[00:36] I figure #ubuntu-server would be appropriate for my problem. I have an eth0 issue. The wire is plugged in and it seems that I b0rked it because I endlessly was trying to bridge few days ago. [00:36] If anybody could help me resolve my issue, that'd be truly great. ;< [00:42] what [00:43] why did you do that [01:00] how can i run /home/chris/bin/psig once an hour as the user chris [01:02] ? [01:04] cron [01:04] how come i don't have a /etc/crontab? [01:05] FFForever, use crontab -e [01:05] or just have a look at the crontab vile [01:06] file* [01:06] it's stored in some obscure place in /var [01:06] it says where in the crontab man file, but for editing it's best to use crontab -e [01:06] Hey folks, me again, I was wanting to change my configuration on my server to go from static ip to dhcp on eth0, I was pointed to /etc/sysconfig/network-scripts/ifcfg-eth0 but it doesn't seem to exist, any ideas? [01:07] tsrk, crontab is an unknown command [01:07] FFForever, it's a command on mine... [01:07] chris@chr1831:~$ crontab [01:07] -bash: crontab: command not found [01:07] FFForever, do you have cron installed? [01:08] i do now :D [01:08] ok, sounds good :) [01:08] how'd you get ubuntu installed without it though? [01:08] did you uninstall it? [01:09] also, i just noticed, there is /etc/crontab [01:09] tsrk, i am using the xen image provided by my vps =\ [01:09] ah i see [01:09] it is missing a lot of normal stuff =\ [01:10] MkJackson: sysconfig you will find on RedHat boxes, /etc/network/interfaces for ubuntu/debian [01:10] MkJackson: http://www.ubuntugeek.com/change-ubuntu-system-from-dhcp-to-a-static-ip-address.html [01:10] Just reverse the directions [01:12] like this?, 0 * * * * /home/chris/bin/pisg [01:12] (for once an hour) [01:29] stickystyle: thanks! [03:10] This is driving me nuts. postfix seems to be ignoring /etc/aliases and .forward and I'm stuffed if I can figure out what is going on. Any suggestions on narrowing down the issue? [03:11] Mail to external addresses is working fine. Mail to root gets sent to root@mydomain, and not to the aliases or .forward. [03:41] Anyone? [03:44] owh: Did you run newaliases after changing them? [03:46] yup [03:48] owh: Anything useful in logs? [03:49] ( /var/log/mail.log) [03:49] genii: No, it just delivers the mail, no errors, nothing indicating that it's trying to use an alias or a .forward and failing, it's just sending it. [03:49] Just not to anything I've specified :( [03:50] genii: If I drop out, my workstation internet is doing something weird, no web-traffic at all :( [03:50] owh: After alias change/newaliases did you refresh/reload postfix? [03:51] genii: Yup [03:51] genii: It's like there's a switch that makes all deliveries remote, rather than try local first. [03:53] owh: Do other aliases in the system work? [03:53] genii: You mean postfix aliases? [03:53] owh: Yes [03:54] genii: There's only one alias. [03:54] genii: All it's supposed to do is map root to administrator [03:55] genii: Then in the .forward for administrator it's supposed to forward mail to external users. [03:56] genii: Just sent an email direct to administrator, hadn't done that yet. It sends it to administrator@mydomain, not to the address in .forward [03:57] owh: I havent had to debug postfix in a long time, apologies. My next guess would be perhaps it isn't using the aliases because maybe it hasn't been specified in it's mailrc or with postconf or such [03:57] (where the aliases file is) [03:58] This machine was running exim4 until a few hours ago. It had the same problem. I wasn't happy with exim for a number of reasons. I purged it and did a fresh postfix install. [03:58] postconf -n shows that it knows about /etc/aliases [03:59] Just found something strange: [03:59] mydestination = localhost.localdomain, localhost [03:59] Hmm. If same issue with two different mail systems, points to some other issue like routing paths or so [03:59] But /etc/mailname shows mydomain [04:00] I wonder if I should change /etc/mailname to localhost.localdomain [04:01] That then means that unqualified addresses are delivered to the endpoint for that. So, root becomes root@localhost.localdomain, which is then delivered locally. [04:01] owh: Past