[00:24] <jrwren> the pvdisplay man page is wrong. *RAGE*
[00:24] <jrwren> can anyone confirm for me. pvdisplay -c third output field is size in 512B blocks NOT size in kB?
[00:25] <sarnold> I wonder if pvdisplay uses the same POSIXLY_CORRECT environment variable that df uses...
[00:26] <jrwren> either way, the man page lies.
[00:29] <bekks> jrwren: Thats correct. It's 512B units.
[00:29] <jrwren> thanks bekks
[00:29] <bekks> Tested on several systems here, not just Ubuntu.
[00:31] <bekks> jrwren: You mind to create a bug?
[00:31] <jrwren> i appreciate it greatly. thank you very much.
[00:31] <jrwren> I was going to LP just now.
[00:32] <bekks> jrwren: Could you provide the bug URL please, after bug creation? So I can subscribe.
[00:32] <thumper> does anyone know how to find out the underlying filesystem for a directory or device?
[00:32] <thumper> I know I can use 'df' to get from dir -> device
[00:33] <thumper> but I want to ask "is this directory on a btrfs volume?"
[00:35] <sarnold> thumper: I don't know of anything pre-built; your best bet might be to use df . to find the mountpoint and then look up the mountpoint in /proc/self/mounts. it doesn't sound fun..
[00:35]  * thumper shrugs
[00:35] <jrwren> bekks: https://bugs.launchpad.net/ubuntu/+source/lvm2/+filebug/dffddb12-950f-11e3-81c3-d485646cd9a4?
[00:35] <thumper> probably better the proc fs than fstab
[00:36] <thumper> /proc/self/mounts lists / as rootfs, not ext4
[00:37] <thumper> actually
[00:37] <thumper> just got pointed to "mount -l"
[00:37] <thumper> which does list type
[00:38] <thumper> do, "df ." gives me /dev/sda1
[00:38] <thumper> then "mount -l " I can parse (relatively easily)
[00:39] <thumper> I wonder how much the output of that has changed over versions
[00:39] <sarnold> 'rootfs'? p_P
[00:39] <thumper> precise is the same at least
[05:11] <Queops> Hey everyone. If I have one public facing VPS with static IP, hosting a couple sites on a couple domains which entries should I have on /etc/hosts besides the localhost one?
[05:15] <Queops> I know the hostname is usually there by default, associated with 127.0.0.1 but the hostname itself doesn't represent anything
[05:15] <Queops> No domain or anything, which I can only assume it serves as purpose to some services
[05:17] <Queops> or maybe I'm using the hostname wrong :)
[05:20] <shauno> Queops: I don't believe apache cares
[05:21] <Queops> I used to have a fully functional postfix and I remember putting my static IP there once, but I don't remember the reason for it anymore. I guess as a lone VPS with no other machines working with it the hostname pretty much doesn't matter right?
[05:22] <shauno> mail might, smtp is a fickle beast.  but for hosting sites you just need vhosts setup in your httpd
[05:23] <shauno> (I personally believe there's no such thing as a correct smtp setup, you're just trying to be broken in the same way as everyone else)
[05:23] <Queops> yeh smtp is a pain in the butt :) specially the ram required to scan for virus and spam, that's why I'm using outside services now
[05:23] <Queops> Thanks shauno
[06:48] <TraderJoes>  Do any of you run Cisco Connect (admin program for Cisco/Linksys routers)?   I wondered if it works in wine or if I'll have to use the web based administration. I checked the wine database and it's not listed.
[06:49] <TraderJoes> or if theres an ubuntu specific tool for such that might help. I didn't see any such thing in the cisco/linksys sites.
[07:05] <_root_> whic one is better is handling and prevention in general what I am asking is which one works better ???  a)Postfix + Amavis-new + Spamassassin + Clamav  OR  PostfixGreylisting????
