[00:00] lol ofc [00:00] is it possible that many pcs access the same nfs share withou;side effects? [00:00] but those ntp servers are accurate. [00:00] aliverius, that's sort of the entire point of nfs [00:01] great cause i need to share data between a kvm host and guest [00:09] qman__: I just dont understand why ntpd listen at UDP port 123 [00:10] to get time (duh)? [00:10] I dont to run it as a server [00:10] dont want === smoser` is now known as smoser [00:11] ntpd serves time on that port [00:11] if you don't want others accessing it, block it in your firewall [00:11] but i dont want to serve [00:11] I did. [00:12] qman__: so I can use my machine as server even that i get that time from other sv? [00:12] that's how ntp works [00:12] you get time from class 2, 3, 4 servers depending on your configuration [00:13] your ntpd then becomes one class above that [00:13] nice :) [00:14] but there should be an option to listen [00:14] and each link in the daisy chain is one class higher, until 16 [00:14] :x [00:14] 16 is considered totally unreliable [00:14] to close listen port [00:14] i meant [00:14] yeh [00:15] most people who run ntpd run it to provide network time [00:15] as ntpdate is good enough for 99% [00:15] i run it to get time [00:15] :p [00:15] the other 1% can block it in the firewall [00:15] k [00:18] hmm, the default ntp config doesn't serve time [00:19] IIRC, it doesn't serve time, but it still listens on the port [00:19] and simply rejects any requests [00:19] ya, cause port 123 is much more than just receive/send time [00:19] it's also the control interface and stuff [00:20] if your paranoid about it listening on port 123, use ntpdate [00:20] Patrickdk: :D [00:20] just make sure you script it properly [00:20] no im paranoid [00:20] i just block 123 port [00:20] * Patrickdk runs a pool server :) [00:21] Patrickdk: if default ntp config doesnt act as server there's not secutiry risk then [00:21] I dunno what happened, but been getting flooded with ntp requests from china for the last month [00:21] there is always a security risk [00:21] and it does act as a server [00:21] ddos [00:21] just cause it doesn't respond [00:21] doesn't mean someone won't find a vaunerability [00:21] ok [00:22] it's just the likelyhood is even less [00:22] thanks for clarifying my doubts [00:23] * Patrickdk has never been scared though of it [00:23] config shows [00:23] # By default, exchange time with everybody, but don't allow configuration. === justin__ is now known as jMyles [00:24] hmm? [00:24] restrict -4 default kod notrap nomodify nopeer noquery [00:24] ./etc/ntp.conf [00:24] noquery makes it not let other people request time [00:24] restrict 127.0.0.1 [00:24] only at localhsot right [00:24] allow everything on localhost [00:24] oh [00:25] # Local users may interrogate the ntp server more closely. [00:25] and how I do that? [00:25] http://support.ntp.org/bin/view/Support/AccessRestrictions [00:25] you want to allow local access? [00:25] no just get time [00:25] restrict 192.168.0.0/16 default nomodify notrap nopeer [00:26] Patrickdk: ok it means that i can access it from subnet [00:26] oh that isn't right [00:26] # If you want to provide time to your local subnet, change the next line. # (Again, the address is an example only.) #broadcast 192.168.123.255 [00:26] restrict 192.168.0.0 mask 255.255.0.0 notrap nopeer nomodify [00:26] is what I'm using [00:27] 0.0? [00:27] hmm, ya [00:28] should I disable restrict 127.0.0.1 restrict ::1 [00:28] ? [00:28] with # [00:28] why? [00:28] i dont want anyone to access it [00:28] you don't trust your own machine it's running on? [00:28] :) [00:28] if you do, ntpq and stuff won't work [00:28] ah k [00:29] you could always password restrict it [00:29] Patrickdk: restrict -4 default kod [00:30] -4 is for ipv4? [00:30] ya [00:30] and default and kod? [00:30] kiss of death [00:30] it won't do really anything, without the limit option [00:30] not sure about default :) [00:30] never really use it myself [00:30] default is to listen all adresses maybe [00:31] no [00:31] OH [00:31] By default, exchange time with everybody, but don't allow configuration. [00:31] so if i want to configure need to remove default [00:31] Patrickdk: so i just need restrict 192.168.0.0 mask 255.255.0.0 notrap nopeer nomodify [00:32] and to open 123 port in router? [00:32] if your subnet uses 192.160.x.x [00:32] and soft firewall [00:32] 192.168.x.x [00:32] yea [00:32] got it [00:33] Patrickdk: and i only want the router [00:33] I put its IP? [00:33] Hey how do you select multiple software packages to install when installing ubu server? [00:33] restrict 192.168.1.69 mask 255.255.0.0 notrap nopeer nomodify [00:34] for example [00:34] ? [00:36] why the odd restrict ip? [00:37] Patrickdk: if router is 192.168.0.1 [00:37] I use that IP? [00:38] why did u use 0.0 as example? [00:38] is it like