[00:06] kirkland: https://blueprints.launchpad.net/~mathiaz/+specs?role=registrant&searchtext=server-lucid [00:19] #cakephp [00:19] hello to all. have a question about bind. i am able to access my server over the web domain "mydomain.com" but "www.mydomain.com" wont work. it looks like i need to added a new entry to the bind 9 file db.mydomain.com [00:21] xperia2: did you add it? did you update the serial? did you reload? [00:22] RoyK: dont want to break the file as i dont know what exactly i have to put for a line in this file [00:22] will upload the file to pastebin so you can see it [00:23] RoyK: okay this here is the file with my entrys http://pastebin.com/d1d5e0401 [00:24] it would be great to know what i need to change so both version of the domain "www.mydomain.com" and "mydomain.com" works ! [00:24] xperia2: the serial at line 8 must be updated for the zone to be reloaded. the usual way to to set this is YYYYMMDDXX where XX is a counter [00:25] to add www, just add [00:25] www IN A 80.254.182.249 [00:25] or even better a cname [00:25] as in [00:25] WWW IN CNAME wificom.ch. [00:25] cname is an alias [00:26] RoyK: wooow very intersting ! [00:26] will change this now as you said ! [00:26] next time, look up the bind docs [00:27] also, set TTL to something lower [00:27] well i have allready the ubuntu wiki pages opened and looked but i dont wanted to break something. [00:27] it's set to 7 days [00:27] set it to 86400 (24 hours) or perhaps even 7200 (two hours( [00:27] s/.$/)/ [00:27] and the wikipages dont really tell what is needed to be done for resolving this problem. [00:28] http://www.bind9.net/manuals is a fine place to start [00:28] RoyK: does "IN" denote an "IN" record, or is that just some command's format for an A record? [00:28] 'internet' [00:28] as in ipv4 [00:28] RoyK: ahh okay :-) i look allways at the wiki pages ! [00:28] Right, thanks [00:30] Actually IN is a DNS record class. [00:30] RoyK: a specific question. if i want also to have subdomains like "www.sub1.mydomain.com" and also "sub1.mydomain.com" what is the better solution ? [00:30] It is the only one in use at the moment [00:31] $ORIGIN sub1.mydomain.com [00:31] www IN A x.x.x.x [00:31] in that order [00:31] or just www.sub1.mydomain.com. in a x.x.x.x [00:31] mark the period at the end of .com. [00:31] no i want to have about 10 sub domains [00:32] that means "don't add the origin defined above" [00:32] but i am wondering if i should use the cname entry [00:32] Wouldn't it be better for each machine to have a "real" name, like "zippy", and then CNAME www to zippy, and for the domain root (example.net) have something smart enough to redirect to www.example.net for port 80 and mail.example.net for port 25, and so on? [00:32] The latter being an MX, I guess [00:33] xperia2: this isn't really an ubuntu question, but keep it short, and I'll answer it. it's a bind question after all [00:33] twb: yes i want this for receiving mails on my server [00:33] create one A and then a bunch of CNAMEs to that [00:34] and create one A for the MX and reference to that from the other domains or subdomains [00:34] okay give me a moment for doing that ! [00:35] The point being that "example.net" isn't a host's name, it's a domain. If you later introduce a second host "griffy", you don't want to have to piss about in zippy:/etc/hosts and such during the migration [00:35] I didn't say he should define everything as the domain name [00:35] the mx points to the domain [00:36] everything else are hosts [00:36] even the domain name itself [00:36] you can even define the domain name as a cname to a host if you want to [00:37] as in [00:37] @ cname somehost [00:37] given $ORIGIN yourdom.com [00:38] but night [00:39] how i can install ubuntu server under windows? [00:39] Campulung: badly [00:39] Campulung: why do you want to do that? [00:39] because i want to use windows too [00:39] in same time [00:40] Campulung: why? [00:40] i have an office aplication who run only in windows [00:40] On a server? [00:40] no [00:40] A single host shouldn't be performing both server and workstation