this is where I currently would have to consult manuals myself on the variables and what gets used in what order there. I recall something about mydomain or overrides, vaguely [04:01] Does that make sense? [04:01] genii: You understand that mydomain is not what it actually says right, it is my actual domain. [04:02] owh: Yes, if you have mydomain it will try fqdn [04:03] So, then if I change /etc/mailname to localhost.localdomain, all should be good I'm thinking. [04:04] Holy crap. [04:04] That looks like it might be working. [04:04] * owh does more testing. [04:04] * genii drinks a decaf and thinks about bed [04:07] Whoot! [04:08] genii: Thank you for prodding me along. [04:08] owh: Glad to have assisted, although you did all the actual diagnoses/work [04:09] genii: Perhaps, but your prodding made me look in other places, which then prompted me to think harder, so without your poking, this would have taken much longer than it already has. So thanks. [04:09] owh: Well, glad it's resolved :) [04:09] Yeah. [04:10] * owh curses /etc/mailname. I suspect it was what caused exim to die also. [04:10] * owh pats ngrep :) [04:12] Neat, I'll have to remember that next time I'm using tshark === hggdh_ is now known as hggdh [04:20] twb: What, ngrep or mailname :) [04:20] ngrep [04:20] twb: I like this: ngrep -d any port 25 [04:20] Oh, so you're just using it like tshark. [04:21] twb: Dunno, never used tshark :) [04:21] Did you also try "nc localhost smtp" and then manually "HELO fred"? [04:21] tshark = wireshark [04:22] twb: You mean like telnet localhost 25 :) [04:22] Yes, except that telnet is evil [04:22] nc/socat are cooler [04:22] Of course it is. [04:23] Now all I need to do is figure out which *&%$ DNS server has fallen over... [04:24] Wonderful, is OpenDNS working for anyone else? [04:24] for i in It gets better. I can ping OpenDNS from a remote host, just not from here - carry on. [04:26] owh: It's working fine for me [04:26] (opendns) [04:26] genii: Yeah, works on a remote server here too, just not locally. No idea why though, time to reset my internet connection. [04:30] Nothing like bodgy network device firmware. [04:32] so in order to install xen, do i need an OS as a base, or i first install xen then guest operating systems on top ? [04:33] No idea. [04:41] orudie: Host OS first, then xen packages, then guest OS within xen. [04:42] thanx [04:42] trying to figure out why my computer resents on rundom... [04:43] last thing i had installed on it was ubuntu server [04:43] trying to think of a piece of hardware that could have caused it [05:32] Im trying to reduce the size of hardy/JeOS. Can I delete /usr/src/linux-headers-* if I have no plans to ever run gcc or build anything on it [05:33] What is a JeOS? [05:34] That's still 350MB for me. [05:34] http://www.ubuntu.com/products/whatisubuntu/serveredition/jeos [05:35] Oh, a kernel for VMs [05:35] Im just looking for a debian based (apt) OS that has a small (not busybox) footprint. Hopefully around 100Mb. [05:35] You'd be better off removing the associated package. [05:35] Ubuntu server is defalt 700Mb, JeOS is nice at 350 but still not ideal for my needs [05:36] ever heard of dpigs? [05:38] oh_noes1: do you have the install in front of you (i.e. ssh or whatever)? If so, is aptitude installed? [05:39] Yep I have aptitude [05:39] OK, you can use that (or debfoster/deborphan) to work out which packages you don't need [05:39] Ubuntu JeOS is just the same as Ubuntu Server, except kernel optimized for VM's and missing all X and other unused libs [05:40] oh cool, ta let me look into that [05:40] You can also use grep-dctrl to list installed packages by package size, so you can remove the biggest ones first. [05:40] The grep-dctrl manpage contains that use case in its EXAMPLES section [05:40] oh_noes1, I think you should be ok removing linux-headers if you are sure that you won't be building anything [05:41] gaveen: I'm surprised its even installed by default [05:41] building some kernel modules require them [05:42] *kernel modules (not just some) :) [05:42] yeh thanks, i'll start by removing the headers and maybe even build-essentials [05:42] then delete /usr/share/doc [05:42] might be a quick win [05:42] but