[08:03] <z1haze> can someone please help me? is it possible to reinstall ubuntu via ssh? i dont have assistance from my server provider and the server is remote
[08:04] <z1haze> i messed up the filepermissions by messing up a chown command for the root folder
[08:05] <Repox> Hi. I realise that this must be both the wrong channel and that I'd might be using bullsh*t buzz words, but I'm hoping someone can point me in the right direction. I'd like to handle 'big data' in a business intelligence relation. I'm unsure what technologies for Ubuntu I'd be looking at. Currently I'm thinking of combining ElasticSearch and MySQL, but I'm unsure of this is the right combination.
[08:36] <mregg> Need help with linux permissions on samba shared folders
[12:12] <hxm> is webmin a good choice?
[12:14] <Adri2000> no
[12:14] <Adri2000> :)
[12:15] <hxm> why is the best option then
[12:16] <hxm> what*
[12:28] <disposab1e> i have a precise server with 3.8 kernel and i would like to set a specific mac address for a bond0 (active-backup). when i use "hwaddress ether x:x:x:x:x:x" line in /etc/network/interfaces, it has absolutely no effect. server still boots up with an address randomly picked from one of its slaves. what is the correct way of setting a mac?
[12:33] <ogra_> disposab1e, /etc/udev/rules.d/70-persistent-net.rules
[12:41] <disposab1e> ogra_: udev rule for a virtual interface?
[12:42] <ogra_> well, you asked where MACs are handled nowadys :)
[12:43] <Patrickdk> heh? you don't set mac in there
[12:43] <Patrickdk> it does set the nic name based on the mac there
[12:43] <disposab1e> ogra_: you got it the other way around. that rule sets name for mac. not mac for name
[13:38] <goiku> Hi, guys. Just a quick question, if I may. Am in the process of moving over from Redhat and have encountered this problem: http://hastebin.com/xowoyumudu.bash - I am sure it's a quick fix, but getting this server up needs to happen already and everything I have tried to far has not helped. Can I get a quick pointer, please? Thanks
[13:44] <jnor> I have two connections: root and user1. As root, if i type "passwd user1" and change the password to 123456, then reconnect as user1 and try sudo and it says password is wrong
[14:14] <jnor> this is so strange.. wtf
[15:50] <hxm> hi, I send emails with "mail -s" from a script with root, but I dont want to see the root@domain, can I define a different alias?
[15:52] <rbasak> hxm: I'm not sure if mail can do that. Have you tried "man mail" or looked further into its documentation? Another approach is to fix it in the MTA. Though usually, for root mail, that's not desirable.
[15:53] <hxm> yes I tried, and there is no information about the sender
[15:53] <rbasak> hxm: to override on a case-by-case basis, you could use "sendmail -oi -t" and write your headers directly in the script, including From:, Sender: etc as needed.
[15:53] <hxm> i know, but the script I made requires root privileges for create some /etc/apache2/sites-available/ files and then send an email
[15:54] <hxm> oh, um, i'll try that, thanks
[15:55] <rbasak> hxm: I suggest that you minimise your use of root privilege. Create /etc/sudoers.d/ rules and individual scripts for the specific actions you need, and run the rest of it (including mail sending) as non-root.
[15:56] <hxm> ok, thank you, I appreciate it
[15:56] <hxm> by the way, what about webmin or openPanel, are those tools recommended?
[15:56] <rbasak> webmin is apparently discouraged.
[15:56] <rbasak> !webmin
[15:56] <rbasak> No idea about openPanel.
[15:57] <rbasak> I would suggest that you can use what you like, but the set of people who can help you with things will drop dramatically. I would use what your support channels are familiar with. If this *is* your support channel, then probably best not to use those.
[15:57] <hxm> oh
[15:57] <hxm> right
[16:05] <yeats> hxm: webmin is probably fine, even though it's not officially supported - the way it does things *does* conflict with Debian/Ubuntu standards though
[16:07] <genii> I think supposed to be ebox/zentyal for webgui now
[16:10] <yeats> genii: cool - I'll check that out
[16:36] <rbasak> rharper: I'm catching up on uvtool. --meta-data is in and I'll upload it shortly. I refactored your code quite a bit though.
