] ok. got it [09:22] nice :) [09:23] it was because lxcbr0 was still enabled in /etc/default/lxc-net [09:23] https://help.ubuntu.com/12.04/serverguide/lxc.html [09:23] found that there [09:23] it's workign now :D awesome [09:26] so now i'm wondering what happens when lxc tries to update a kernel within the container [09:26] when i do an apt-get upgrade [09:30] Congratulations :D [09:35] hi guys. unsure how to add aliases for a loopback device, i'm wondering if there's a way for me to *generate* /etc/network/interfaces based on the current interface setup (i.e. i've added the aliases using the `ip` utility) [09:47] anyone? this is an example, but not working: http://pastebin.com/USqgBYZ7 (line 6 is a misspelled option - 'address'?) [09:48] based on this http://www.linuxhomenetworking.com/wiki/index.php/Quick_HOWTO_:_Ch03_:_Linux_Networking#Creating_Interface_Aliases [10:00] can you alias lo? [10:22] aliases for lo are pointless [10:23] once you have 127.0.0.1 you have 127.0.0.1 -> 127.255.255.254 [10:23] try pinging random ip from that range :) === shadeslayer_ is now known as shadeslayer [11:00] smb: I tried reproducing the skb error using netcat and inetd echo earlier, but that didn't do the trick. Am now trying with redis pubsub. [11:09] Okay, can reproduce it with redis, but I'll respond to the bug. [11:24] hi there, is this the right channel to ask about issues accessing az3.clouds.archive.ubuntu.com from a cloud instance ? (As in wget hangs at HTTP request sent, awaiting response) [11:51] did we end up implementing a kernel cleanup mechanism? === geowany_ is now known as geowany [12:12] kirkland: i found a thread you started about this ↑ in 2012 on ubuntu-devel but it led to a dead-end. anything change? [12:13] FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! 70943 [12:13] FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! 62488 [12:13] FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! 83055 [12:13] FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! 16288 [12:13] FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! 5669 [12:13] FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! 15688 [12:13] FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! 23815 [12:13] FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! 87886 [12:14] FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! 7981 [12:14] FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! 24824 [12:14] FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! 74771 [12:14] FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! 16577 [12:14] !ops [12:14] Help! Channel emergency! soren, lamont, mathiaz, Pici, Daviey, Tm_T or pmatulis [12:14] FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! 31868 [12:14] FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! 53599 [12:14] FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! 24633 [12:14] FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! FUCK YOU ALL!!! 5348 [12:14] Tm_T: Thanks [12:15] Thank you Tm_T :) [12:15] sorry for being slow [12:44] stephank, Ah ok, I will read up there. Sorry was away for a bit. [12:50] kirkland, congrats for going public ! [12:59] pmatulis, apt-get autoremove (--purge) should get you to keeping 2 (or 3, I cannot remember exactly) kernel versions since I believe Saucy === medberry is now known as Guest51278 === Guest51278 is now known as med__ [14:05] smb: oh really? will look thanks [14:57] hi, when i execute "java -version" i get meaningful result: "java version "1.8.0" ..." if i call the "java" command from within a bash file located in the same directory, i get "java: command not found" [14:58] and? [15:01] why? [15:01] java is installed [15:02] so the bash script including the java call should work [15:05] ah i have also to modify bashrc and not only .profile [15:19] nevermind === medberry is now known as Guest139 [15:28] i have still the same problem, "java" from within bash script does not work if i execute the bash script [15:29] if i call "java -version" from the shell for example, it works [15:29] (same directory) [15:33] try putting the full path to java in the script [15:34] sounds like its in the path for a interactive shell but not in the non-interactive shell path === Guest139 is now known as med__ [15:36] rberg__: i have already the full path in .profile file [15:37] i have also set the full path in .bashrc [15:38] and putting the full path in the script isnt