| zul | adam_g, eventlet uploaded | 00:33 |
|---|---|---|
| === dendro-afk is now known as dendrobates | ||
| === dendrobates is now known as dendro-afk | ||
| === n0ts_off is now known as n0ts | ||
| === n0ts is now known as n0ts_off | ||
| === Lcawte is now known as Lcawte|Away | ||
| ClientAlive | has anyone here worked with scm manager before? | 02:36 |
| ClientAlive | what does a guy have to do to learn how to use scm manager? | 02:40 |
| ClientAlive | I can find very little about it on the internet | 02:40 |
| === n0ts_off is now known as n0ts | ||
| === n0ts is now known as n0ts_off | ||
| uvirtbot | New bug: #1034228 in maas-enlist "maas-enlist script hard codes the path to the maas server" [Undecided,New] https://launchpad.net/bugs/1034228 | 03:01 |
| === n0ts_off is now known as n0ts | ||
| === n0ts is now known as n0ts_off | ||
| === cpg is now known as cpg|away | ||
| === reidmv is now known as reidmver | ||
| === reidmver is now known as marut-lurking | ||
| === marut-lurking is now known as turam | ||
| hadees | i'm building a file server for home use, it'll have 18 terabytes using ZFS. I have 8 gigs of ram and 2 30 gig SSD drives that i'm going to mirror as the boot disks. | 04:16 |
| hadees | i'm wondering how much swap space I really need. | 04:17 |
| hadees | since the SSD drives are so small, 16gigs of swap seems like a lot. | 04:17 |
| === n0ts_off is now known as n0ts | ||
| ScottK | You probably won't need any. | 04:30 |
| \sh | ScottK, swap space is always good | 04:31 |
| ScottK | Sure. | 04:33 |
| ScottK | I'd suggest a little, just in case, but 16GB is overkill. | 04:33 |
| hadees | 4? | 04:33 |
| hadees | thats what i'm planning on for now | 04:33 |
| \sh | http://www.linuxjournal.com/article/10678 (good explanation) | 04:34 |
| \sh | anyhow, 8 gigs I give at least 2G, when you do a lot of heavy memory consuming work, give more | 04:34 |
| hadees | \sh: thanks | 04:36 |
| hadees | i'm also mirroring my boot drive, does swap need to be mirrored too? | 04:36 |
| ScottK | No. | 04:37 |
| hadees | so i guess I only need 2gig on each drive if i want 4gigs of swap total right? | 04:37 |
| ScottK | That'd be less efficient then one 4gig swap partition, but sure. | 04:38 |
| ScottK | Odds are you won't need much if any of it. | 04:38 |
| === n0ts is now known as n0ts_off | ||
| \sh | http://www.infoworld.com/t/linux/managing-swap-space-linux-systems-196068?page=0,0 also an interesting article | 04:39 |
| hadees | ScottK: well considering I only have 2 30gig drives for my boot drive 1 4gig swap would waste space right? | 04:39 |
| ScottK | Why are you mirroring the boot drive? | 04:40 |
| ScottK | O/S reinstall isn't very hard to do if one dies and this isn't exactly a mission critical application where a bit of down time matters. | 04:41 |
| hadees | ScottK: it was cheap and i figured why not, also I know I don't need it, honestly I don't need so much space but this is as much about building it as it is using it. | 04:42 |
| stgraber | ScottK: I'd agree that in this case, mirroring is probably overkill, though if going down that road anyway, I'd strongly recommend mirroring the swap space too as it's really quite annoying to have a server that can continue working on drive faillure kernel panic on you because your swap wasn't mirrored | 04:43 |
| ScottK | Good point. | 04:43 |
| \sh | hadees: swap has nothing to do with your boot drives. swap is an addition to your real memory...to give the kernel the possibilty to 'swap' real memory to the disk...(see https://www.linux.com/news/software/applications/8208-all-about-linux-swap-space/ the 'How big should your swap space be' paragraph) | 04:43 |
| hadees | \sh: sorry i'm using boot loosely, i'm talking about the drive that is going to hold the OS | 04:44 |
| stgraber | though nowadays on machine with enough of RAM, I tend to go with no on-disk swap and zram-config to compress some RAM into SWAP. That still lets the kernel swap pages as usual but doesn't kill your disks. | 04:44 |
| \sh | well, disks are replaceable ;) and not that expensive...on the other hand, memory is not expensive, too, so 32G or 64G for a 18TB fileserver sounds good...especially to hold a lot of filecache data ;) | 04:45 |
