[00:22] hitsujiTMO: (sorry had dinner) here ya go - http://pastebin.ubuntu.com/6280144/ [00:43] hmm, so conf.d is ignored [00:45] gdos: can you paste /etc/apache2/conf-available/dwww.conf and /etc/apache2/conf-available/dhelp.conf and the output of: ls -l /etc/apache2/conf-enabled [02:11] hitsujiTMO: (sorry had dinner) here ya go - http://pastebin.ubuntu.com/6280144/ [02:39] i'm thinking this broken package is causing my apache2 headaches - http://pastebin.ubuntu.com/6280144/ - how do i remove it? [03:37] Is it possible to get an unused device in /proc/mdstat? I've installed a fresh copy, installed mdadm and there is no raid and no unused device :( [05:46] sarnold: haven't written much, but I can answer if you have questions === freeflying_away is now known as freeflying [06:08] I'm trying to set up a basic samba print server and need to add the drivers. When I use the command "net rpc rights grant Everyone SePrintOperatorPrivilege -U root" I get: Could not connect to server 127.0.0.1 Connection failed: NT_STATUS_ACCESS_DENIED [07:12] Hi all How to setup a VPN server ? [07:25] babinlonston: https://help.ubuntu.com/community/OpenVPN [07:26] it's a good number of steps, but it's pretty straightforward [07:26] Does anyone know how a crontab can be executed 10 minutes late ? [07:27] jkitchen: configured there are some issue troubleshooting it i used this like to setup it https://www.digitalocean.com/community/articles/how-to-setup-your-own-vpn-with-pptp === freeflying is now known as freeflying_away [07:29] mac_nibblet: if the clock changes, cron tries to be smart and catch anything it missed [07:29] jkitchen: it has not changed [07:29] i scheduled something for 19:54 it was executed 20:06 [07:30] by what clock did you record that it ran at 20:06? [07:30] The records where updated in mysql at that time [07:30] on the same machine [07:30] so it's not a clock missmatch [07:31] is it possible that something held up the mysql update? [07:31] jkitchen: nopp [07:31] weird [07:31] you said the records were updated in mysql at that time, how were you checking this? [07:31] CLI [07:31] just up-entering on a query until it gave you the updated record? [07:32] what does syslog say regarding this cron job? [07:32] let me check [07:33] babinlonston: let me try to understand. You used https://www.digitalocean.com/community/articles/how-to-setup-your-own-vpn-with-pptp to set up a vpn and you're having trouble? [07:33] jkitchen: interesting.... [07:34] if that's the case, then I'm sorry, I've only ever worked with openvpn on the server end of things (and to be honest, I don't see why I would want to use anything different :) [07:34] jkitchen: the command is being run at 19:54:01 [07:34] mac_nibblet: it's possible there's some caching going on? you should also see when the cron started and when it stopped [07:35] jkitchen: http://caps.pmg.se/caps/e8122b.png [07:35] (by looking at the cron:session entries in /var/log/auth.log, that is) [07:36] oh [07:36] :p [07:36] why would you screenshot text? :) [07:36] uh, it's not a screenshot ^_^ [07:36] well it is, but gtkgrab [07:37] mac_nibblet: screenshot of text :) [07:37] anywho [07:37] look in auth.log to see when that cron started, and when it stopped [07:40] jkitchen: http://caps.pmg.se/caps/e34cd6.png [07:40] that does not make any sense [07:40] because the crontab took around 7 seconds to complete [07:42] I'm guessing caching on the mysql server. [07:42] or that your clocks are off by about 10 minutes :) [07:44] anywho, good luck! I must acquire sleep. === psivaa-afk is now known as psivaa === zz_Gurkenmaster is now known as Gurkenmaster [13:14] hello people, when you are in a hotel or in an airport, there is always a wifi hotspot that offers you to gain access to the internet, but only when you pay a fee, you will be given a username and password in order to gain access to the internet. Is there a similar solution in ubuntu? === freeflying_away is now known as freeflying [13:38] how do I get a remote server to use static dns addresses without ifdown ifup? [14:05] hi [14:06] i'm trying to install samba4 on the latest ubuntu server 3.10 [14:06] 13.10 i mean [14:06] but i get an error [14:08] whats the error buscon? [14:09] hitsujiTMO: just a second, maybe i've found the error [14:09] :) === freeflying is now known as freeflying_away [15:01] Is it possible to get an unused device