for a JeOS it's kinda odd to have it. true [05:42] gaveen: you mean it's there for bloody vmware-thingy-tools? [05:43] Grrpmnh [05:43] I think so lol [05:43] oh_noes1: you should NOT start out by deleting parts of the filesystem [05:43] oh_noes1: start out by removing unnecessary packages. [05:43] I'm 99% confident vmwaretools (lets ignore it's 100Mb) needs it to build an additional 150MB into the OS [05:43] twb, possibly. plus there are other modules,.. like device drivers requiring them [05:43] gaveen: it's a VM! [05:43] twb: there are no unnesscary packages, headers, build-essentials and thats it [05:43] it [05:44] oh_noes1: why do you need build-essential and kernel headers? [05:44] Its not a life server, it's for an ISV and obviously i can create as many of these VMs as I want to test [05:44] breaking anything isnt a problem [05:44] twb: i dont ... vmbuilder must be auto installing them [05:44] "du -mx / | sort -nr | head -$((LINES - 1))" will tell you which directories consume the most space. [05:45] *sigh [05:45] 126 /usr/lib/vmware-tools is the 8th on the list, closely behind /usr lol [05:46] oh_noes1: I suggest you install open-vmware-tools instead of the proprietary one. The former are pre-compiled by Debian, so you don't need to build them and thus don't need the build tools. [05:46] thanks i'll look into that [05:46] Also, the vmware tools are really only useful for two things: hgfs (not available in vmware-server) and opengl tunnelling (not useful for a server VM). [05:47] ironically, we only use vmware tools for the shutdown of the guest OS from ESX. That's it [05:47] Well, from the service console. [05:47] ssh $guest halt [05:48] Note also that open-vm-tools is split into GUI and CLI sections (unlike VMware's version), so you don't need to install all those X libs. [05:49] Oh, maybe you DO need to build open-vm-source using m-a... [05:49] But you could do that once, roll the .deb, then just install that deb into each VM [05:51] yeah, I'll investigate open-vm-tool more [05:51] even if it means building a cut down lean CLI only specific version for myself [05:51] that just halts the OS [06:09] help [06:13] dhruba: insufficient data [06:14] dhruba: C [06:14] try turning it on and off again [06:14] (it crowd, hehe) [06:14] dhruba: what's happening? [06:20] what's an LTS Server is that the same as Desktop server? [06:33] quizme: LTS is a policy, not a thing. [06:33] quizme: an LTS release is a bit like having an extended warranty [06:33] * ajmitch guesses there could have been confusion with LTSP [06:34] There is no connection between LTSP and LTS -- they just happen to have similar acronyms. [06:34] very similar [06:35] Of course, LTSP runs on Ubuntu, including the LTS releases... [06:35] * tonyyarusso thinks you're making it worse [06:35] tonyyarusso: that's what I do [06:35] oh [06:35] ajmitch: hehe [06:35] LTS = Long Term Support [06:36] oh oh [06:36] when i install postfix then TLS and SASL are compiled into postfix already? [06:36] Compiled yes. Configured no. [06:37] * ajmitch prepares to break the world by uploading merges [06:43] ajmitch: Don't forget the alpha freeze. [06:44] soren: I know, I hadn't seen any of the usual freeze announcement in #ubuntu-devel yet though [06:45] so far it's just apache2 & squid, since they tend to be useful for me [06:46] ajmitch: It was announced on ubuntu-devel-announce. [06:46] Haven't seen anything in #ubuntu-devel either.. [06:46] yes, the tuesday-thursday thing [06:46] * ajmitch was reading mail also :) [06:46] *nod* [06:47] besides, I wasn't going to upload these straight away because of that [06:48] I was even thinking of getting someone to quickly check them first, since it's been a little while since I uploaded to main [06:51] * ajmitch wonders if apr-util just needs a giveback on i386 [06:53] ajmitch: You don't need no stinkin' check overs! This is the devel version we're talking about - upload a new libc6 and live on the edge! [06:53] because who cares if the package compiles, right? [06:53] EXACTLY [06:54] And quite frankly, what business do they have worrying about whether things will compile if they can't even boot? I mean honestly people, quit yer fretting. [06:55] sad to say, but the apache & squid packages compiled, installed & even worked for me [06:55] I can't upload them in a state like that [06:55] ajmitch: Compile lighttpd and polipo and just name then apache2 and squid and upload those then. [06:56] I may as well have djbdns Conflict/Replace bind [06:57] and replace ssh with rsh! Or better yet, telnet! Yay! [06:59] now you're going too far [06:59] :( [07:00] And this is why I'm not the one with main upload privs? [07:01] for the good of mankind [07:12] is xen supported in current vm-serveR? [07:13] dom0, no. domU, yes. [07:14] thanks, soren [08:07] Hello, anyone around ? [08:10] Has open-vm-tools been backported to hardy? [08:12] I was using windows as a webserver but switched to ubuntu and I dont have static ip, and I was using everydns but I dont think they have a program for linux [08:13] oh_noes, look in backports to find out ... [08:14] huh? [08:18] Sangrial, are you goign to ask a question? [08:18] other then 'anyone around' [08:18] i did ask =P [08:18] I was using windows as a webserver but switched to ubuntu and I dont have static ip, and I was using everydns but I dont think they have a program for linux so I need a dynamic dns updater [08:18] no, you gave us a sob story :P [08:19] any recommendations? [08:19] Sangrial, try something like `apt-cache search dynamic dns` [08:19] ok thank you [09:14] Hello anyone around ? [09:18] <_ruben> !ask [09:18] Please don't ask to ask a question, simply ask the question (all on ONE line, so others can read and follow it easily). If anyone knows the answer they will most likely reply. :-) [09:18] I have set up Ubuntu 9.04 Server in a PC and configured LDAP, Samba, DNS, DHCP and NTP on it. I am trying to configure a Ubuntu 9.04 Desktop. I received positive response from the desktop through ldapsearch. I am able to see the users I had created through phpldapadmin in the server. But when I am trying to switch over to any user, the logon is failing [09:21] <_ruben> never used ldap auth myself, so cant help you there .. i assume you read the server guide? [09:24] I have gone through the Server Guide thoroughly. But unfortunately, there is very little related to joining of a linux client to a linux server. [09:29] ldap-auth-client [09:29] auth-client-config -l (to list the available profile in /etc/auth-client-config) [09:30] auth-config-client -a -p lac_ldap for default ldap auth [09:30] dhruba: see above [09:30] ok i'll try [09:31] in the server guide chapter 6.1.7 ldap auth [09:31] not as easy as I wrote above but fairly the same [09:32] is this applicable to 9.04 also? [09:32] bo reason it shouldn't be [09:32] no [09:34] I saw this guide. But stuck-up as the client gz file was not available [09:35] gz? [09:35] yes. it refers to a file to download [09:36] you are building from source? [09:37] Yes I was trying to. But had to abort [09:37] using the deb wouldn't be easier? [09:38] sudo apt-get install ldap-auth-client [09:39] Yes. Later I did so [09:40] By the way, this server guide you referred is of 9.04? [09:40] <_ruben> you should always consult the server guide matching your install os version ;) [09:41] <_ruben> especially when it comes to stuff like ldap and ad integration, which are fast moving targets [09:41] Yes I always do so [09:44] But it is very difficult from the document to identify how to configure a ubunto client to join the ubuntu domain === \sh_ is now known as \sh === Nicke_ is now known as Nicke [10:48] Hello. Anyone around? [10:49] alot people around [10:49] /names ... 182 users in here O_O [10:54] While configuring Squid, I am getting only http: sites. It is not opening gmail or gtalk. Also VPN is not passing through. How should it be configured? [10:55] <_ruben> transparent proxy or not? [10:55] Yes [10:55] <_ruben> transparent proxy only works for port 80 [10:55] Ok [10:56] <_ruben> transparent proxy for https would be a man-in-the-middle "attack" [10:56] How should I configure it securely and for all above services? [10:57] <_ruben> not as a transparent one :) [10:57] <_ruben> as in: manually configured on each client [10:57] Is any document available on it? === mcasadevall is now known as NCommander [12:31] New bug: #373999 in samba (main) "CIFS loops