| hadees | maybe i'll just use one drive for the OS and the other one to play around with some of ZFS's caching features | 04:47 |
| ScottK | Reminds me it's about time to go buy hard drives for my file server again. The ones I have in there now are nearing their 5th anniversary. | 04:48 |
| hadees | ScottK: if you can hold off a year i would, the prices still haven't come down from the flood | 04:48 |
| hadees | my old file server died 8 months ago and I got sick of waiting for prices to drop | 04:49 |
| ScottK | Well. 5 years is about their nominal service life. | 04:49 |
| ScottK | I might be able to wait another year, I might not. | 04:49 |
| \sh | ScottK, for desktop work ;) | 04:49 |
| ScottK | Capacitors get old and stuff. | 04:50 |
| ScottK | Although the motherboard/cpu/ram/power supply they are with were made in 1999. | 04:50 |
| \sh | ScottK, honestly we should go with the time, and use the klout ;) | 04:51 |
| ScottK | Hardy is the last Ubuntu release it can run, so it'll be gone in 9 months one way or another. | 04:52 |
| \sh | .oO(until it explodes like the . com bubble in the beginning of the new century) | 04:52 |
| ScottK | \sh: http://packages.ubuntu.com/precise/owncloud | 04:53 |
| ScottK | Cloud is great, until it's not: http://www.wired.com/gadgetlab/2012/08/apple-amazon-mat-honan-hacking/ | 04:54 |
| ScottK | Anyway ... | 04:54 |
| ScottK | Bedtime for me. Good night. | 04:54 |
| === n0ts_off is now known as n0ts | ||
| === n0ts is now known as n0ts_off | ||
| === cpg|away is now known as cpg | ||
| === n0ts_off is now known as n0ts | ||
| === n0ts is now known as n0ts_off | ||
| Marcio | what do you all think of this graphic card: http://tinyurl.com/8jyv778 ? | 06:39 |
| === n0ts_off is now known as n0ts | ||
| === smb` is now known as smb | ||
| === n0ts is now known as n0ts_off | ||
| === n0ts_off is now known as n0ts | ||
| === n0ts is now known as n0ts_off | ||
| === n0ts_off is now known as n0ts | ||
| === n0ts is now known as n0ts_off | ||
| Nafallo | hey. does anyone know if omcmd (omapi client to do dhcp on-the-fly changes) is packaged yet? | 09:13 |
| === n0ts_off is now known as n0ts | ||
| lifeless | Nafallo: yes. | 09:23 |
| lifeless | Nafallo: or rather no, but that is itself a wrapper | 09:23 |
| lifeless | Nafallo: there is a lower level omapi tool which iworks just fine | 09:24 |
| === n0ts is now known as n0ts_off | ||
| Nafallo | lifeless: omshell? found that already | 09:48 |
| lifeless | yah | 09:49 |
| lifeless | you should be able to drive it via here input. | 09:49 |
| === railsraider__ is now known as railsraider | ||
| feisar | hi, does dump work with ext4 on ubuntu 12.04? | 11:00 |
| jamespage | utlemming, walinuxagent only has context on i386 and amd64 right? | 11:27 |
| === n0ts_off is now known as n0ts | ||
| === n0ts is now known as n0ts_off | ||
| === zyga is now known as zyga-food | ||
| bhosmer | I'm looking for some resources and guides to get me started and point me in the right direction. I have a server with a wireless router connected to it. This server will only be used to connect users to http traffic, but not connect to the outside world. I want to redirect any port 80 traffic to a name, like my-server.name is this bind, DNS or a mixture of both? I know how to set up virtual hosts in apache, and it is listening on the ip address, but | 11:59 |
| bhosmer | like users to be able to at least enter a name in their browser instead of the ip for the machine. I also don't want to block access to the internet at large when I do connect it to a network so that I can run updates and install additional software. I guess my question is, where do I start? | 11:59 |
| ikonia | bhosmer: you mean you want a web proxy server ? | 12:01 |
| bhosmer | ikonia: I suppose, I am running my own web server, but it is entirely self-contained. I am using it for classes I am teaching when wifi is terrible. | 12:02 |
| ikonia | bhosmer: so you just want users to be able to connect to your server ? | 12:03 |
| ikonia | to see classes ? | 12:03 |
| ikonia | is that what you are asking ? | 12:03 |
| === n0ts_off is now known as n0ts | ||