wildcard? [00:38] cause anything that is a 0 in the netmask is pointness to define in the ip [00:38] and lots of programs will bomb out if you do [00:38] bomb out? [00:38] cause mine was *correct* and yours is, odd :) [00:39] crash, syntax error, fail, ... [00:39] Patrickdk: i just want the router to access the sv [00:39] so i just put router ip [00:39] is that correct? [00:39] NO [00:39] or i use my local ip [00:39] ? [00:40] NO [00:40] NO [00:40] learn how ip subnets work [00:40] tell me teacher [00:40] google, ip subnet [00:40] :D [00:41] * old_keyboard reading in is native language [00:43] Class C [00:44] B in that case [00:44] ignore anything talking about classes [00:44] B 10 128.0.0.1 191.255.255.254 255.255.0.0 /16 [00:45] it's kinda advanced [00:48] Patrickdk: I found this config on a site http://forums13.itrc.hp.com/service/forums/questionanswer.do?admit=109447627+1301185010084+28353475&threadId=1357094 [00:48] restrict default ignore restrict 127.0.0.1 restrict your.ntp.server [00:51] and in here http://www.debianadmin.com/ntp-server-and-client-configuration-in-debian.html they say to use restrict 192.168.1.0 mask 255.255.255.0 nomodify notrap [00:52] man, you really can't be bothered to learn what anything does? [00:52] just blindly follow whatever anything tell you to do? [00:53] i am learning [00:53] you say something other ppl say something different [00:53] :) [00:53] I told you exactly what you wanted [00:54] those are generic [00:54] and the first one has nothing to do with anything you want [00:54] ok [00:54] restrict 192.168.0.0 mask 255.255.0.0 notrap nopeer nomodify [00:54] you could remove the nopeer if you want [00:54] gonna get hacked :( [00:55] why? [00:55] what is the nopeer [00:55] option [00:55] no connections? [00:55] you shouldn't allow anyone using a 192.168.x.x network range on your network [00:55] it's for when two ntp servers agree to talk to each other [00:55] they can use each other as clients or servers [00:55] so i remove that I could mess it up [00:56] no [00:56] hm [00:56] ok [00:56] it means another ntp server on your network could become the source of time [00:56] and if that person is screwing with you [00:56] could theoredically cause your server to get bad time [00:57] nice [00:58] Patrickdk: and restrict 127.0.0.1 is for unlimited access [00:58] ? [02:00] ya [02:00] see i've learned something new today [02:01] then sudo service ntp restart [02:02] Patrickdk: what about the -4 and -6 should I remove it? [02:02] doubt it matters, dunno [02:02] doesnt make a difference since i only use ipv4 [02:02] I haven't bothered [02:02] k [02:03] the ORDER in the file matters a LOT though [02:03] k [02:03] i comment out those 2 lines and add restrict 192.168.0.0 mask 255.255.0.0 notrap nomodify [02:04] so now your open to everyone? [02:04] not yet [02:04] well, if you commented out the restrict default lines [02:04] but added restrict 192.168.0.0 mask 255.255.0.0 notrap nomodify [02:04] just add the restrict 192.168.0.0 line AFTER those [02:04] #restrict -4 default kod notrap nomodify nopeer noquery #restrict -6 default kod notrap nomodify nopeer noquery [02:05] ya, so now since no one is restricted by default :) [02:06] gonna get hax00red [02:07] or maybe it doesn't [02:07] but that never opened it up for me [02:07] probably have lots of other things going on with my config, cause it's a few hundred lines long [02:08] so im open to my subnet [02:08] and the rest of the world if i forward port in router [02:08] ? [02:12] all the options here http://www.eecis.udel.edu/~mills/ntp/html/accopt.html [02:35] hello people. I need assistance with changing the IP addess of my ubuntu server, I have obtained a seet of ip's from my isp and have them running on the network, I just checked the ip of the server and it's still runing the old one, how to I change it? [02:36] that is a google-able question [02:36] the current ip is 192.168.0.8 and I need to change it to *.*.*.139 (the * rep the puplic ip) [02:36] * patdk-lap still doesn't get why people bother hiding ip addresses [02:36] they are *public* knowledge, and of a small enough space to *guess* [02:37] so I can't help on this item in the ubuntu-server HELP channel? [02:37] use a VPN [02:37] technically, the answer is simple, edit /etc/network/interfaces [02:37] last time i sent a public ip here I ended up with DOS attacks [02:37] but the question you didn't ask, and can't do much about really is [02:37] you have to change the config of all your programs also [02:38] that I can get help from the program website, as it's one master program that changes everything, just wont do the eth ip [02:38] they overloaded your internet connection? [02:39] yes [02:40] ok it seems that the ip address is obtained automaticly, how do I do a release renew in ubuntu-serer? [02:40] the interface is eth1 [02:41] you run a server via dhcp? [02:41] I