in /proc/mdstat? I've installed a fresh copy, installed mdadm and there is no raid and no unused device :( [15:17] how easy is it to expand an encrypted partition? [15:18] i've managed to fill mine up so i'd like to expand it quite a bit [15:18] i've already expanded the VHDx [16:50] Whenever I try to run "sudo net rpc rights grant Everyone SePrintOperatorPrivilege" I get: "Could not connect to server 127.0.0.1 [16:50] Connection failed: NT_STATUS_ACCESS_DENIED [16:50] " [16:51] I'm trying to set up printer drivers with Samba [16:59] RoyK: cool, thanks :) I'm thinking of buying a handful of 3tb drives and wondering if you'd recommend a raidz1 or raidz2 configuration for commodity consumer hardware or mirrors of raidz1 or raidz1s of mirrors or ... there's just so many different ways to get there. :) [16:59] RoyK: I'm thinking it'd be nice to keep it under $1000 for the storage and it'd be nice to have enough space to not think about it for a few years :) [17:00] RoyK: I was thinking of using a 500-gig ssd for slog and l2arc, would that be overkill or underkill? :) [17:21] adam_g: fyi, I am doing raring cinder, keystone and nova security updates [17:22] adam_g: cinder is a no change rebuild for -security over what was there before, so you don't have to do anything for your proposed package [17:23] adam_g: nova is to fix bug #1212179 which is in your proposed update already [17:23] Launchpad bug 1212179 in nova/folsom "[OSSA 2013-024] nova should check the is_public of flavor when creating an instance" [Critical,Fix committed] https://launchpad.net/bugs/1212179 [17:23] adam_g: keystone is to fix bug #1179955 (part of your update) and bug #1202952 (doesn't seem to be part of your update) [17:24] Launchpad bug 1179955 in keystone/folsom "Disabling a tenant would not disable a user token" [High,Fix committed] https://launchpad.net/bugs/1179955 [17:24] Launchpad bug 1202952 in keystone/folsom "[OSSA 2013-025] PKI tokens are never revoked using memcache token backend (CVE-2013-4294)" [High,Fix committed] https://launchpad.net/bugs/1202952 [17:40] CVE-2013-4428 [17:40] ivoks: ** RESERVED ** This candidate has been reserved by an organization or individual that will use it when announcing a new security problem. When the candidate has been publicized, the details for this candidate will be provided. (http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2013-4428) [17:40] \o/ [17:40] adam_g: also, I'm fixing glance bug #1235378, which is part of your proposed update as well [17:40] Launchpad bug 1235378 in glance/folsom "[OSSA 2013-027] 'image_download' role in v2 causes traceback" [Undecided,Fix committed] https://launchpad.net/bugs/1235378 [17:48] ivoks: looks like uvirtbot queries something like launchpad or NVD or MITRE for cve information; the ubuntu UCT database or the debian CVE database may have newer information, as they do in this case. [17:48] ivoks: see e.g. http://people.canonical.com/~ubuntu-security/cve/2013/CVE-2013-4428.html or https://security-tracker.debian.org/tracker/CVE-2013-4428 [17:49] sarnold: ** RESERVED ** This candidate has been reserved by an organization or individual that will use it when announcing a new security problem. When the candidate has been publicized, the details for this candidate will be provided. (http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2013-4428) [17:49] sarnold: ** RESERVED ** This candidate has been reserved by an organization or individual that will use it when announcing a new security problem. When the candidate has been publicized, the details for this candidate will be provided. (http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2013-4428) [17:50] quiet you silly bot [17:53] heh [18:02] hi all. When using software raid, is it possibly to define a scheduler for the 'mdX' device or only on the underlying disk? [18:25] Hey, Propably a newbie question, but it was hard for me to find an answer to this online: Do apache2 logfiles really show the full traffic stats? Let's say I have a 1GB file, as soon as someone starts the download it shows 1GB traffic, right? If yes, is there a tool to measure the real traffic? [18:28] Daniel12: this looks useful: http://httpd.apache.org/docs/2.2/mod/mod_logio.html [18:28] (change the 2.2 in the url to reflect whatever version you have available, apache is good about making urls work as you expect :) [18:29] sarnold: great! Thank you! Will look into this. [18:31] Daniel12: thanks for asking the question, I've never looked into it before, but you made me wonder. hehe. :) [18:32] heh? [18:32] apache doesn't log how much was downloaded, till the download completes and is over [18:32] and that download size does not include headers by default [18:34] patdk-wk: oh. So an aborted download will show the (real) traffic it caused? [18:34] as long as real, doesn't mean, http headers, yes [18:34] patdk-wk: thanks! [18:36] patdk-wk: that's what made me wonder -- surely apache wouldn't wait to log until the request was over, what if you had a few hundred thousand connections drawing a byte per second or something? the log files would never report anything... so of course it'd log immediately.. [18:37] it never logs immediantly :) [18:37] .. but then I figured admins would need -some- way to realize that no one ever finishes downloading their 50megabyte background image, but how would they.... -> MUST GO FIND OUT :) hehe [18:37] no? [18:37] I have log entries happen a good day later [18:38] only way to know is by loading the status pages [18:40] "status pages"? [19:40] sarnold: ping === bitnumus is now known as Guest57978 [20:15] Hello all,I need some help,I was installed ubuntu server 13.10 and default installed LAMP pkg,when i was installed phpmyadmin,and login it ,The phpmyadmin tell me it need mcrypt,how to make it?thanks [20:16] apt-get install php5-mcrypt [20:17] the system auto installed it [20:17] but php.ini is auto config [20:18] i can't find the php5-mcrypt path [20:19] php.ini mcrypt.modes_dir= and mcrypt.algorithms_dir= I don't know the path [20:19] nsh001: did you install php5-mcrypt? [20:19] whoopsies, yes, you did [20:19] Mavken Thanks for you [20:20] @TheLordOfTime php5-mcrypt is auto installed [20:21] nsh001: if you're looking for the mcrypt configuration it gets installed to /etc/php5/conf.d/mcrypt.ini [20:21] nsh001, no prob...sorry...in another chat, having some network issues [20:21] (just an FYI) [20:21] nsh001: the .so for it is somewhere in /usr/lib/php5/ but i'd need to dig around in the 13.10 packages [20:22] nsh001: i'd say refer to the mcrypt.ini file it might already have those? [20:22] nsh001: also, known issue with php5, sometimes those plugins don't refresh after install, so doing a restart of your php5 instance might not hurt [20:22] i finding it [20:23] TheLordOfTime thanks for help and sorry for my english [20:23] your english is fine compared to some of the people i've worked with :) [20:23] let's me try [20:23] (not joking, it's actually pretty decent) [20:27] TheLordOfTime thanks for you,I can't find it.I need to sleep my time zone is +8 4:26am see you next time [20:27] Have a good time for all!bye [20:28] see ya [20:28] ooo too late :/ [20:28] oh well, was on the phone :/ [20:29] RoyK: pong :) [20:29] RoyK: a few more questions :D do you use de-dup ever? do you use copies=2 ever? do you pick-and-choose which datasets you use compression on? which compression algorithm? [20:30] I've used dedup, but without sufficient memory [20:30] dedup is *very* memory hungry [20:30] I haven't been using zfs in prod for a couple of years, but back then I used lzjb compression by default [20:31] some datasets gzip, some without [20:31] jz4 is probably what I'd used as default today [20:31] I haven't used copies=2 for anything except testing so far [20:32] how come no ~/.bash_profile in ubuntu? [20:32] sarnold: before choosing zfs, keep in mind that it's not as flexible as mdraid [20:32] you can't just add a drive to a raidzN set and have the raidzN grow with another drive [20:32] Beatstreet: .bashrc [20:33] same thing? [20:33] Beatstreet: it's run on login [20:33] RoyK: oh, you moved away from zfs? [20:33] so is .profile [20:34] sarnold: I switched jobs [20:34] RoyK: oh! [20:34] sarnold: had some 350TiB on zfs in the last job [20:34] RoyK: congratulations :) how far out of date am I? hehe [20:34] RoyK - I dont see a .profile [20:34] sarnold: only 15 months ;) [20:34] RoyK: hunh :) [20:34] sarnold: working for hioa.no now [20:35] Beatstreet: just add it to .bashrc [20:35] will work [20:35] add what? [20:35] RoyK: ah, I thought you used zfs for hioa .. (not that I ever could have recalled the name..) [20:35] Beatstreet: whatever you want it to run at login [20:36] sarnold: no, I used to work for