| bhosmer | Yes, I have that now. The server doesn't talk to the outside world, because I am using a wireless access point that only connects the my server which isn't connected to anything else. It works now, but users need to enter the ip in their browser. I want to have a name instead. | 12:04 |
| bhosmer | Does that make sense? | 12:04 |
| ikonia | so either setup a dns server (overkill) or just put the hostname in the users hostfile | 12:05 |
| ikonia | bhosmer: how many people are connecting to this ? | 12:05 |
| bhosmer | Basically my own self-contained internet. | 12:05 |
| bhosmer | I was trying to not edit the hosts file on each machine. I assume I need DNS for this. I have about 75 connecting reliably and downloading files over http. It worked well and was incredibly fast. | 12:06 |
| bhosmer | The network I've solved. I am using three routers and letting the last one in the chain handle dhcp. | 12:06 |
| ikonia | is there a reason you aren't happy with the IP ? | 12:06 |
| ikonia | as setting up a DNS server for this seems overkill | 12:06 |
| ikonia | especially if it's internal | 12:07 |
| ikonia | and most likley getting IP's from the routers which point at other DNS servers | 12:07 |
| bhosmer | My users aren't that savvy and a lot are windows users. That is one of the reasons I would rather not edit each host file as well. | 12:07 |
| bhosmer | Yes, it is all internal and won't talk to the outside world at all. | 12:07 |
| ikonia | bhosmer: why is it a problem just putting the IP in the browser | 12:08 |
| bhosmer | For class efficiency and the users being fairly nooby, it would be nice to give them a name instead of an ip. The ip works, but I want to add the additional polish. | 12:09 |
| bhosmer | Most of them won't remember the ip. | 12:09 |
| ikonia | you need to setup a dns server then | 12:09 |
| jacobw2 | i'm looking at mysql.conf for upstart, how does upstart determine which user to run the process as? | 12:10 |
| ikonia | but although this is a simple task, it's going to be a massive headache from the environment you've described | 12:10 |
| ikonia | jacobw2: it's in the script it calls to start up | 12:10 |
| bhosmer | Okay, that is what I was looking for ikonia thanks. What do I need to watch out for as far as headaches? | 12:10 |
| ikonia | bhosmer: well you've got multiple access points/routers that I'm assuming are acting as a dhcp server and offering "their" dns servers | 12:11 |
| jacobw2 | ikonia: http://paste.ubuntu.com/1135994/ | 12:11 |
| ikonia | bhosmer: so yours will conflict with theirs and not contain the right dns servers | 12:11 |
| jacobw2 | ikonia: can you please tell me which line does that? | 12:11 |
| bhosmer | I only have one acting as DHCP, the other two are bridged into the one so only one is acting as the dhcp server. I was thinking of actually letting my server do dhcp instead. | 12:12 |
| jacobw2 | ikonia: thanks for your very quick response :) | 12:12 |
| ikonia | jacobw2: look at /usr/bin/mysqld | 12:12 |
| ikonia | /usr/sbin/mysqld sorry | 12:12 |
| jacobw2 | ikonia: setuid? | 12:12 |
| ikonia | bhosmer: ok, well, it will cause a conflict with that one rotuer/access point | 12:12 |
| ikonia | jacobw2: is it a binary or a script | 12:12 |
| bhosmer | What if I turn dhcp off an all three then? | 12:13 |
| === n0ts is now known as n0ts_off | ||
| ikonia | bhosmer: then no-one will get IP addresses | 12:13 |
| bhosmer | I can't let the server assign ip's? | 12:13 |
| jacobw2 | ikonia: it's a binarz | 12:14 |
| ikonia | bhosmer: the access points may not accept addresses from external devices | 12:14 |
| ikonia | bhosmer: you'll need to research | 12:14 |
| ikonia | jacobw2: that's interesting, so it's no longer calling mysqld-safe | 12:14 |
| ikonia | jacobw2: (I don't have an ubuntu box to look myself) | 12:15 |
| bhosmer | I'm fine with that. Thanks for the nudge in the right direction though. You've answered my question about where to start. I was a little confused whether I needed to use DNS for this. | 12:15 |
| === cpg is now known as cpg|away | ||
| ikonia | bhosmer: dns is the only real option for name resolution | 12:15 |
| bhosmer | Awesome, that was what I needed. | 12:16 |
| bhosmer | jacobw2: I have a box I can peek at, what were you looking for? | 12:17 |
| jacobw2 | bhosmer: i'm looking for how the upstart script determines the user to run mysql as | 12:18 |