set static via the router. but I need to get the ip on the server first [02:41] it hasn't failed me yet [02:42] ifconfig -a will show you its address [02:42] does the router have a public ip? [02:42] and it's going route that public ip into the same network your mixing private ip? [02:42] and you have no issue with that? [02:43] i don't know if he realizes the issue.... [02:43] all devs on my network will be running the public ip addresses [02:43] I have firewalls on all systems [02:44] *facepalms hard* [02:44] are firewalls not good enough [02:44] no, you want to subnet public from private [02:45] how do I do that? [02:45] what type of routers do you have? [02:45] netgear [02:45] model numbers :) [02:45] dg834g [02:45] but it will be changing [02:45] serial numbers... [02:46] u need that for? [02:46] oh, home system crap [02:46] you cant get the info on the model? [02:46] nvm.... I didn't realize that was model... [02:46] yea... your need more commerical system... I'm using a home system netgear... so wish i could stick a different firmware on it [02:46] I can get the serial number if you need it [02:47] no need [02:47] ok recomend? [02:49] something you can flash pfsense onto... [02:52] yeah I can't recommend somethign... I'm only a security student atm... [02:52] oh ok. sorry guys I am new to this type of networking [02:53] get shoved into a sys admin role? [02:54] not really, I am starting a hosting company, so learning on the fly BEFORE I get customers [02:55] ooohhh... erm... whats your computer knowledge? [02:55] shaggy2: how are you connected to your isp? dedicated line? [02:55] 2 decicated lines [02:55] ok [02:56] Just configure each machine on your network with a different ip from the range you have been assigned [02:56] I have 2 ADSL2+ connections atm, first in line when fiber comes into my town (if your from australia then you'll know what I am on about if not the lookup the Australian NBN) [02:57] haha i live in Arkansas, we just retired the tin can and string [02:57] I assume you have a router on each line [02:57] so on a system that I don't want on the public network, I just set for eg 192.168.0.3 on the same subnet as the public systems? [02:57] i live in south dakota... we still have pony express [02:57] router---->main box----->switch---->rest of the boxes [02:58] the main box can do your firewalls and nat [02:58] *cough* pfsense [02:58] hehe ok [02:58] i will be quiet now [02:59] oh hey... keep going :_ [02:59] I was just suggesting using pfsense [02:59] both modem are going into a linux box that goes to a router all in the main room, then in my server room trunk cable (5 line) into a switch then into 3 systems and a server [03:00] sounds like you know what your doing, I will be quiet [03:00] everything runs from the main modem, the 2nd only kicks on on reduncy [03:01] make sure to test that it'll kick in aka unplug the main, so you don't discover its not working [03:01] already done, I do a weekly test [03:02] the router after the linux box in the dg834g [03:02] I run my network as i showed above [03:02] i'll be quite now too... look up some linux sys admin/network admin books if you need to look into subnetting and stuff [03:02] my main box has the public ip the rest are private [03:02] main box does firewall, nat, web and mail but mostly just tosses packets around [03:02] well thats how I had it before, but had to make changes so I can run a DNS server [03:02] mine is schools network -> crapy router -> switch -> boxes [03:02] that is what got me unstuck [03:02] just put the dns server on the main box also [03:03] I could view all domians on my network, but the people that was outside couldn't get to see them cause it was directing to local address's [03:03] 123/tcp closed ntp [03:04] shouldn't it respond as open [03:04] ? [03:04] if you are hosting other folks then i would assume you would have those boxes with public ip's also [03:04] so that they can get to them [03:04] well I only had 1 pub ip untill today [03:04] still the configuration i gave you will do that too [03:04] nod [03:05] so anyways back to changing the ip on the ubuntu server [03:05] I can manualy set it [03:05] sure [03:05] just need someone to tell me how [03:05] this is much more of an issue than that [03:06] you need to resign your whole network [03:06] http://www.howtogeek.com/howto/ubuntu/change-ubuntu-server-from-dhcp-to-a-static-ip-address/ [03:06] and I don't think that *router* you have will let you do it [03:06] well this is the only thing that "NEEDS Static" [03:07] it's allready done, I was on the phone with netgear support for 3 hours having them telling me that it wont do it, and then after 3 hours saying do this and this and this and now it works [03:07] dont forget to make a backup before you edit the file [03:07] paid them $89.95 for them to tell me to buy a new mobem, only to then turn