roles. [00:41] i can run ubuntu server under windows? [00:41] in same time? [00:42] Campulung: that is technically possible, but it is stupid. [00:42] RoyK: i have changed the file now. http://pastebin.com/d41283b30 however i am a litlle unsure about the mail entry for the subdomain gallery [00:42] i need ubuntu server for ehcp [00:44] What should I do? [00:44] Inherit a lot of money from a dead aunt, move to Jamaica and sit on the beach all day drinking [00:45] twb: nice one but even this dont make happy the whole life :-) [00:47] hey guys, I am running ntpd on a server, 9.04 / Jaunty. time is becoming further and further out of sync with the NTP servers quite fast. It seems like about every 10 seconds it becomes 1 second out of date. Does anyone know what would cause this? === rgreening_ is now known as rgreening [00:49] jetole: not running an ntp daemon [00:50] jetole: oh, if it's losing one-tenth of a second to the second, that's different [00:50] That's "buy a new motherboard" kind of territory. [00:50] twb, it is running and I am running the nagios check ntp check now but on the server in real time with the watch command and I am watching it [00:50] If you weren't running ntp, I would expect you to lose maybe a minute to the day. [00:51] twb: it's a VM and it's the only vm out of 1/2 dozen that this is occuring to [00:51] jetole: OK, VM's are a whole different ballpark [00:51] why? [00:51] jetole: which virtualization technology? VMware? [00:51] kvm [00:51] I don't know the right fix for kvm. [00:51] I don't think it is kvm [00:51] First of all try booting with notsc. [00:52] (Booting the VM, that is.) [00:52] why would this be the only vm running on this host [00:52] what is notsc? [00:52] jetole: I don't know. [00:52] notsc means "don't trust the hardware clock" [00:52] ok, let me try, it's a mail server so I can reboot it and proper mail servers will send again [00:53] I don't expect that to help much... [00:53] twb: do you know if the last line in my bind9 config file is right for receiving mails to my subdomain "gallery.mydomain.ch" ? http://pastebin.com/d41283b30 [00:53] hmm [00:55] jetole: for other VM technologies, particularly VMware, I have seen that running ntp inside the VM is "wrong", and you are instead supposed to use a funky bridge that makes it sync via the virtualization layer [00:55] Which in vmware is done with the vmware-tools kernel blob crap [00:55] twb, well perhaps but again, this is one among more then a half dozen running on the host server with this issue [00:56] twb: actually I think it can be done trusting the hardware clock with ntp turned off [00:56] Shrug [00:56] I would ask the KVM people if they know of this issue [00:56] well the hardware clock for a vm is the software clock of the host [00:56] yeah good idea [00:57] Sorry, that was what I was implying earlier, I should've said so outright [01:00] stgraber: LXC support should be enabled in libvirt in karmic [01:00] it's [01:00] though apparmor blocks it [01:00] and the userland tools for LXC were outdated [01:00] stgraber: oh? poke jdstrand about that [01:00] once I updated LXC, then libvirt is outdated but work is going on upstream [01:01] so for Lucid I have good hopes to have both the userland tools working correctly (already have that with a git snapshot in my PPA) and libvirt is being worked on to follow these lxc changes [01:03] apparmor isn't a big deal to fix, it's just about allowing libvirt to call two additional binaries IIRC though there's not much point in fixing that in Karmic as userland LXC doesn't work correctly [01:08] * stgraber currently has 3-4 users at the office running LXC on netbooks or atom-based thin clients with 10 contexts or so each ;) [01:08] ubottu: LXC [01:08] Sorry, I don't know anything about LXC [01:09] twb: Linux containers. http://www.stgraber.org/2009/11/06/lxc-containers-or-extremely-fast-virtualization [01:10] Is that a competitor for OpenVZ? [01:11] it may very well be its replacement as it's entirely included in the upstream kernel [01:11] Bloody awesome [01:11] yep [01:11] I hate openvz [01:11] I liked OpenVZ but not being able to run it on your laptop was kind of frustrating [01:11] where i can find unbuntu 32b ? [01:12] LXC's tools aren't as good as OpenVZ's but with the libvirt module we shouldn't need them anyway ;) [01:13] from where download ubuntu 32-bit ? [01:14] *ubuntu server [01:14] I don't like OpenVZ's tendency to hang the whole system when I mount -t aufs [01:16] Campulung, www,ubuntu.com [01:16] Campulung, www.ubuntu.com [01:17] here is only 64b version [01:17] i need 32-bit [01:18] look: http://www.ubuntu.com/getubuntu/download-server [01:19] Campulung40: there's an "alternative download options" link there which will show you the 32-bit version [01:21] tx ajmitch ... === erichammond1 is now known as erichammond [01:48] mathiaz: https://bugs.edge.launchpad.net/ubuntu/+source/eucalyptus/+bug/454405 [01:48] Launchpad bug 454405 in eucalyptus "the CC is returning incorrect networkIndex values on describeInstances" [Critical,Fix committed] [01:54] how do I configure what service runs at which runlevel? [01:54] I'm aware of the presence of /etc/rc.d script hooks, but I'm looking for a tool [01:55] mathiaz: http://www.google.com/intl/en/press/pressrel/20091110_free_airport_wifi_holiday.html [01:58] mathiaz: since you use virsh a lot, feel free to add helpful hints to: https://help.ubuntu.com/community/KVM/Virsh === erichammond1 is now known as erichammond [03:07] how do you refresh /etc/hosts in commandline [03:09] stgraber: please file bugs for LXC and libvirt. these have to be simple profiling bugs for libvirtd since only the qemu driver has guest protection [03:10] stgraber: I can then get lucid fixed at least, and get the changes pushed upstream [03:11] stgraber: though, if I have another libvirt SRU, I might just fix it in karmic too (the profiling bugs) [03:14] bug 480478 [03:14] Launchpad bug 480478 in libvirt "libvirt's apparmor profile doesn't allow execution of /usr/lib/libvirt/libvirt_lxc" [Undecided,New] https://launchpad.net/bugs/480478 [03:14] jdstrand: ^ [03:15] stgraber: thanks! [03:15] np [03:15] I think there is another binary or too-- I'll fix them [03:19] two [03:20] stgraber: btw, would 0.7.2 be good? I figure I'll do the merge sometime soon [03:21] New bug: #480478 in libvirt (main) "libvirt's apparmor profile doesn't allow execution of /usr/lib/libvirt/libvirt_lxc" [Medium,Triaged] https://launchpad.net/bugs/480478 [03:23] jdstrand: I'm not sure, that will include a whole lot of upstream changes in the lxc driver for libvirt but they did a whole lot more afterwards and changelog isn't very clear on that. [03:24] jdstrand: I see it's in unstable so I may just upload a backport to my PPA and do some tests with it tomorrow (if I find the time) [03:24] * stgraber would love to have some working to show at UDS (including libvirt support) [03:25] stgraber: I'll try to get the merge before UDS [03:25] I'm going to upstream a couple of other things too, so maybe we'll have 0.7.3 soon in lucid (though not UDS soon) [03:26] New bug: #480479 in backuppc (main) "package backuppc 3.1.0-4ubuntu1.1 failed to install/upgrade: subprocess post-installation script returned error exit status 2" [Undecided,New] https://launchpad.net/bugs/480479 [03:57] can anyone point me to a good site or help me setup this.. i want to make it so that, say user bob and user bill have different domains for the email address.. i have a few domains on this server... how can i do this. [03:58] smackd: tell postfix to accept email for both domains [03:58] smackd: then go to the pub and have a drink [04:02] lol [04:02] pub [04:02] too far [04:03] twb, the accepting part is set-up.. its when an email is sent [04:03] it only shows the main domain [04:04] On the envelope From, or the message From? [04:05] ok sorry.. forgot to say im using squirrelmail as the client for this [04:09] figured it out [04:10] * smackd kicks himself === jono_ is now known as jono [05:24] any one here know about secure file storage? I need to create a client portal solution for accountants to share sensitive data with their clients and visa versa. I would be willing to pay a