nilu.no and they have a few zfs installs [20:36] oooo [20:37] RoyK: would mdraid let you just add another drive and have it grow? [20:37] sarnold: yes [20:37] RoyK: the ability to grow over time is one of the reasons why I've considered a handful of mirrored drives instead of raidz levels.. [20:37] obviously I need to re-read more about mdraid :) I just liked how simple zfs looked. [20:37] RoyK: what filesystems do you use now? [20:37] sarnold: you can add raidz or mirror VDEVs in zfs too [20:38] sarnold: but it won't rebalance the data across the drives, so writes won't be as quick as you'd want them (and reads, perhaps to a less extent) [20:38] sarnold: mostly ext4/xfs on a dell equallogic nas - *not* my choice [20:38] but then - it mostly work [20:39] RoyK: I've thought I owed xfs a look too... there's a general "eww" factor to nearly everything oracle-related, but zfs just looked so simple. :) hehe [20:40] RoyK: did you use different 'pool' styles on a single machine for different datasets? or is it best to just pick a level of redundancy and speed you're willing to live with and use that for everything? [20:40] sarnold: zfs isn't very simple, really, but if you plan it well, it's very good indeed [20:40] sarnold: but I chose not to use zfs for my home server - just using mdraid now [20:41] RoyK: did you rely much on the smb exports or nfs exports? [20:41] and some filesystem on top of that [20:41] mostly using samba/netatalk [20:41] RoyK: do you miss the resilvering? [20:41] no [20:41] but I miss real scrubbing ;) [20:41] checksumming would be nice [20:41] yeah, the checksumming feels pretty awesome [20:42] it's got a cost too, though [20:42] oh? [20:42] zfs will never be as fast as ext4 or xfs on mdraid [20:42] I figured CPUs were quick enough these days.. [20:42] it adds a tiny bit of latency [20:42] so effectively, it's slower [20:43] imho it's fast enought, though [20:43] usually, for a home server, the network's the bottleneck anyway :P [20:43] any of you running ubuntu server on digitalocean? [20:44] yeah, as I expect mine will be. cheapo-switches... [20:45] RoyK: all the guides suggest weekly scrubbing.. will that run a risk of driving hardware to premature extinction? [20:46] RoyK: would you recommend an ssd slog or just let the slog happen on the spindles? [20:52] sarnold: not really a danger [20:53] sarnold: slog or not (or l2arc or not) depends on i/o use pattern [20:53] for general purpose (your movies, porn, pics blah) slog/l2arc won't be worth much [20:53] for virtualization slog should be considered [20:54] et cetera et cetera [20:57] RoyK: aha, I was thinking a giant pile of the thing would go towards having a local ubuntu archive for the times I want to grep -r the world :) [20:58] RoyK: and I figured l2arc wouldn't help there much, aside from unpack-then-grep, but that might happen better on local machine memory.. [20:58] RoyK: did you do much ARC tuning to make more memory available to applications? [20:59] RoyK: does the arc duplicate efforts of the linux caching mechanisms? [21:00] RoyK: ooh ooh is the atime handling affected by the linux relatime mount option? [21:01] sarnold: if you want to grep -r whatever's there, perhaps solr would be better? ;) [21:02] sarnold: not really any ARC tuning - I was running this on openindiana (now not very active, illomos-based opensolaris fork) [21:02] RoyK: maybe? :) as it is I don't have a copy of it anyhow.. the debian codesearch is awesome though, it covers 90% of my questions easily... [21:02] RoyK: oh, I thought you used the zfs-on-linux... boy learning a lot today :) hehe [21:03] sarnold: otherwise - try #zfsonlinux - I've been running zfs on linux too, but not in large scale production [21:03] (not production at all) [21:05] RoyK: thanks for answering my pile of questions! :D [21:05] :) === Gurkenmaster is now known as zz_Gurkenmaster === Guest37141 is now known as Monotoko [22:43] Does anyone know what all is needed to create a classroom style domain controller? Essentially I would like to have a group of desktops where a user can sit down at any computer and login and have access to all their files and settings. Unlike a thin client I would like to leverage the power of the desktop not the server and remote connection needs to be available as well. I think this can be done with samba/ldap but I w [22:43] ould also like to pxe boot