around after abusing them then it now works [03:08] my advice is worth what you paid for it [03:08] hehe [03:08] take a look at that artice, it should fix you up [03:08] sweet thank you, but how do I make a backup of that file? [03:09] hey how can i see if my raid is mounted or not? [03:09] cp filename filename.original [03:09] also make sure you look at the comment from [03:10] Nickname007 and do that too [03:10] could also set static dhcp per mac [03:10] otherwise you wont have any name resolution [03:10] hope that helps [03:11] * patdk-lap still wonders how using a static ip inside the nat part of that router will work [03:16] shaggy2 you hosting webservers? [03:19] shaggy2: if your changes dont work you can always cp filename.original filename and be back where you started [03:23] formatting my server's 1.5tb raid... zeroing... omg... so much time... [04:58] What does the automatic generation of /etc/ssl/private/dovecot.pem? [05:10] hi [05:11] Do you know a good website to search mother board model? [05:17] New bug: #683591 in sensors-applet (universe) "sensors-applet randomly picks up duplicates" [Undecided,Invalid] https://launchpad.net/bugs/683591 [05:17] hm [05:35] tonyyarusso: The postinst maintainer script in the dovecot-common package [05:37] jmarsden: Ah. Is there an easy way to re-run it (to change the hostname), or do I have to walk through things manually? (The primary system hostname is johnmarty.org, but I want Dovecot's certificate for mail.johnmarty.org) [05:37] I'm not sure. Look at /var/lib/dpkg/info/.dovecot-common.postinst and maybe you can copy it somewhere, edit that script to do just the certificate generation, and run your modified copy? [05:38] seems reasonable [05:38] Make that /var/lib/dpkg/info/dovecot-common.postinst [06:33] jmarsden: I ended up having to do a bit more than that, but got it now. [07:42] New bug: #743484 in nut (main) "libupsclient.pc contains unresolved symbols" [High,Confirmed] https://launchpad.net/bugs/743484 [09:46] hello [09:46] anyone know anything about frame routes. or where I can get help [10:08] Atomhunter: are you there? [10:09] shaggy2: how did changing to a fixed ip go? [10:10] not goot [10:10] all ips are done [10:10] why? [10:10] but my router dun support framed routes [10:10] fucken netgear [10:10] ah [10:10] lookin for a decent router that will do it, was pointed at cisco [10:11] it's a shame [10:11] the netgear dg834g has been in use for about 4 years now across heaps of different isp's, and this is the first time it hasn't been able to do something [10:12] sorry to hear that [10:12] it is gunna be a sad day to see it go [10:12] never missed a beat [10:13] no matter how good it was, it's no good once it fails to keep up. :) [10:13] had a problem connecting once my ISP said I need a new modem, I said no check everything else first, thats when they found out the my line was pulled from the ADSL socket in the exchange [10:14] how dodgy is that [10:16] so can anyone recomend a router for me [10:17] cant your isp recommed what you should be using? [10:18] shaggy2: so, you'r looking at an adsl modem + router combined? [10:18] yeah [10:18] I've never liked that combo, rather have a bridged adsl modem than wont messup + standalone router. [10:19] dun have to be, I do have a dlink adsl2+ that I can use [10:19] currently I use a bridged zyxel + apple's time capsule, howeverm the timecapsule is not that great. [10:20] the bridged zyxel on the other hand doesn't fuck up. it just does it's thing. :) [10:20] cool [10:21] I just need something that supports framed routes... I have never played with this before so it's all new to me [10:21] (this is the cheapest basic version of the zyxel adsl modem, just one port in, one out, and in bridged mode, basically all it's features are disabled and can't break) [10:22] but I know nothing about framed routes, so can't recommend anything there. :/ [10:22] lol same with my ISP who sell them [10:24] afk. gott go get my house built. :) [10:24] lol [10:24] seriously. [10:24] it ain't gonna build itself. :) [10:31] lol [10:38] hey guys, i am having a bit of a problem with apache displaying php errors in browser, I have set my my php.ini varialbles to display errors [10:56] what's the problem kickar ? [10:56] lenios, hi, I can't get php error message displayed in browser [10:56] if there is an php error i get either plain white screen either [10:56] "The page you are trying to view cannot be shown because it uses an invalid or unsupported form of compression." [10:59] kickar what browser [11:00] KMFrog, every well know browser [11:00] IE, Mozilla, CHrome [11:00] they all say that exact thing? [11:00] yes error 330 Error content decoding failed [11:01] on my gentoo machine displayes mysql conn. error [11:01] i believe it is php/apache related issue [11:01] can you wget / telnet it to see the raw output? [11:02] KMFrog, yes