[16:36] <rbasak> Is there anything else I've forgotten about?
[16:36] <rbasak> Oh - I fixed the template.xml thing for acpi etc too.
[16:44] <DeltaHeavy> Hey, I have an nginx server running on Ubuntu 12.04, and as far as #nginx is concerned it's configured correctly. However when I try to run 'curl staging.decorgroupinc.com', I get 'curl: (7) couldn't connect to host'. Is port 80 probably being blocked? I flushed netfilter and changed all chain's default policies to ACCEPT. What else could it be?
[16:45] <DeltaHeavy> Sorry, accidently parted.
[16:47] <jpds> DeltaHeavy: Have you checked that nginx is correctly running with: "sudo netstat -lntp" ?
[16:47] <DeltaHeavy> jpds: It appears to not be...I see sshd and mysqld but yeah...nothing on port 80
[16:47] <TJ-> DeltaHeavy: If there are no netfilters rules blocking it, then the other thing to check is which IP addresses/interfaces nginx has bound to
[16:48] <jpds> DeltaHeavy: Check the logs. :)
[16:48] <DeltaHeavy> Awesome, I think this is it. Thanks!
[16:49] <DeltaHeavy> Nope, nvm. Nothing in here. Those are errors from before =/
[16:50] <TJ-> DeltaHeavy: You're sure the process is running? It hasn't silently failed?
[16:51] <DeltaHeavy> TJ-: Process is running
[16:51] <TJ-> DeltaHeavy: and bound to port 80 on the IP(s) you expect?
[16:52] <DeltaHeavy> TJ-: netstat -lntp     doesn't come up with anything about nginx
[16:52] <jpds> DeltaHeavy: But 'ps aux | grep nginx' shows it running?
[16:52] <TJ-> DeltaHeavy: which suggests the process isn't binding to any ports, if the process is running
[16:54] <rbasak> DeltaHeavy: are you running netstat as root? -p won't see nginx otherwise (but it should still see the port open)
[16:54] <DeltaHeavy> rbasak: I'm running it under 'sudo -s'
[16:55] <rbasak> And what about curl using the IP directly from the host itself?
[16:55] <DeltaHeavy> https://www.refheap.com/c0383a26d60db41edf1e24beb
[16:56] <DeltaHeavy> rbasak: Won't work, nginx isn't configured to return anything unless the actual server name is used
[16:56] <TJ-> DeltaHeavy: So use a manual telnet session
[16:57] <DeltaHeavy> TJ-: telnet staging.decorgroupinc.com 80
[16:57] <DeltaHeavy>         gives me 'session refused'
[16:57] <rbasak> DeltaHeavy: doesn't matter. Try it anyway. If nginx is working, it should give you something different from "couldn't connect to host" and that is useful.
[16:57] <DeltaHeavy> connection refused*
[16:57] <TJ-> DeltaHeavy: "telnet <ip address> 80"  ... "GET / HTTP1.1/" ...  "Host: <hostname>"  ... ""   ... ""
[16:57] <DeltaHeavy> rbasak: Same result
[16:57] <DeltaHeavy> Same with telnet
[16:57] <TJ-> DeltaHeavy: So, the process isn't listening, so start the nginx deamon manually and make sure it is providing verbose output
[16:58] <DeltaHeavy> Ok, trying to figure out how to start nginx with verbose
[16:59] <TJ-> DeltaHeavy: "nginx -t" will test the config
[16:59] <DeltaHeavy> Yeah, already did that a bunch of times, and just now again. All is fine. That's not starting nginx with verbose though.
[17:00] <rbasak> DeltaHeavy: with the server IP, from the server machine itself?
[17:00] <DeltaHeavy> rbasak: Yep
[17:00] <TJ-> DeltaHeavy: pastebin the config
[17:01] <rbasak> DeltaHeavy: then it's iptables or your nginx config.