reward for the right solution [05:28] Chris_Davis: That's a very broad question; does your specification have all the details about what "secure" means in this context, and whether you care about secure *transmission* as well as secure *storage* of the data? What legal issues are there that have a bearing on this in your country/jurisdiction? To get a useful answer you will probably need to point people at the detailed RFP... [05:34] it's for the USA and for both transfer and storage [05:34] there are HIPPA guidelines that people often talk about [05:52] Chris_Davis: There are several open source "web file manager" apps that could work for you, appropriately secured over SSL/TLS. Things like http://boxroom.rubyforge.org/ [05:52] secure enough to store client SSN#s? [05:54] Chris_Davis: If you're seriously concerned about security, before you have stuff on the actual internet, you really, really need to have some serious design expertise and security review. It's not a problem you'll solve with casual IRC chat. [05:55] ScottK, true, but if it isn't worth it and will cost too much in legal fees, etc to do [05:55] I can save my time and money before going to a lawyer and paying them $200/hour too look up what a GB is [05:59] Chris_Davis: You are wasting other people's time if you ask questions like "secure enough to store client SSN#s?" before you define your requirements. We're back to "what secure means in this context" ... you have not provided that definition... define "secure enough"... secure from what class of attackers with what resources for what time period, to start with... overall you need to define your project requirements [05:59] in more depth. Write a clear RFP, or point us to the existing one. [06:00] secure period, can't access the data period [06:00] I need a way for cpas to share data with clients and only that client [06:00] with audit trail [06:00] secure period implies you cannot do it on the Internet period. [06:01] Have them ship data by secure courier in briefcases chained to the bonded couriers wrist, etc. [06:02] If you know so little about security you think that "secure period" is feasible among networked computers, you are totally unqualified for this kind of work. [06:03] it isn't a line of work for me [06:03] I just do project management [06:03] I wanted to know what was possible [06:03] and get a trail to follow [06:03] luckily, another channel was kind enough to just mention PKI, and I've been happily reading ever since [06:07] Chris_Davis: Not "Secure. Period.", BTW. Anyone who is giving you security answers in absolultes is either leading you on or doesn't know what they're talking about. [06:08] Seriously, you need to have someone figure this out who understands it. If you're actually worried about security, do it yourself makes about as much sense as do it yourself brain surgery. [06:08] I'm not trying to do it myself, but others have had to work out similar problems [06:08] at the moment they are agreeing that PKI is the "best" way [06:09] In general, that's true. [06:09] right, I'm not expecting to get a fool-proof answer [06:09] but this isn't my area at all [06:10] and before I post a job saying "something to do with file storage" [06:10] I'd prefer to narrow it down [06:10] in this case to PKI [06:10] then I can learn advantages and disadvantages, etc [06:20] New bug: #478333 in mysql-dfsg-5.0 (main) "package mysql-server-5.0 (not installed) failed to install/upgrade: le sous-processus nouveau script pre-installation a retourn? une erreur de sortie d'?tat 1" [Low,Invalid] https://launchpad.net/bugs/478333 [06:21] Ctrl+z pause & put in background a task, how to just put a task in background and get it back ? [06:27] maxagaz, I belive it is: fg NUMBER [06:27] maxagaz, the number was shown right after the CTRL+Z [06:28] maxagaz: ctrl-Z and then fg (to get it back in the foreground) or bg (to get it running in the background. Default jobspec is the current job, which is the one you just created with ctrl-Z, so there should be no need to specify a number for this. help fg [06:37] HI there, I have a