the os. So that I can just plug in a new computer and boot from ethernet and its ready to go. Remote connections should not be pxe booted however but something similar to rdp would be nice. thank you [22:44] also single sign on to all the services available to that user account would be needed as well. So all the user has to do is signin to ubuntu and thats it [22:47] philip550c: I don't know if anything is ready-made for your situation.. investigate LTSP and MAAS. MAAS is mostly intended for server uses, but might be applicable to your case.. [22:49] im messing with MAAS right now but it seems to be more about provisioning, I see that I can make it provision desktops but it doesnt act as a domain controller. Thanks for the suggestion, Ill look into ltsp more, I thought that was for thin clients but I might not have understood that correctly [22:49] no, you'd need to go to the effort of configuring samba or whatever for that end of things yourself [22:50] if you combine juju with maas, you just might be able to do something like "juju deploy -n 20 terminals" -- the end result _might_ be awesome :D but it'd probably be a fair amount of work on your part to get there. [22:52] im experimenting with maas/juju and servers, ill see what I can do with it for desktops [22:52] cool! if you don't mind, please report back, this sounds fun. :) [22:54] yeah, I will. Are you always on here? this is actually my first time messing with irc. Are some of you on here all day everyday? [22:55] i run my own xmpp server but i see that irc has a different use case now [22:55] philip550c: I'd wager 80-90% of the people here just leave an irc client connected all the time. /lastlog -hilight is just too useful.. [22:56] sarnold: or they're on bouncers, i know for a fact that's what i do :P [22:57] ok how does /lastlog -hilight work? I just entered it and nothing happened as far as I can tell. Sorry irc newb [22:58] oh i see nm [22:58] philip550c: hrm, in most clients it'll select just the bits of conversation aimed at you, referencing you, or mentioning any of the words that you might have put highlights / alerts on... [22:59] (and /lastlog -clear ought to get rid of it) [22:59] ok thanks [22:59] TheLordOfTime: hunh, I figured you for a screen+irssi user :) [22:59] heh [22:59] sarnold: need the bouncer so i can access IRC from the phone [22:59] so is it proper etiquette to address people by username before speaking? [22:59] TheLordOfTime: ah, makes sense... [22:59] can't ssh+screen+irssi from that very well ;) [23:00] sarnold: even then, i have the ZNC actively save the logfiles so I never actually miss anything, even if my buffer playback doesn't show everything [23:00] philip550c: it's more important the busier a channel is. when it's mostly quiet it doesn't matter much, but if it is quiet enough that your other party has wandered off, then the nicknames are useful again. :) hehe [23:00] like here, it's not super useful to always highlight someone's name, but when it's a tad busier it never hurts to identify who you're tlaking to :) [23:01] exactly [23:01] i usually always identify who at the first line in a string of lines. [23:01] now, #ubuntu, on the other hand, you prepend your destination's name on everything because that moves that much more rapidly [23:01] thanks guys [23:02] my project is actually more in depth than what I explained before but if I can pull it all off ill report back to here so you can hear about it [23:03] philip550c: cool! thanks. :D [23:34] jamespage: please can you subscribe ~ubuntu-server to php-json? [23:35] rbasak, [23:35] yes [23:36] rbasak, done [23:39] Thank you! [23:45] hey, i just installed userver on a vm and forgot my username to login with, what are my options? [23:45] oscalation: if you can boot with "single" on the kernel command line (grub or whatever..) you can get a root prompt pretty easy, that you can then use to inspect the system to find username, authorized_keys, password hashes, etc [23:50] sarnold, luckely i got it by guessing [23:50] thank you for the suggestion [23:54] what is the easiest way to get php mail working without having a full blown mail server? [23:54] it is used for sending out e-mail confirms and password resets [23:56] MavKen: I think PHP PEAR offers an inprocess mail() mechanism that doesn't rely on local host MTA support... [23:56] ok thanks, will check it out