| ikonia | jacobw2: can you look at the process list and grep for mysql and pastebin it please. | 12:18 |
| bhosmer | If I remember right, doesn't upstart basically use shell scripts or something to start them? Let me look on mine. | 12:19 |
| jacobw2 | ikonia: http://paste.ubuntu.com/1136004/ | 12:19 |
| bhosmer | jacobw2: a ps of my mysql shows the process is owned by the mysql user. I looked in /etc/init.d and found a mysql entry, but I I haven't been able to see where the user is set yet. | 12:27 |
| uvirtbot | New bug: #1031479 in python-django-openstack-auth (universe) "not lintian clean" [Medium,Fix released] https://launchpad.net/bugs/1031479 | 12:36 |
| jacobw2 | how can this not be documented anywhere | 12:39 |
| hallyn | ahs3: debian bug 684211 - it built fine in quantal, and locally on sid. But I assume the patch is in fact really needed (to build in debian). | 12:42 |
| uvirtbot | Debian bug 684211 in netcf "FTBFS: netcf (0.2.0-1)" [Normal,Open] http://bugs.debian.org/684211 | 12:42 |
| === groupcat is now known as konk | ||
| hallyn | jdstrand: so for netcf MIR, all the knobs have been twiddled and I can make libvirt build-dep on it? (bug 904014) | 12:46 |
| uvirtbot | Launchpad bug 904014 in netcf "[MIR] netcf" [Medium,Fix committed] https://launchpad.net/bugs/904014 | 12:46 |
| jdstrand | hallyn: yep | 12:47 |
| jdstrand | hallyn: 'Fix Committed' means it is on you to seed it or to add it as a dependency of some sort of something already in main | 12:48 |
| hallyn | \o/ thanks :) | 12:48 |
| jdstrand | np | 12:49 |
| === zyga-food is now known as zyga | ||
| === railsraider_ is now known as railsraider | ||
| === railsraider_ is now known as railsraider | ||
| === mcclurmc_away is now known as mcclurmc | ||
| ahs3 | hallyn: :(. yeah, it is. it built fine in pbuilder, too. | 13:31 |
| hallyn | ahs3: do you want to just push with that patch, or do you want me to send a debdiff? | 13:31 |
| ahs3 | hallyn: i can just use that patch. that work for you? | 13:33 |
| hallyn | ahs3: absolutely, thanks | 13:34 |
| chmac | I have two 12.04 machines, both running apparmor 2.7.102-0ubuntu3.1, with identical mysql configs, and one machine is denying mysql access to /etc/mysql/certs/ while the other allows it. | 13:39 |
| chmac | I could obviously modify the offending machine's /etc/apparmor.d/usr.sbin.mysqld, but I'd rather figure out what the underlying issue is. | 13:39 |
| chmac | How do I debug what's causing the denial? | 13:40 |
| chmac | Aha, simpler than I thought, apparmor simply isn't running on one of the servers! | 13:45 |
| chmac | It won't start, but not sure why, and nothing I can find in the logs | 13:45 |
| uvirtbot | New bug: #1033637 in nova (main) "VM reboot fails with EC 42 (natty)" [High,Incomplete] https://launchpad.net/bugs/1033637 | 13:46 |
| zul | hallyn: ping | 14:00 |
| jamespage | utlemming, ping re walinuxagent when you start | 14:09 |
| utlemming | jamespage: pong | 14:10 |
| jdstrand | chmac: I recommend you update /etc/apaprmor.d/local/usr.sbin.mysqld-- that way you won't be prompted on upgrades | 14:10 |
| jamespage | utlemming, hey - so I wanted to check something | 14:10 |
| chmac | jdstrand: Ok, so that's like a local override of the default policy? | 14:10 |
| jamespage | I'm assuming that walinuxagent only works on i386 & amd64 right? based of the dep which is arch restricted as well | 14:11 |
| chmac | jdstrand: Very useful to know, I'll make that change into my puppet config. | 14:11 |
| utlemming | jamespage: the agent will run on any arch, but I restricted it to x86 arches because that is all that will run HyperV | 14:12 |
| jamespage | utlemming, OK - so reason is that it shows up in the problem report ATM | 14:13 |
| jamespage | utlemming, http://people.canonical.com/~ubuntu-archive/testing/quantal_probs.html | 14:13 |
| jamespage | its un-installable on non-x86 archs (its not restricted ATM) | 14:13 |
| jamespage | utlemming, should we switch it to i386/amd64 only? | 14:13 |
| utlemming | jamespage: I thought it was.....yes | 14:13 |
| jamespage | utlemming, OK - leave it with me then | 14:14 |
| jdstrand | chmac: re policy overrides> yes. you might be interested in /etc/apparmor.d/local/README | 14:14 |