endlessly when remote host is down" [Undecided,Incomplete] https://launchpad.net/bugs/373999 [13:22] hi everyone ... I have a general question but it stays unanswered in other chats so I will try it here ... I am trying to remove the package squirrelmail via aptitude purge ... this also wants to remove apache2 (as it is not needed according to aptitude) ... I want to block apache2 from being removed ... there must be a way; I have disabled autoremove via apt.conf.d [13:22] but it seems to be ignored [14:23] can i install phpbb as a package [14:23] ? [14:26] oruwork: yep there's phpbb3 or phpbb2 packages [14:27] sommer-> so its apt-get install phpbb3 ? [14:28] oruwork: yeppers === asac_ is now known as asac [14:44] sommer-> cant seem to find korean language pack for phpbb3 any ideas ? [14:53] oruwork: it may not be packaged I guess... should be able to download it from the phpbb site I'd think [14:54] sommer-> i just installed phpbb3 as a package on hardy, not sure how to access it in the browser i tried www.myurl.com/phpbb3 didnt work [15:02] oruwork: not sure, I'm not familiar with the pacakge, but there's probably an apache.conf in /etc/phpbb, or somewhere similar, that needs to be enabled by apache [15:02] oruwork: usually /etc/appname/apache.conf files are symlinked to /etc/apache2/conf.d/ then when you restart apache2 the new config is used [15:02] oruwork: so you might try just restarting apache2 :) [15:20] sommer-> restarted and still nothing [15:23] oruwork: not sure then, but you might take a look at the community wiki: https://help.ubuntu.com/community/PhpBB2 [15:28] sommer-> very outdated article :( [15:39] sommer-> i have apache.conf database.inc.php styles inside /etc/phpbb3 [15:43] oruwork: /usr/share/doc/phpbb3/README.Debian.gz should have further instructions... or where ever the README.Debian file is located [15:59] so is there any word about further movement towards upstart? [16:10] anyone know how to disable dhcp client with out removing it ? === dendrobates_ is now known as dendrobates [16:16] New bug: #375493 in tomcat6 (main) "tomcat6 needs debug start mode with jpda" [Undecided,New] https://launchpad.net/bugs/375493 [16:19] oly: which do you use? maybe this helps: /etc/init.d/dhcdbd stop [16:19] oly: Um... What are you trying to do, exactly? [16:20] well basically i have ubuntu server installed [16:21] i modified /etc/network/interfaces with a static ip [16:21] which sticks for about 10 minutes [16:21] then it gets replaced vid dhcp with another ip [16:21] <\sh> oly: you need to kick your dhcp server to not use the static ip in your dhcp range [16:21] i would like to leave dhcp-client on the system in case i need it in future as i can run it manually [16:22] <\sh> oly: or you configure your dhcp with the mac addr of your server nic and push it a static ip via dhcp [16:22] should i not be able to stop ubuntu from trying to retrieve one though ? [16:23] <\sh> oly: afaik ubuntu server only activates dhclient when it's set via /e/n/i [16:23] <\sh> oly: if it's still running after /etc/init.d/network restart and you set a static ip in /e/n/i then it's a bug...but it never happened here at my place with the server flavour [16:23] also the dhcp server is giving out ip addresses of 192.168.0.100 upwards and i have a static ip of 192.168.0.4 set on the server machine [16:24] i did restart /etc/init.d/networking but it was still replacing [16:24] perhaps a reboot will stop it [16:25] although i tend to try and avoid them :) [16:25] but its a fresh install i am setting up so no harm [16:26] \sh i have setup servers before and not had this problem [16:26] it my first 9.04 server install though [16:28] oly: pkill dhclient [16:29] is there a permenant way ? [16:30] oly: If you reboot, it'll pass. [16:30] oly: The problem is this: [16:30] okay thats good [16:30] ifupdown does not remember how an interface was configured when you take it down. [16:30] So... [16:30] You brought it up while it was configured to use dhcp. This started the dhcp client. [16:31] aha that makes sense now [16:31] Thenyou changed your config, ifdown'ed it, and ifup'ed it. When you ifdown'ed it, ifupdown had no clue it had a dhcp client associated with it. [16:31] ...so