I can [11:02] it says "Warning: mysql_connect(): Unknown MySQL server host 'http' (1) in /var/www/qa-db.php on line 50" [11:02] you doing your own compression? [11:03] in php [11:03] i have tried to install a cms [11:04] have you seen that before [11:04] ah, ok, CMS is maybe trying to do its own gzip compression and its all failing because of the error [11:04] you will need to remove the http:// from your config i would guess by that message [11:04] if its http://localhost it needs to be localhost [11:04] for the db address [11:05] KMFrog, i am not trying to fix this particular issue, I am trying to get apacge to show [11:05] the error in browser [11:05] the CMS probably says its a compressed page, but then the error is in plaintext so the browser dies [11:07] it is the same on the test script that i wrote [11:08] KMFrog, do you have any idea? [11:09] fixing the mysql error [11:09] I mean at all [11:09] check http://otgovorimi.com [11:09] this is the ubuntu server [11:10] this is the gentoo machine [11:10] http://dhwebservices.com/web/ [11:11] it is the same script [11:12] oh that's odd. your first url has something in the first four bytes which kills my browser [11:13] shauno, they are the same script [11:13] the first is ubuntu server [11:13] the second one is gento [11:13] gentoo* [11:13] the good page starts ef bb bf .. and then into the html. the bad one has an extra byte, ef bb bf 0a ... [11:14] shauno, maybe I have miskaten something in php / apache configuration [11:15] KMFrog, welcome back [11:15] srry, client died -_- [11:15] did you get my last 2 lines ? [11:15] ubuntu server: http://otgovorimi.com [11:15] genoo server: http://dhwebservices.com/web/ [11:15] the same script [11:15] the same error [11:15] oh, 0a's just a linefeed. I've no idea why that's making any difference [11:16] maybe apache conf? [11:18] kickar I would say its a case of the CMS not handling errors very well, if you fix the error it will probably work [11:18] what cms is it [11:18] KMFrog, the same with WP, Joomla, and so on [11:18] it is some free q&a cms [11:19] just for testing [11:19] my goal is not to fix this particular error, but to fix apache [11:19] apache is probably ok [11:19] it looks like php issues [11:21] maybe i have missed a module [11:21] kickar, try to do a test page in php and see if errors are displayed [11:21] what php modules I should install to have a good production enviroument [11:22] good test page: [11:22] anyname.php [11:22] error page: [11:23] i might have a syntax error besides the missing quote [11:24] http://otgovorimi.com/info.php [11:24] works fine [11:25] http://otgovorimi.com/error.php [11:25] displays the error fine now [11:25] makes me think again that CMS is not dealing with errors very well [11:25] cms related, yes [11:26] KMFrog, but what about the other url? [11:26] the are both the same script on different systems set the same way [11:26] kickar if error.php is .. then its normal it wont work :p [11:27] i mean about [11:27] for example i have set up a wp here [11:27] can you give phpinfo(); for both servers? [11:27] just a second [11:29] kickar: have you asked the guys in #wordpress as I suggested [11:29] kickar: you seem to be repeating the same question as in #ubuntu [11:29] ikonia, just looking for an answer [11:29] i have asked in #php as well [11:29] kickar: did you ask the guys in #wordpress who I told you HAD the answer and explained it to me [11:30] ikonia, their solution is adding a ling in .htaccess to show error, but this is not going to fix my entire server [11:30] http://dhwebservices.com/info.php [11:30] lenios, here is the gentoo one [11:30] maybe it's just not your server [11:30] was the .htaccess line to disable gzip? [11:31] lenios: it's not a server issue [11:31] lenios, i have had the same issue with WHMCS, Joomla!, Wordpress etc. [11:31] my RHEL and Fedora machines to it too, it's how PHP clasifies errors [11:31] the guys in ##php should be able to explain this, and the guys in #wordpress explained this to me approx 2 days ago [11:32] my error test page is displayed as expected, anyway [11:32] the settings in the php.ini need to be set a certain way, it's not just enabling logging and restarting [11:32] ikonia, thet gave me that link now: http://codex.wordpress.org/User:Sivel/FAQ [11:32] they* [11:32] and it is not a bit of a help [11:33] ikonia, can you do me a favor and pastebin me you php.ini file? [11:33] kickar: why ? [11:33] so i can compare mine to yours [11:33] why ? mines not got logging enabled [11:34] ok [11:34] pastebin yours, we can look over it [11:35] http://gist.github.com/889114 [11:35] kickar: this is ubuntu 10.04 ? [11:36] Any idea why I only see ./powerbtn in /etc/acpi/event in ubuntu server 10.04 ? I want to to run a script on power change. [11:36] 10.10 [11:38] kickar: why did you set the error_reporting = -1 [11:39] ikonia, -1 should be everythinh [11:39] really, I thought that was nothing ? [11:39] I know mine didn't use -1 [11:40] ok, now is : E_ALL | E_STRICT [11:40] the same [11:41] I didn't change that [11:41] do you remember what did you change, or maybe point me to some reading [11:46] $ sudo mdadm -r /dev/md0 /dev/sdb [11:46] mdadm: hot remove failed for /dev/sdb: Device or resource busy [11:46] how do i remove this from my raid 1 array? [11:47] i erroneusly made the whole /dev/sdb a mirror while i should have made just one partition [11:52] aliverius: is it still building the array [11:52] no [11:52] kickar: I don't remember, I got the info from #wordpress guys [11:53] ikonia: i marked it as faulty [11:53] aliverius: but is the build/sync process totally complete ? [11:53] and it still doesnt remove [11:54] how do i unmark it faulty? [11:54] aliverius: you have to remove and hot add [11:54] marking it as faulty should have removed it though [11:54] how do i unmark it? [11:54] you don't [11:54] you need to remove and re-add it [11:54] both drives?! [11:55] no, the one you have marked as faulty [11:55] and I don't mean physially remove it [11:55] * aliverius fears he will render the 1st disk non bootable [11:55] remove it from the array [11:55] sudo mdadm -f /dev/md0 [11:56] i marked all the array faulty :( [11:56] ooh, you marked the array, not the disk [11:56] i mucked up right? [11:56] well, it's not the end of the world [11:57] please help me [11:57] can you pastebin the output of cat /proc/mdstad please [11:57] can you pastebin the output of cat /proc/mdstat please [11:58] http://pastebin.com/FNXwnyuM [11:58] aliverius: great, no problems, was it disk sdb you wanted to remove ? [12:00] yes [12:00] shall i mark that as faulty? [12:00] yes, mdadm -f /dev/sdb [12:00] $ sudo mdadm -f [12:00] mdadm: an md device must be given in this mode [12:01] this is what made me mark the whole array [12:01] ok lets do as you said [12:01] aliverius: mdadm /dev/md0 -f /dev/sdb [12:01] you don't want to fail the whole array, just the one disk [12:02] mdadm: hot removed /dev/sdb [12:02] ty ikonia [12:02] there we go [12:02] now i can partition it and make it like the first disk [12:02] :)) [12:02] excellent [12:11] but how do i partition it exactly as the first disk? [12:11] when i built the array [12:12] i thought it would use part of sdb not the whole sdb... [12:13] aliverius: just make a partition the same size as the one you want to mirror [12:13] be aware though that you may have performance issues if you've using part of a disk for a mirror and part for something else [12:13] mirroring is normally done disk/disk mirroring partitions like for like [12:13] really? nobody ever told me [12:13] not a partition on a disk for mirroring and a partition for something else [12:14] but ok [12:14] it may boot a bit slower and that's all [12:14] i need to mirror only 700G [12:14] the other 2x1.3G are precious [12:14] as space [12:15] not the data inside them [12:15] no [12:15] general performance may suffer [12:15] * aliverius is wondering if he will be able to run the server headless forever [12:15] :( [12:15] anyway, i cant do otherwise now [12:15] it won't be "that" bad [12:16] I don't mean your machine will grind to a halt [12:24] i will have to logout [12:24] so, bb and thanks ikonia === dendro-afk is now known as dendrobates [15:27] hello guys === IdleOne is now known as Idle0ne [15:43] how do I enable and disable a daemon at boot? e.g. smdb. I am confused by all those possibilities with upstart, update-rc, runlevels [15:43] seems unnecessarily complex. [15:45] hardwired, it depends on the daemon [15:46] lenios: OK, what are the possibilities, and how can I identify which daemon uses which method? [15:46] and... why does it depend on the daemon? can't there be one single way for all daemons? === Idle0ne is now known as IdleOne [15:49] do I edit files in /etc/init/ ? won't they get overwritten at the next upgrade? [15:51] sysv-rc-conf [15:52] iceflatline: thanks, I'll check that out. [15:53] iceflatline: this doesn't show smdb, yet it is started at boot. [15:54] iceflatline: and it has no markers for cron, yet cron is started at boot. [15:55] iceflatline: oh wait, it has smbd. [15:56] iceflatline: but no marker for cron. [15:58] aha; [15:58] CAVEATS [15:58] sysv-rc-conf only manages the symlinks in the "rc{runlevel}.d" directories. It's possible that packages may have [15:58] other ways of being disabled or enabled. [15:59] I think cron is started in user space but I don't recall. [16:00] what do you mean started in userspace? [16:00] of course cron runs in userspace and not in the kernel [16:00] I think I got the rc stuff covered with that sysv-cr-conf, thanks! [16:01] now I need to fugure out the other half, the upstart stuff. [16:02] initctl list gives a nice list of what is running [16:04] sysv-rc-conf will only