| jamespage | utlemming, OK - I uploaded a new version | 14:15 |
| utlemming | danke :) | 14:15 |
| chmac | jdstrand: Ok, will check it out, thanks | 14:18 |
| hallyn | zul: 'sup? | 14:19 |
| zul | hallyn: for that libvirt bug you added for nova im going to mark it wont fix for nova since its at least 3 versions behind now | 14:19 |
| hallyn | zul: do you figure it is in fact nova's doing? | 14:20 |
| zul | hallyn: could be but they should be running at least precise | 14:20 |
| hallyn | wondering whether libvirt bit should be marked wontfix or invalid | 14:20 |
| hallyn | hm, landscape isnot treating me will this morning | 14:20 |
| soren | hallyn, zul: Which bug is this? | 14:22 |
| zul | soren: https://bugs.launchpad.net/ubuntu/+source/libvirt/+bug/1033637 | 14:23 |
| uvirtbot | Launchpad bug 1033637 in nova "VM reboot fails with EC 42 (natty)" [High,Incomplete] | 14:23 |
| soren | zul: Ok, thanks. No idea what's going on there :) | 14:25 |
| === railsraider_ is now known as railsraider | ||
| zul | soren: neither do i...and its quite low on my list of priorities | 14:28 |
| addisonj | okay... is it normal behavior for mdadm to choose unused block devices and auto create arrays for me? because its realllly annoying | 14:56 |
| xnox | addisonj: did the unused block device still has mdadm metadata stuck in the headers / not-wiped? | 14:56 |
| rbasak | addisonj: sounds like it's picking up on the superblock and autostarting raid | 14:57 |
| addisonj | this is on EC2, new EBS volumes | 14:57 |
| addisonj | here is the relevant section of my user-data script: https://gist.github.com/3295661 | 14:58 |
| addisonj | just added the output of what mdadm gives me when trying to create the raid and /proc/mdstat | 15:00 |
| addisonj | as you can see, I tried doing a stop of all arrays beforehand, but that didn't work... so confused! | 15:01 |
| glebaron | I am getting two error messages that are filling up my syslog that I can't figure out. Hoping that y'all might be able to provide some insight. | 15:05 |
| glebaron | avahi-daemon[919]: server.c: Packet too short or invalid while reading question key. (Maybe a UTF-8 problem?) | 15:05 |
| rbasak | I think you need to run mkinitramfs -u at the end. Otherwise the next time you reboot your EBS volumes already have superblocks on them and are being autostarted to md127. Maybe. | 15:06 |
| rbasak | err, update-initramfs | 15:06 |
| rbasak | s/reboot/start a new instance | 15:06 |
| glebaron | NetworkManager[1315]: <info> Unmanaged Device found; state CONNECTED forced. (see http://bugs.launchpad.net/bugs/191889) | 15:06 |
| uvirtbot | Launchpad bug 191889 in firefox "[MASTER] [WORKAROUND] "Offline Mode" feature fails to detect proper online state for networks that are managed outside of network manager." [High,Fix released] | 15:06 |
| addisonj | rbasak: this is on initial boot... using cloud-formation templates to being up the box with the 8 EBS volumes | 15:07 |
| glebaron | I have read the bug, but I can't figure out what the fix is. | 15:07 |
| rbasak | yeah but you're not giving it fresh EBS volumes, are you? | 15:10 |
| med_ | Is Ubuntu Server certified to run under kvm? Where is that certification listed. (Yes, I know it works, but certification is more about testing/agreements/etc.) | 15:10 |
| med_ | http://www.ubuntu.com/certification/server/ doesn't list KVM. | 15:10 |
| addisonj | yep, new provisioned EBS volumes, not remounting old volumes | 15:11 |
| patdk-wk | med, kvm isn't an external piece, it's part of ubuntu | 15:12 |
| patdk-wk | you normally don't certify you work on yourself | 15:12 |
| === Skaag_ is now known as Skaag | ||
| uvirtbot | New bug: #1034428 in samba (main) "package samba 2:3.6.3-2ubuntu2.3 failed to install/upgrade: subprocess installed post-installation script returned error exit status 128 (dup-of: 1022385)" [Undecided,New] https://launchpad.net/bugs/1034428 | 15:20 |
| med_ | patdk-wk, nod, but it turns out, other distros also ship kvm. | 15:24 |
| patdk-wk | the, a certification for every single version of every single distro would be required | 15:32 |
| arrrghhh | hey all, i'm trying to scp some files between two ubuntu boxes. i get 'write failed: broken pipe' then 'lost connection'. immediately... it doesn't even start to transfer anything. | 15:50 |