it didn't kill it. [16:31] ifup gave it the static address.. [16:31] okay makes sense now [16:31] but dhclient lives on, and doesn't know to back off. [16:31] thanks for that explanation [16:31] Sure thing. [16:31] you can just kill the dhclient process and ifdown+ifup the interface, and you'll be fine. [17:29] hey I cannot find any documentation man pages or anything on how to provision the samba4 and openchange packages anyone know where to look? [17:47] i did a upgrade but it failed on configuring systraq. i moved the old dir and ran dpkg --configure -a [17:48] as suggested. how do i continue the upgrade? [17:48] or whas that it? [17:50] anyone please? the mailserver is down atm [17:53] I would assume that you would restart the upgrade process now that you dealt with that package manually [17:55] btw systraq seems to be a server monitoring and reporting tool [17:55] see http://packages.ubuntu.com/hardy/admin/systraq [17:56] hope that helps I seem to be getting no where on my problem [17:57] a_ok: I would had to give you wrong advice. But if you downloaded all the packages and it failed when configuring them and then you ran a dpkg --configure -a I think you should be done [17:58] ssd7: yeah i noticed that it was the only thing that had to be done and rebooted just fine. thanks anyway [18:06] ssd7: do i need to report the failure somewhere? [18:25] where is apache2 config located? [18:34] wizardslovak: /etc/apache2 [18:36] thx [18:36] is there way i can backup apache2 config file? [18:36] wizardslovak: read /usr/share/doc/apache2/README.Debian.gz for an explanation of the structure [18:37] i dont really have time to read i neet to know to back it up [18:37] wizardslovak: Pretty much all the stuff in /etc/apache2 and it's subdirs [18:38] so if ill copy it with scp to other computer ,i wont erase it in server right? [18:39] wizardslovak: it's just files, back them up any way you would any other directory full of files. Note the actual docs served up are not in that directory, so it tends to be small. Yes, you can just scp -r the directory somewhere else, that won't change anything on the source host. [18:42] so just scp -r file host@ip:/locationd? [18:45] yes, where "file" is spelled "/etc/apache2", hmm, reading the scp man page, -r follows symlinks, so you might end up with some stuff duplicated. Something that preserves symlinks might be better. tar into a temp file ans scp that, or rdist -a, [18:46] thhx rfm [18:46] i gotta go [18:46] thx again [18:48] I'm currently using Citadel as a groupware server. It supports SSL for the text client connections and the HTTP, but as far as I can tell the XMPP does not. So, I want to set up stunnel or similar to forward encrypted connections on 0.0.0.0:5223 to 127.0.0.1:5222 in the clear. Can I do that? Anyone know how? [18:49] stunnel seems to be more at home with an inetd environment than one where there's just free daemons. [18:56] is ubuntu server based off debian-unstable? [18:57] afaik, yes [18:57] mostly [18:58] So how should one incorporate that information into an evaluation of ubuntu server as a server platform compared with centos/rhel, or debian stable? [18:59] debian compares well with centos, both being community distros with a rock solid base. [19:00] rhel is a bit on its own in this lineup as it is rock solid but commercial [19:00] ubuntu is both community and commercial, the commercial part is something debian lacks, and is also rock solid stable, just not quite the granite of debian. You bet newer packages though. [19:00] s/bet/get/ [19:01] sure .. except tis based off unstable which denotes a lower testing and qa time doesnt it? [19:01] unstable includes packages that are stable on their own, just the system itself is not tested for three years to make sure its *stable* stable. [19:03] it's on a six-month release cycle, and is considered stable by Ubuntu's standards, which are somewhere between RedHat and Fedora standards. [19:05] fedora has standards? [19:05] heh [19:05] didnt they release with abroken version of mysql? [19:05] SirStan, I just made that comment in another dev channel [19:05] ubuntu's done worse [19:06] fedora kind of does a rolling release with no proper package freeze, so