working? or $(which java) [15:39] reberg__ i will try it [15:40] yep, putting full path in client bash script is working [15:41] cool.. are you exporting PATH in your script? possibly overriding the one from .profile? [15:42] now, iam exporting PATH in the client bash script, yes, but before u told me, there were no export statements in it [15:42] i mean [15:42] i set the -cp though [15:42] java -cp ./bin:./bin/lib/* ... [15:43] but this should not affect if java is working in general, i guess [15:44] I dont know much about java really.. is cp class path? [15:44] yep, additional class pathes, e.g. directing to directories containing depending jar files === medberry is now known as Guest17095 === esde_ is now known as esde [16:36] I have a couple upstart jobs created using the pattern described at the bottom of the section http://upstart.ubuntu.com/cookbook/#another-instance-example [16:36] basically a dummy job with no exec that starts some workers [16:36] hii anyone can help me how to transfer/ copy a file from VPS [SERVER] to LOCAL [MY-PC] using SCP please [16:37] this works fine upon boot, however *very* consistently when I "stop workers" and then "start workers" the start will hang in start/starting state with no way for me to see what's happening or recover [17:55] hey all, i have rsyslog watching some files and for some odd reason it just stops logging [17:56] sarkis: did those files get rotated or renamed or something similar? [17:56] i mean it doesn't write the watched files to /var/log/syslog as it should, no errors.. i did notice there is a profile being loaded up in /etc/default/rsyslog and then the profile is present in /etc/apparmor.d/disable [17:56] sarnold: ah they do get rotated... interesting. [17:57] sarnold: i think that is the problem, it works fine until the files are logrotated and then it stops ;( [19:09] I have simple dnsmasq server setup and am unable to load images from twitters cdn. abs.twimg.com [19:10] And if I disable dnsmasq everything works [19:10] images from abs.twing.com will load, that is [19:11] It happens to a few other sites, but most definitely with twitter's cdn services [19:11] Hello i was here yesterday and i talked to someone joesomething or somethingjoe, not to sure. Anyone know who i am talking about and if he is a regular around here? [19:12] Joey Jojo Shabadoo? [19:12] N0C: that may have been dasjoe? [19:12] yes that sounds right [19:12] It was about an N54l microserver and making a BIOS update and some other things [19:12] ah, that's what it was :) [19:13] So [19:13] i got the update going [19:13] it wrote a bunch of stuff [19:13] but now i am just left with a c:> blinking [19:14] -probably- you just hit the power button -- I don't think freedos has any 'sync' or 'umount' or 'shutdown -h now' equivalent commands [19:15] Any ideas? [19:16] It looks like this https://imgur.com/xzPPtlY [19:16] I am fuckin scared to brick the thing [19:16] Question (cross-post on #ubuntu): How can I disable ufw/table to filter inter-bridge traffic? I tried "iptables -I FORWARD -m physdev --physdev-is-bridged -j ACCEPT" but it doesn't help. [19:20] vlad_starkov: what version of Ubuntu are you using? [19:21] IT WORKED! GREAT SUCESS [19:21] Ubuntu Server 14.04 64bit [19:21] I'm also using bonds [19:21] so the config is eth -> bond -> bridge [19:21] vlad_starkov: you might be interested in https://bugs.launchpad.net/ubuntu/+source/ufw/+bug/573461/comments/12 [19:21] Launchpad bug 573461 in ufw "UFW blocks libvirt bridged traffic" [Undecided,Invalid] [19:22] jdstrand: nice :-) [19:24] hi all, I see some ICMP echo request from serverA to serverB, how can I find out which process is sending these ICMP packages? [19:26] jdstrand: made changes to /etc/sysctl.conf. now everything works :-) [19:26] jdstrand: thanks [19:54] nice! [19:56] I wonder who might be interested in running a server with nested raids on ubuntu :P [20:02] Well, i have a couple of these hanging around, so maybe: http://www.networkstorageantics.com/products/jet_stor_416s.php [20:02] Actually, 5 altogether but 2 are different model [20:08] [20:08] whoops [20:09] Hi, i wonder what is the difference between using postfix as an own mailserver or going threw my ISP's mailserver? My ISP has blocked port 25 :P [20:10] it's not suppported to run email servers at home :) [20:11] i mean, is there a difference when running it threw my ISP ? [20:11] there isnt a problem if i do it like that