| arrrghhh | i am able to copy stuff from my workstation to the server in question... but from ubuntu to ubuntu it's not workin. | 15:51 |
| arrrghhh | rsync errors out similarly - write failed broken pipe, connection unexpectedly closed | 15:59 |
| uvirtbot | New bug: #1034454 in php5 (main) "12.04 Dist-upgrade breaks PHP extensions > pdo_mysql.so symbols" [Undecided,New] https://launchpad.net/bugs/1034454 | 16:01 |
| arrrghhh | *crickets* | 16:03 |
| hallyn | stgraber: ok, seccomp committed to ubuntu:lxc. Did you have any other changes you wanted to stage? (I saw the lxc-start-ephemeral one) | 16:10 |
| stgraber | hallyn: nope, start-ephemeral was the only one | 16:11 |
| hallyn | ok | 16:11 |
| arrrghhh | i just disabled UFW on both the client and server | 16:15 |
| arrrghhh | still no dice | 16:15 |
| arrrghhh | i get asked for a password, so the network connectivity between the two boxes is there... | 16:15 |
| arrrghhh | anyone? | 16:29 |
| resno | funkyHat: ! | 16:31 |
| === lotia-away is now known as lotia | ||
| hallyn | arrrghhh: whacky thought, but check your .ssh/config, | 16:33 |
| === ksx4system_ is now known as ksx4system | ||
| hallyn | it sounds to me like you might have an entry doing ssh forwarding, and the forward isn't working | 16:33 |
| arrrghhh | hallyn, 1 sec thx | 16:34 |
| arrrghhh | hallyn, neither box has a .ssh/config | 16:36 |
| arrrghhh | ~/.ssh/config | 16:36 |
| arrrghhh | ? | 16:36 |
| hallyn | arrrghhh: weirdness. i'd dry 'ssh -vvv to get debug info | 16:37 |
| hallyn | yeah ~/.ssh/config would be it | 16:37 |
| arrrghhh | ok | 16:37 |
| arrrghhh | scp -vvv works? | 16:37 |
| arrrghhh | i'm trying to sftp | 16:37 |
| hallyn | oh, dunno | 16:37 |
| arrrghhh | ssh is disabled, this user should only have sftp rights. | 16:37 |
| hallyn | oh | 16:37 |
| hallyn | how exactly did you accomplish that? | 16:38 |
| hallyn | so did you say that 'sftp x user@host' also does not work? | 16:38 |
| hallyn | i'm not sure you can do sftp without being able to ssh... | 16:38 |
| arrrghhh | hallyn, 1 sec | 16:39 |
| arrrghhh | hallyn, http://solderintheveins.co.uk/2011/03/ubuntu-sftp-only-account-how-to/ | 16:39 |
| arrrghhh | hallyn, i am able to connect via WinSCP and copy files using that limited user account | 16:40 |
| arrrghhh | so i don't know why it would work from my workstation but not the server....:/ | 16:41 |
| hallyn | arrrghhh: perhaps you need to specify the ssh key to use (with -i)? failing that, this is too config-specific - cool, but i'd have to try it myself to see where it would fail lik ethat. | 16:43 |
| arrrghhh | no keys | 16:44 |
| arrrghhh | just password auth | 16:44 |
| arrrghhh | i'm trying to keep is simple lol | 16:44 |
| arrrghhh | i guess i can remove all the sftp-only stuff | 16:44 |
| arrrghhh | just odd that it works from my workstation... | 16:44 |
| hallyn | arrrghhh: yeah, it is weird... | 16:46 |
| ernetas | Hey guys. | 16:46 |
| ernetas | I'm running Ubuntu Server 12.04 on Amazon EC2 with 2 NICs. | 16:47 |
| ernetas | Currently the configuration is under DHCP. | 16:47 |
| === mcclurmc is now known as mcclurmc_away | ||
| ernetas | I'm using ip command to make both NICs work. But I can't automate it for system boot, because the default network adapter (the one which has gateway in "route" output) is always randomized from eth0 to eth1 on boot. | 16:49 |
| ernetas | How do I make eth0 or eth1 the default one? | 16:49 |
| ernetas | Anyone? | 16:51 |
| arrrghhh | ernetas, sorry that's way above my pay grade... this room is slow today as well. | 16:56 |
| arrrghhh | might want to post in ze forums | 16:56 |
| genii-around | ernetas: It's in the /etc/udev/rules.d/70-persistent-net.rules ...change in there eth0 to eth1 and change eth1 to eth0, after a reboot they will be swapped | 16:58 |
| guest1 | I want to implement Ubuntu Server 12.04 as a hardware firewall with 2 hard nics, 1 wireless. Nic#1==WAN, NIC#1==LAN, WirelessNic=LANWireless. I found tutorials for installing Squid Proxy (Web Cache), and a little about how to configure Ubuntu as a DHCP/DNS/Internet Server. My problem is that I have not found any info talking about how to implement Codel, the latest Active Que Management (AQM) system. Bufferbloat.net does have "Binary | 17:02 |