it's always on the bleeding edge [19:06] SirStan, I think that came from upstream mysql to be fair and actually, it *is* broken for debian/ubuntu right now [19:06] myself, on Ubuntu Desktop I make heavy use of PPA repositories and GetDeb.net and even some source compiling, so I get bleeding edge too. [19:09] poor old mysql, sucked up by SUN and now in the hands of Oracle. [19:10] It's gonna either die or turn in to Oracle Lite. The King (MySQL) is dead! Long live the King (PostgreSQL) [19:11] cellofellow, doubtful that it will die anytime soon [19:11] oh, sure, it'll take a while [19:11] cellofellow, it would be a bad idea for oracle to kill mysql actually [19:11] and maybe postgres isn't the new king, but that fork of mysql run by the original developers. [19:11] i got the #centos kids all in a tiffy. [19:12] of course, but it's still gonna be Oracle Lite. [19:12] lol [19:12] what'd you say? [19:12] Asked for a quantitive summary of the merits/quality of RHEL/centos over ubuntu/other platforms. [19:12] and got shit like "redhat supports more foss" [19:12] as metrics. [19:12] lol, yeah right [19:12] im not pro-rhel or centos. [19:12] I was looking for an informed summary of the quality of the two. [19:13] hello [19:13] sure, red hat commits more to the kernel than canonical, but what gives? Canonical cares more about gnome than the kernel, big deal. [19:13] not fud about releases per year, or number of foss advocates employed. [19:13] sorry for missing meeting, i had a business meeting :/ [19:14] hey, is there a posiblity to show realtime response time in apache2 in graphs per requested url ? [19:14] hagedorn: apache doesn't capture processing time [19:14] firebug has a nice graph showing load times per url in a web page [19:14] hagedorn: it would have to be a function of your development platform [19:15] thats true.. i assumed you wanted a server side summar [19:15] yy [19:15] SirStan: with %s and %T it does! over the logfiledefinition [19:15] SirStan: mod-securtiy do it to [19:15] really [19:16] bur we have to big logfils, lots of gigs per day couse of high traffic website [19:16] hagedorn: url? (I cant google %t) [19:16] ah .. nice [19:16] write a cricket graph generating perl script that tails the file [19:26] New bug: #375593 in samba (main) "cannot browse samba shares without editing smb.conf" [Undecided,New] https://launchpad.net/bugs/375593 === Adri2000_ is now known as Adri2000 [19:47] Interesting. "TCP: Treason uncloaked! Peer X.X.X.X:37017/37125 shrinks window 1592966069:1592966319. Repaired." (IP censored). Haven't seen this previously in dmesg [19:55] how can I check the last 15 log in attempts? [19:57] fevel: Might be in /var/log/auth.log [19:59] ok [20:02] My postfix server is bouncing back emails with the error "Relay access denied" [20:03] What would be causing this? [20:09] jumbers: it thinks that the client is coming from an unapproved network [20:10] lamont: I'm sending the test mail from gmail to my domain email [20:10] jumbers: /var/log/mail.log will tell you. [20:10] then it doesn't accept mail for that domain [20:10] or it doesn't accept mail from outside, at all [20:11] ScottK: you gonna be at UDS? [20:11] ivoks: ? [20:11] ivoks: I'm using a default setup based on this: https://help.ubuntu.com/9.04/serverguide/C/postfix.html [20:11] lamont: Unless something happens in the next two weeks to convince me not to bother, yes (the chances of that are non-zero, BTW). [20:12] ScottK: cool. postfix bof sometime, eh? [20:12] Sure [20:12] * lamont _will_ be there [20:12] yay! :) [20:12] lamont: yes? [20:13] ivoks: shorthand for "same question"... [20:13] jumbers: so, i suspect you wrote 'mail.example.com, localhost.localdomain, localhost'? [20:13] lamont: will be there [20:13] \o/ [20:13] ivoks: Correct. Changing it to my domains obviously [20:14] jumbers: and steve? :) [20:15] ivoks: Changed that too [20:16] jumbers: there's a note 'Replace mail.example.com with your mail server hostname, 192.168.0/24 with the actual network and class range of your mail server, and steve with the appropriate username.' :) [20:17] ivoks: Yes. I changed those [20:17] then that's it [20:19] I've isolated the problem. It works if I send to mail.example.com, but