work on sysv scripts, not upstart ones [16:04] AFAIK there is no simple, easy way to manage upstart scripts, you must edit them manually [16:05] qman__: I think i just found one: [16:05] Hello all [16:05] With newer versions of Upstart, you can make use of override files and the manual stanza to achieve the same result in a simpler manner: echo "manual" >> /etc/init/myjob.override [16:06] Does someone know a good tutorial on setting up Dovecot+Postfix+SASL+LDAP on a recent Ubuntu sever? [16:06] I seem to be unable to do it despie an overkill of information [16:06] i need some virtual users [16:06] fefwerf, that's covered in the server guide [16:07] fefwerf: I suggest you find some preconfigured config files and use them as a base for your system [16:07] qman__ looking at the right as we speak [16:07] fefwerf: that's what I did with dovecot+exim+postgres [16:07] it's under postfix [16:07] it covers SASL and postfix+dovecot [16:08] https://help.ubuntu.com/10.04/serverguide/C/postfix.html did not work. Should I just install the postfix-dovecot package an add on LDAP&virtual users? [16:08] what do you mean by "did not work"? [16:09] postfix/smtpd[4657]: fatal: no SASL authentication mechanisms [16:09] I don't understand [16:09] do I have to activate SASL somewhere, Dovecot and Postfix are running [16:10] did you do the configuration in dovecot, and restart dovecot? [16:10] ye [16:10] Sorry i just checked, dovecot-auth is NOT running [16:11] I will try to fix that [16:11] eh, that guide needs updating [16:11] it uses /etc/init.d to restart services [16:12] auth-worker(default): Killed with signal 15 (by pid=1 uid=0 code=kill) [16:13] I did postfix/dovecot/sasl with that guide a couple of months ago. I don't remember stumbling across anything that got in my way [16:13] looks like somehow my dovecot is not configured right, I'll look into it [16:13] dovecot main process (5061) terminated with status 89 [16:15] qman__> it uses /etc/init.d to restart services [16:15] ha! [16:15] the transission to upstart is a mess :-) [16:15] yeah [16:15] really not a fan of upstart [16:15] what they should have done is look at the NetBSD rc.d system [16:16] it is so simple and effective [16:16] systemd is pretty brilliant too [16:16] Sorry hardwired&qman___ looks like I broke my dovecot.conf, will correct and try again [16:17] yep. the rc.d lacks parallel startup. that's a big drawback these days [16:17] fefwerf: not need to apologize :-) [16:17] I think parallel startup is no big issue for most servers [16:18] But what do I know ;) [16:26] very true [16:26] my upstart is not recent enough for the above-mentioned method to work [16:27] OK fixed my dovecot.conf, looks like Postfix can use SASL now [16:28] Now i need to add on LDAP&Virtual users [16:39] I want a combination of local system users and virtual LDAP users for my Dovecot-Postfix setup [16:39] how should I do that? [16:41] is https://help.ubuntu.com/community/PostfixCompleteVirtualMailSystemHowto a good starting point? [16:44] laters [18:16] Hallyn??? [18:16] You awake? [18:20] I created an Upstart service, but it is unable to start for some reason, while if I run the same command I pass to exec from the command line, it works fine. How can I debug this? [18:20] (by unable to start, I mean the service immediately terminates) [18:40] There was a way to restrict a user to his home dir by typing a special home dir in /etc/passwd. I can't remember it. Could anyone remind me of it? [18:40] There was an additional character to prepend/append to home dir in /etc/passwd [18:46] hallyn: i got a box that can amd qemu-kvm now fyi [19:08] ikonia: are you there? [19:17] aliverius: yes [19:20] ikonia: i partitioned the second disk with exactly the same size [19:20] do i proceed normally? [19:20] yup [19:23] sudo mdadm -a /dev/md0 /dev/sdb4 [19:23] ? [19:23] depends on your partion layout and raid type [19:23] read up on the options this time [19:23] raid 1 and the partition will be sdb4 [19:24] i am browsing the man pages but still i am not confident [19:24] what is not clear to you ? [19:24] if that is the correct syntax [19:24] what makes you think it's not [19:25] the fact the man page doesnt make it clear [19:25] but from my experience it should be the right one [19:26] ok, what are you worried is not right [19:26] and what part does the man page not make clear ? [19:26] (trying to get you confident in what you're doing) [19:26] New bug: #743763 in exim4 (main) "package exim4-config (not installed) failed to install/upgrade: le sous-processus script post-installation installé a retourné une erreur de sortie d'état 1" [Undecided,New] https://launchpad.net/bugs/743763 [19:32] ikonia: nothing is unclear, maybe i wanted to have an example command, just to be sure [19:32] but oh [19:32] mdadm [mode] [options] [19:32] beggining of the man page :p [19:32] ok now i am sure [19:32] lets build the array! [19:33] mdadm: /dev/sdb4 not large enough to join array [19:33] oooops [19:35] http://pastebin.com/rghs01j4 <--- /dev/sdx4 are the two partitions that would become an array [19:36] is it ok if i give the second disk partitio a few megabytes more? [19:36] sure [19:37] isnt it strange since the blocks are exactly the same? [19:40] aliverius: the disks aren't [19:41] no they are [19:41] both same model [19:43] that's odd [19:43] not uncommon though [19:50] ikonia: http://pastebin.com/XAgdjStA :) [19:50] afterwards is there anything more to do? [19:50] nothing more [19:50] aliverius: great news, nice job. [19:51] sit back and enjoy [19:51] thanks a lot [19:51] i dont see you in lfs [19:51] got bored of it? [19:59] hi, anybody here install server on a USB stick? [20:01] you can do that, but very frew people would [20:01] * centHOGG server NAS [20:01] yeah [20:01] ever seen freeNAS? [20:01] runs off a USB stick [20:01] for boot [20:03] don't want to run a nas of usb [20:04] actually freenas is pretty clever.. just too much FS overhead [20:50] hello to all. i have heavy prolems here with ubuntu server i installed just yesterday [20:51] problem is the hybrid disk that i use inside that server [20:51] after the installement of ubuntu everything worked like it should [20:51] but then after several reboots ubuntu wanted check the disk with fscheck [20:51] here it stuck now [20:52] somehow ubuntu crashes full when it try to fscheck the hybrid disk [20:52] before it worked everything all fine but this fscheck breaks now everyting [20:56] New bug: #743821 in samba (main) "package winbind 2:3.5.8~dfsg-1ubuntu1 failed to install/upgrade: le sous-processus dpkg-deb --fsys-tarfile a retourné une erreur de sortie d'état 2" [Undecided,New] https://launchpad.net/bugs/743821 [20:59] is anybody here to help me ? [21:04] whaz wrong [21:04] xperia: not that I'm not willing to help, but alas I have no idea hwat a hybrid disk might be [21:04] ditto [21:04] *what [21:17] xperia: I don't know much about the hybrid disks..I'm running a Seagate Momentus XT 500 GB 7200RPM SATA 3Gb/s 32 MB Cache 2.5 Inch Solid State Hybrid Drive ST95005620AS-Bare Drive Seagate Momentus XT 500 GB 7200RPM SATA 3Gb/s 32 MB Cache 2.5 Inch Solid State Hybrid Drive ST95005620AS-Bare Drive with no problems on an 64 bit 10.04 Desktop version. [21:18] kewl [21:19] xperia: I would test for the drive being bad. Then try to re install ubuntu..did try different kernel version during boot? [21:37] thesheff17: i have the exact harddisk here is it a Seagate Momentus XT 500 GB 7200RPM SATA 3Gb/s 32 MB Cache 2.5 Inch Solid State Hybrid Drive [21:38] xperia: what version of ubuntu are you using? [21:38] i am using it on the newest ubuntu server natty alpha 3 release [21:38] it boot all fine till fscheck [21:38] but then it crash full [21:38] screen goes off and nothing happen [21:39] xperia: try 10.10 or even 10.04 which is an LTS...why alpha? [21:39] LTS [21:39] LTS = long term support [21:39] * centHOGG concur [21:40] thesheff thanks for tip [21:40] Hey everyone. I have a question about dovecot. [21:40] looking just right now if disk is good [21:40] When postfix saves an email, it saves it into /home/vmail/domain/user/new (or cur or tmp). [21:40] But when I try to access it via imap, dovecot goes to /home/vmail/domain/user/Maildir/new (or cur or tmp). [21:40] How do I correct this? In dovecot.conf, it is set to: mail_location = maildir:/home/vmail/%d/%n [21:45] storz this is more a postfix / doecot question [21:45] normally all the info configuring this software is availble on the net [21:45] if you dont have luck possibility is to ask in the chanels of postfix and dovecot [21:46] xperia. Thanks. I know. I was just hoping that one of you have run into this before. Figured it'd be a bit quicker. [21:55] what is the command to configure network on Ubuntu server? [21:55] Al-Bundy: see https://help.ubuntu.com/10.04/serverguide/C/network-configuration.html [21:57] lets say i decide to boot my raid 1 without one disk [21:57] then i reboot woth both disks connected [21:58] what happens? does it sync automatically to the newst disk? [22:11] New bug: #743858 in openssh (main) "sshd not appending to /var/log/btmp" [Undecided,New] https://launchpad.net/bugs/743858 [22:12] thesheff17: i booted the server now with a live cd. maked a fsck on the disk and ubuntu server load all fine now [22:13] after the reboot [22:13] looks like a problem with fsck in ubuntu natty alpha 3 [22:13] okay have to do see you all next time bye [22:44] ikonia: you there? i made some questions above [22:56] New bug: #743883 in samba (main) "cannot load packages. Broken packages" [Undecided,New] https://launchpad.net/bugs/743883 [23:56] New bug: #743920 in tftp-hpa (main) "package tftpd-hpa 0.49-1 failed to install/upgrade:" [Undecided,New] https://launchpad.net/bugs/743920