| guest1 | source: http://www.bufferbloat.net/projects/codel/wiki | 17:02 |
| guest1 | I found a tutorial for installing SQUID: https://help.ubuntu.com/community/Squid, yet can't find any info about if codel only uses its AQM for the local server's data, or if it can be used for the DHCP/DNS/Cache/Hardware Firewall portion as well? | 17:03 |
| arrrghhh | hallyn, the plot thickens. i undid everything, and now when i type my password in it just hangs on rsync... what have i done. | 17:07 |
| hallyn | strace -f -ooutout rsync <...> - see where rsync is hanging | 17:08 |
| hallyn | kees: have you done any code review of mosh.mit.edu? | 17:08 |
| arrrghhh | hallyn, what am i supposed to see from that strace command? i don't see anything different in the console. | 17:12 |
| arrrghhh | i type in my password, goes to the next line... and stalls out. | 17:13 |
| arrrghhh | let me see if straight up ssh works now | 17:13 |
| arrrghhh | it deso | 17:13 |
| arrrghhh | does* | 17:13 |
| arrrghhh | lol but it hangs on login. | 17:13 |
| rbasak | addisonj: is http://ubuntuforums.org/showpost.php?p=10907831&postcount=6 related to your problem? | 17:20 |
| addisonj | rbasak: this, this was it, just found it a minute ago: http://dev.bizo.com/2012/07/mdadm-device-or-resource-busy.html | 17:22 |
| addisonj | but yes, I do need to make that work for reboot as well, thanks for your help! | 17:23 |
| rbasak | addisonj: thanks - sounds like an mdadm or kernel bug | 17:23 |
| addisonj | yeah, not sure where the fault lies, looks like the underlying cause is udev scanning the devices deciding if it should do something with em, apparently, that is new behavior to 12.04 | 17:25 |
| rbasak | addisonj: would you mind searching for a bug report and filing one if there isn't one with your instructions to reproduce, please? | 17:26 |
| rbasak | addisonj: it would be good to get the information into one place in case other people hit the same thing | 17:26 |
| addisonj | yeah, sounds like a good idea, spare someone else the 12 hours to figure that one out | 17:27 |
| rbasak | and also to help gauge how common this problem is | 17:27 |
| rbasak | thanks! | 17:27 |
| arrrghhh | hallyn, it's "working" | 17:30 |
| arrrghhh | i had to blow out the user and recreate him | 17:30 |
| arrrghhh | and i haven't locked him down yet. not sure if that guide isn't accurate or what... | 17:30 |
| addisonj | rbasak: https://bugs.launchpad.net/ubuntu/+source/mdadm/+bug/1030354 one already exists | 17:33 |
| uvirtbot | Launchpad bug 1030354 in mdadm ""device or resource busy" on mdadm create" [Undecided,Confirmed] | 17:33 |
| addisonj | quite old, never triaged | 17:33 |
| addisonj | err nm, getting it confused with another bug I saw while searching | 17:34 |
| rbasak | addisonj: thank you! | 17:34 |
| rbasak | adac: ? | 17:35 |
| rbasak | sorry | 17:35 |
| rbasak | addisonj: ? | 17:35 |
| rbasak | addisonj: looks like the same problem to me | 17:35 |
| rbasak | addisonj: am I missing something? | 17:35 |
| addisonj | no, I mean it is, i said it was old, its not that old a report | 17:35 |
| addisonj | looks like it just needs triaged | 17:35 |
| addisonj | all is well | 17:36 |
| rbasak | Ah OK | 17:36 |
| rbasak | It's a foundations bug, so I don't think it's appropriate for me to mess with it. I can't see it realistically being bumped in priority unless it affects more people as a workaround is available. But at least it's there to track it. | 17:37 |
| rbasak | Thanks for looking! | 17:38 |
| === Aaton_off is now known as Aaton | ||
| roaksoax | Daviey: howdy!! any particular reason why delta was dropped for bug #1018001? | 18:04 |
| uvirtbot | Launchpad bug 1018001 in rabbitmq-server "Please sync rabbitmq-server 2.8.4 from Debian" [Medium,Fix released] https://launchpad.net/bugs/1018001 | 18:04 |
| Altoms | anyone here got RHCE ? | 18:18 |
| _ruben | might as well have a poll for MCSE and CCNA as well... | 18:18 |
| Altoms | how long did you study for rhce ? | 18:20 |
| Altoms | for a medioce sysadmin it says probably around 1month study time req | 18:20 |
| arrrghhh | Altoms, this is a support channel... | 18:21 |