right? [20:12] there will be issues no matter what you attempt [20:12] like what? [20:14] i guess this is cool, right? https://pastee.org/f5za3 :-/ [20:16] dunno, I don't use lvm [20:18] im sure its fine.. [20:20] "Don't worry. It is a problem of os-prober. Here an old quote way back from 2009:" https://forum.manjaro.org/index.php?topic=5438.0 [20:20] meh [20:21] i dont understand it but whatever :p [20:25] tho maybe should file a bug report..... hrm [20:28] Or, just add to bug 1313784 [20:28] Launchpad bug 1313784 in grub-installer "File descriptors leaked on lvs invocation" [Undecided,Confirmed] https://launchpad.net/bugs/1313784 === medberry is now known as Guest30963 === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha [21:07] does anybody know where ubutntu sets up which lxc groups are auto-started? [21:08] by default nogroup is started.. i want to autostart groups on boot but can't seem to find out where it happens [21:08] i guess newer lxc just reads the autostart flag [21:09] Macer: some details on autostart here: https://www.stgraber.org/2013/12/21/lxc-1-0-your-second-container/ [21:20] sarnold: yeah i read that but it doesn't explain how to autostart by group [21:20] Macer: ah :) I know just enough to be dangerous, hehe [21:20] heh [21:20] like if i set lxc.start.auto = 1 in the container.conf [21:21] then it will start but if i add the conainer to a group it does not [21:21] and i dont' understand where you set which groups auto start or how to get groups to auto start [21:21] stgraber,hallyn, Macer has a question about lxc autostarting groups that I don't know how to answer :) [21:22] By default only containers without a lxc.group set will be affected. [21:22] from the lxc-autostart man page [21:22] but it doesn't explain where that is set or where to set it or how to do it [21:22] nor does the lxc.conf man page.. nor does any other documentation i've read so far heh [21:23] i'm just trying to set which groups start on boot [21:25] Macer: that is set in th e conatiner's configuration file [21:26] Macer: it's actually hopefully going to be augmented soon so you can specify lxc.group = onboot,whatever or lxc.group = ,whatever , in either case it woudl be autostarted at boot [21:26] (so you could ahve a container in a group but still have it autoboot) [21:26] Macer: but if you haven't added a group to it, then it'snot in a group [21:26] so just lxc.start.auto = 1 whould work [21:27] (see the lxc.container.conf(5) manpage) [21:27] yeah i saw that. and i added the container to a group [21:27] but i don't see a way to autostart on boot [21:27] correct, right now you can't autostart on boot if it's in a group [21:27] oh ok [21:28] well, you can by editing /etc/init/lxc.conf ... [21:28] well. :) there's 30 minutes of my life i won't get back heh [21:28] yeah, sorry, it's a bit convoluted. hopefully it'll be cleaned up in the next few weeks [21:28] lol. np. but you might want to throw that out there in big bold letters hehe.. i was looking for some setting like above for the lxc.conf [21:28] something like [21:29] lxc.onboot.startgroups = group1 group2 etc [21:29] like you stated earlier heh [21:29] but thanks. i'm glad someone told me before i spent more time on it [21:30] yeah - ii think that'd be a nice feature, but i dno't remember where we ende dup on that. (autostart is stgraber's baby, i don't use it :) [21:30] must run - o/ [21:30] thanks hallyn :) [21:30] so just going to throw stgraber under the bus huh? :) [21:31] thanks a lot hallyn .. appreciate the info [21:31] ok ... well i guess i can move on and start working with this container then [21:32] awesome stuff. been curious about having a linux "jail" that wasn't a chroot [21:37] Macer: by all means do look at the seccomp, apparmor, and unprivileged container bits. sarnold may be able to help you, or i shoudl be back ina few hours, if you have any questions [21:38] heh, under the bus - no i quite like the way he did it, we just need to find 'just the right' api, and i'm leaving that to him since he cares more than i do :) [21:38] * hallyn out [21:42] hallyn: lol. maybe soon. i'm setting up a container now to play with [21:43] genii: ahh, will do, thanks [22:04] how do i apt-get install specific package when there are multiple ones with the same package name? (added via ppa) === RaptorJesus_ is now known as RaptorJesus [23:32] ubuntu server is surprisingly good [23:56] Macer, yes - less is more