[17:01] <DeltaHeavy> iptables is disabled
[17:01] <DeltaHeavy> Or all is set to ACCEPT ad a default policy with no rules
[17:03] <rbasak> I suggest that you stop making self-defeating assumptions here, like "won't work" or "iptables is disabled". Clearly one of your assumptions is wrong, else you wouldn't have a problem. Describe exactly what you're typing, and exactly the results you see on the screen - not your own interpretations of these things.
[17:04] <DeltaHeavy> TJ-: https://gist.github.com/anonymous/33b0ec963c479d22f629
[17:05] <DeltaHeavy> rbasak: It's not self-defeating to say that iptables has all default policies set to 'ACCEPT' with no rules, or that my nginx 'won't work' because it won't.
[17:05] <rbasak> Let's hope that somebody else can help you then.
[17:06] <DeltaHeavy> Have my netfilter config here too - https://gist.github.com/anonymous/53c76863ae9004f4dc56
[17:06] <DeltaHeavy> rbasak: I don't really understand why you're being so condecending...but whatever.
[17:06] <TJ-> DeltaHeavy: Does nginx have permissions to read /usr/share/nginx/ ?
[17:06] <jpds> It's not the firewall that's at fault here.
[17:07] <DeltaHeavy> TJ-: It didn't, I just changed that with 'chown -R webdev:webdev /usr/share/nginx/'
[17:07] <DeltaHeavy> Seems to not have done anything
[17:07]  * jpds isn't sure if that was a good idea.
[17:08] <rbasak> DeltaHeavy: http://www.chiark.greenend.org.uk/~sgtatham/bugs.html "If a programmer asks you for extra information, don't make it up! Somebody reported a bug to me once, and I asked him to try a command that I knew wouldn't work. The reason I asked him to try it was that I wanted to know which of two different error messages it would give. Knowing which error message came back would give a vital clue. But he didn't actually try it - he just mailed me 
[17:08] <rbasak> DeltaHeavy: it's very frustrating when I try to help, but you knock me down because you don't understand why I'm asking. Sorry.
[17:09] <jpds> DeltaHeavy: Well, looking at http://packages.ubuntu.com/precise/all/nginx-common/filelist - there's nothing interesting in that directory anyway.
[17:09] <DeltaHeavy> rbasak: Whatever
[17:09] <TJ-> DeltaHeavy: Did you look at "/var/log/nginx/error.log" for clues? How far does it get logging when it starts?
[17:09] <DeltaHeavy> jpds: It's my document root
[17:09] <DeltaHeavy> TJ-: I did, nothing is being written there.
[17:10] <DeltaHeavy> Last error is when I made an error in my conf a few hours ago.
[17:10] <jpds> Ew.
[17:10] <TJ-> DeltaHeavy: so, you need to enable verbose logging of nginx to gain detailed information
[17:10] <DeltaHeavy> jpds: I have multiple websites on this host, or plan to.
[17:10] <jpds> Who keeps their docroot in /usr? That's what /var/www/ and /srv are for.
[17:10] <DeltaHeavy> jpds: nginx does by default. Nothing wrong with it as far as I can tell
[17:10] <DeltaHeavy> /usr/share/nginx/ makes a lot of sense IMO
[17:11] <DeltaHeavy> TJ-: I was looking on how to get nginx to start with verbosity but I see no such option. Am I misunderstanding you?
[17:12] <jpds> Makes no sense to me.