| arrrghhh | #ubuntu-offtopic perhaps? | 18:21 |
| _ruben | a rh-oriented channel would be better i'd say ;) | 18:23 |
| arrrghhh | lol that too | 18:25 |
| roaksoax | jamespage: still around? | 18:32 |
| roaksoax | lynxman: ping | 18:32 |
| jamespage | roaksoax, yep | 18:53 |
| roaksoax | jamespage: any ideas if we still need rabbitmq-stop and rabbitmq-erlang-client | 18:56 |
| roaksoax | ? | 18:56 |
| jamespage | roaksoax, hmm | 18:56 |
| jamespage | I asked about this the other day - apparently not | 18:57 |
| roaksoax | jamespage: uhmmm how so? | 18:57 |
| jamespage | although it does remove 'package configured' options from rabbitmq IMHO | 18:57 |
| === cpg|away is now known as cpg | ||
| uvirtbot | New bug: #1000710 in nova (main) "Attaching volume during instance boot doesn't work" [Medium,Triaged] https://launchpad.net/bugs/1000710 | 19:31 |
| ChmEarl | wew - got sound playing using xmms2 and abraca with no X11 install | 19:54 |
| === bitmonk_ is now known as bitmonk | ||
| === viezerd- is now known as viezerd | ||
| halvors | Anyone knows a about a good domain hosting panel as the virtualadmin module for webmin? | 20:25 |
| === Skaag_ is now known as Skaag | ||
| === mcclurmc_away is now known as mcclurmc | ||
| === bitmonk_ is now known as bitmonk | ||
| arrrghhh | hey. i have samba setup JUST how i want it on one machine | 21:38 |
| arrrghhh | can i just copy the smb.conf to another? | 21:39 |
| arrrghhh | 12.04 desktop --> 12.04 server if it matters | 21:39 |
| hallyn | grrr. i think my bzr merge failed to merge debian/control | 21:51 |
| RoyK | arrrghhh: no difference between desktop and server in that matter | 22:00 |
| RoyK | arrrghhh: so just copy the smb.conf, and it should be fine, unless you have AD integration or similar setup | 22:01 |
| arrrghhh | RoyK, i do not | 22:07 |
| arrrghhh | i copied the smb.conf | 22:07 |
| arrrghhh | and did a service smbd restart | 22:07 |
| arrrghhh | and my shares aren't showing... | 22:08 |
| arrrghhh | bleh. i'm headed home, i'll bb when i get home. | 22:08 |
| arrrghhh | thx | 22:08 |
| === arrrghhh is now known as Arrrghhh | ||
| === Aaton is now known as Aaton_off | ||
| === Aaton_off is now known as Aaton | ||
| === Aaton is now known as Aaton_off | ||
| === Arrrghhh is now known as arrrghhh | ||
| arrrghhh | RoyK, there? | 22:38 |
| arrrghhh | anyone else know where samba config is? i thought /etc/smb.conf was it, evidently not. | 22:52 |
| Exopaladin | arrrghhh: Should be /etc/samba/smb.conf | 22:56 |
| arrrghhh | derp | 22:56 |
| arrrghhh | that's what i meant. | 22:56 |
| arrrghhh | Exopaladin, let me explain what i'm trying to achieve. | 22:56 |
| arrrghhh | i setup ubuntu desktop because i thought the proxy would prevent me from getting ubuntu-server onlilne. | 22:56 |
| arrrghhh | so i used the GUI 'sharing' tool in ubuntu desktop to setup these shares. | 22:56 |
| arrrghhh | now i'd like to move that config to the ubuntu-server setup - i assume just copy /etc/samba/smb.conf, right? well that file doesn't seem to contain any of the data required to setup the shares. | 22:57 |
| arrrghhh | i guess i can just go set it up manually. i was trying to be lazy, and i'm always curious when my laziness doesn't quite work out like i expect it to :P | 22:57 |
| rbasak | arrrghhh: GUI shares are usershares. They sit in /var/lib/samba/somewhere. Copying that might work but I don't think it's a good idea | 22:58 |
| funkyHat | resno: ! | 22:59 |
| rbasak | arrrghhh: use "net usershare" to manipulate usershares. There's a net manpage | 22:59 |
| arrrghhh | rbasak, ahhh. ok. i'll just go and set them up manually then. | 22:59 |
| arrrghhh | it would be better to configure the shares in smb.conf, right? | 22:59 |
| arrrghhh | so it's not user-specific | 23:00 |
| rbasak | I would | 23:00 |
| arrrghhh | k | 23:00 |
| arrrghhh | thx | 23:00 |
| === Aaton_off is now known as Aaton | ||
| arrrghhh | rbasak, you can *kinda* use those files as a guide for smb.conf | 23:44 |
| arrrghhh | but it certainly isn't something that can just drop into the smb.conf and work. | 23:44 |
| === james__ is now known as jmleddy | ||
| === arrrghhh is now known as Arrrghhh | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!