it doesn't like example.com [20:20] Is it possible for it to accept @example.com? [20:20] then open main.cf, and add example.com [20:21] for mydestination [20:22] sommer: is it possible to change server guide now? :/ [20:22] sommer: 'Replace mail.example.com with your mail server hostname' isn't quite correct... [20:25] ivoks: ummm, ya we can make changes in an SRU [20:25] ivoks: it's usually a somewhat slow process though [20:25] ok [20:26] what should be changed exactly? [20:27] just a second [20:27] * sommer standing by :) [20:31] phone :/ [20:32] heh, I had one of those once [20:33] doh... [20:33] where were we? :) [20:34] Instead of 'Replace mail.example.com with your mail server hostname' there should be something like [20:34] 'Instead of mail.example.com, put the domainname for which you'll accept email' [20:34] or something like that [20:35] ah gotcha... I'll be sure to add that to the karmic docs [20:35] native english speeking postfix gurus, please help :) [20:35] I'll file a bug for jaunty and when there's an SRU for the docs package I'll be sure to get it included [20:36] ivoks: of course, virtual domains make that sentence not quite precise either [20:36] for the pain [20:36] that's correct [20:36] but in this howto we don't setup up virutal domains [20:36] yay [20:36] should we cover virtual domains? [20:36] I trust we also state that virtual domains are "beyond the scope of this howto" [20:37] sommer: no.. [20:37] sommer: not yet, hopefully, for karmic [20:37] "you must be this tall to use virtual domains" [20:37] ivoks, and we'll have to work on howto regarding to heartbeat :P [20:37] though, I admit, that sign is much lower than the "signals" sign of the same format [20:37] lamont: well, if we pull out DIT for ldap [20:37] lamont: heh, okay... I don't think they're mentioned at all in the current content [20:37] RoAkSoAx: :) [20:38] sommer: in the cases where we hand out answers that we know full well are going to screw someone stepping beyond the scope of the doc, we should at least put some caveat in there somewhere to say "uh, ignoring that here, kthx" [20:39] RoAkSoAx: sorry, i'll answer your questions in mail right now [20:39] lamont: sure, I'll add that to the list as well :) [20:40] because it's painful to have some noob point at a simplistic howto and say "BUT IT SAID TO DO THAT HERE" when they've clearly strayed into deep-and-shark-infested waters [20:40] ivoks, np :) I didn't even realized you were online until few secs ago.. i thought you were already sleeping [20:40] anyway, afk for a while [20:41] lamont: thanks, later on [20:41] sommer, btw.. would you like me to include the howto's on heartbeat in the bzr branch or in a wiki? [20:42] RoAkSoAx: I'm good with either... if you're not familiar with DocBook it'll probably be faster to create a wiki page [20:43] RoAkSoAx: we can always translate the wiki to DocBook [20:43] sommer, ok cool :) === Iceman_B1SSH is now known as Iceman_B|SSH [21:36] what does the "basic ubuntu server" package contain thats avaliable during the minimal cd installation? [21:36] New bug: #375657 in mailman (main) "Please add "-q" to logrotate.d script" [Undecided,New] https://launchpad.net/bugs/375657 [21:42] oluu: I would imagine pretty much nothing in the way of pre-installed services - just a minimal system to build from. [21:49] are you looking for this oluu? http://packages.ubuntu.com/jaunty/ubuntu-minimal [21:51] New bug: #375669 in libnss-ldap (universe) "auth.log boot cron[5014]: nss_ldap: could not search LDAP server - Server is unavailable" [Undecided,New] https://launchpad.net/bugs/375669 [21:55] mathiaz: I'm guessing that those packages which have been merged should be removed from the roadmap? [21:55] ajmitch: yes :) [21:56] * ajmitch got a couple in before the freeze hit too hard [21:57] mathiaz: you wouldn't happen to have a list of less-trivial merges that need done, would you? [21:58] ajmitch: merges.ubuntu.com? [21:58] ajmitch: or is that list too big? [21:59] yeah, I was looking at that, hard to know if people are working on things unless I go through & contact people for interesting packages [22:00] php5, there's one that should be fun [22:17] hello does anyone here use openvz