[17:12] <jpds> Anyway. <opinion />
[17:14] <jpds> DeltaHeavy: https://rtcamp.com/tutorials/nginx/debugging/
[17:15] <TJ-> If the nginx process isn't shown to be listening on port 80, then something fundamental isn't configured
[17:17] <DeltaHeavy> I'll look into jpds tutorial
[17:17] <TJ-> DeltaHeavy: Have you symlinked your configuration files to "/etc/nginx/sites-enabled/" correctly  "ls -l /etc/nginx/sites-*/ | pastebinit"
[17:18] <DeltaHeavy> Should be in the pastebin I linked you under 'Shell' - https://gist.github.com/anonymous/33b0ec963c479d22f629
[17:19] <TJ-> DeltaHeavy: Nope, only "ls" not "ls -l"
[17:19] <jpds> DeltaHeavy: Have you tried removing this line? "listen [::]:80 default_server ipv6only=on;"
[17:19] <DeltaHeavy> TJ-: lrwxrwxrwx 1 root root  7 Feb 14 10:31 default -> default
[17:19] <DeltaHeavy> lrwxrwxrwx 1 root root 25 Feb 14 10:46 staging.decorgroupinc.com -> staging.decorgroupinc.com
[17:20] <jpds> DeltaHeavy: It's a link, onto itself?
[17:20] <TJ-> DeltaHeavy: There's your problem
[17:21] <DeltaHeavy> No idea why =S https://gist.github.com/anonymous/3dfced1c2943713f4c43
[17:21] <TJ-> DeltaHeavy: If you used the command correctly...!
[17:22] <DeltaHeavy> Fixed it
[17:22] <DeltaHeavy> Getting this now, going to go about fixing it - 'Restarting nginx: nginx: [emerg] could not build the server_names_hash, you should increase server_names_hash_bucket_size: 32'
[17:23] <DeltaHeavy> Fantastic...it all works.
[17:23] <DeltaHeavy> Thanks guys!
[17:28] <rharper> rbasak: cool.  I don't think so. IIRC you already merged the fixes for when specifying a specific bridge name
[17:52] <rbasak> hallyn, zul: shall I just upload http://paste.ubuntu.com/6914206/ then?
[17:52] <hallyn> rbasak: yup
[17:52] <hallyn> (i assume it didn't make it into zul's upload then?)
[17:52] <rbasak> OK, doing it now - thansk.
[17:52] <rbasak> It seems not.
[17:57] <rbasak> hallyn, zul: uploaded.
[20:09] <p3rsist> Hi guys. I want to run a command with user B env variables but with root permissions from a user A shell. How can I do it?
[20:10] <sarnold> p3rsist: how do those environment variables get set? replicate those steps..
[20:15] <p3rsist> Sarnold I want to run a command in user B environment but with root permissions from a user A shell. I know I can use sudo -i -u userB sudo -E <command>... I was wondering if there are other ways
[20:53] <DeltaHeavy> Is pastebinit a potential security threat?
[20:54] <TJ-> Not if you don't install it
[20:55] <sheptard> allowing anyone to access the machine including yourself is a potential security threat
[20:56] <DeltaHeavy> TJ-: Of course :p I'm wondering if I should uninstall it when I'm done with it, and reinstall it as I need it.
[20:56] <DeltaHeavy> sheptard: Yeah, I just want to minimize things.
[20:56] <sheptard> it's not a network service
[20:56] <sheptard> and it's not installed suid...
[20:57] <sheptard> and if you want to 'minimize things' run OpenBSD.
[20:57] <sheptard> no holes in a default install in at least a week.
[20:57] <TJ-> DeltaHeavy: Or you could do what I do: "sudo chmod o-x $(which pastebinit)"
[20:58] <DeltaHeavy> TJ-: I only have one user on the server anyway
[20:58] <TJ-> DeltaHeavy: But it's not different to having 'cat' available
[21:56] <Atw> Question, I am having issues with environment variables/$PATH, it started with me getting a different PATH value (checked with 'printenv') on other accounts (example, ssh deploy@1111, printenv, PATH=/usr/local/rbenv/shims:..., then 'su User', printenv, PATH=/usr/local/sbins:..., then even changin back to 'su deploy' I'm stuck on this other PATH). I seem to have patched the problem by
[21:56] <Atw> changing '/etc/environment' so now it seems like changing users gives me the path I want, BUT if I open a remote terminal from NAUTILUS while the server is mounted via ssh I still get sidelined on this other PATH I don't want? How can I stabliize my global PATH the right way?