question..i just install ldap in ubuntu server but i cant find the file sldap.conf, Is no was installed.Somebody know what i can resolve this problem? [06:39] roxy007: WHat are you trying to do, and why do you think there *should* be a file called sldap.conf ?? [06:50] hi jmarsden, yes ...i install before in another servers and always was this file and i needed also to configurate this file [06:51] i want install a ldap server [06:51] sldap.conf? or slapd.conf? [06:52] Maybe http://www.howtoforge.com/installing-and-configuring-openldap-on-ubuntu-intrepid-ibex will give you some ideas? [06:53] I think you are looking for slapd.conf, you asked for sldap.conf -- those are different things! [06:57] yes ... [06:57] thats rght [06:57] do you know how i can resolve it? [06:58] Did you read the HOWTO? It includes a sample slapd.conf -- so that would be one way. [06:59] Also see https://help.ubuntu.com/9.10/serverguide/C/openldap-server.html [07:03] roxy007: Note that the slapd.conf approach is old and no longer considered a good way to configure slapd, as I understand it. You can and should be using the cn=config stuff instead. That is why there is no slapd.conf provided by default. [07:05] ah..ok, i undestand now..thanks a lot [07:05] No problem. [07:17] can CUPS be used without a GUI [07:19] asentaja: Yes. [08:26] hi [08:26] any one used ntop ? [09:13] I want to format and repartition my external hard drive. I do not want to use NTFS or FAT32 but I still want windows to be able to access files off of it. Is this possible? And if so, what file system do I use? [09:22] if you format it with ext3/4, you would have to get that utility for windows [09:22] which is slooow [09:35] New bug: #480332 in munin (universe) "package munin-node 1.2.6-13ubuntu3 failed to install/upgrade: subprocess installed post-installation script killed by signal (Interrupt)" [Undecided,Invalid] https://launchpad.net/bugs/480332 [09:46] what command do I use to update my server via ssh terminal? [09:47] apt-get update [09:47] or sudo apt-get update [09:47] if not sudo su root [09:47] and that will actually download and install any updates? [09:47] no [09:47] sudo apt-get upgrade [09:47] will install em [09:48] update just gets the list of files [09:48] perfect, thank you. [09:49] np. [09:54] <_ruben> upgrade wont install new packages though, which is the case with kernel upgrades for instance, use apt-get dist-upgrade instead to cover those as well [09:58] _ruben, if I use sudo apt-get dist-upgrade will that upgrade me to 9.10 from 9.04? [10:00] well, sleep time it is. Goodnight [10:06] no [10:06] do-release-upgrade will [10:14] <_ruben> crohakon: it would only do so if you'd had altered you sources.list as well [10:19] <\sh> moins [10:20] I have an obscure error when trying to install dhcp3-server on an arm board [10:20] Can't exec "/tmp/dhcp3-server.config.23361": Permission denied at /usr/share/perl/5.10/IPC/Open3.pm line 168. [10:20] open2: exec of /tmp/dhcp3-server.config.23361 configure failed at /usr/share/perl5/Debconf/ConfModule.pm line 59 [10:20] It then fails to preconfigure, and doesnt start, any ideas? [10:27] Ahh i think I worked out the problem, fstab had /tmp mounted no exec === bigbrovar_ is now known as bigbrovar [10:52] * soren goes to lunch [10:56] how to uninstal ubuntu desktop from server? [10:57] apt-get remove ubuntu-desktop [10:57] And than tons of packages that can be removed too [10:57] probably with 'apt-get autoremove' [10:58] Reading package lists... Done [10:59] Package ubuntu-desktop is not installed, so not removed [10:59] but is instaled [11:00] Well, no. It isn't :) [11:01] but i have the grafic interface [11:01] and all the options from desktop [11:03] How did you install that than? [11:05] apt-get install ubuntu-desktop [11:13] ok [11:13] So you probably de-installed that allready === Jare__ is now known as Jare [11:28] is my printer device URI that specified in lpinfo -v [11:29] ? === mrchrisadams_ is now known as mrchrisadams [12:04] how can i find out my printer URI [12:04] to anyone who uses LDAP: how can I change the cn=admin,dc=domain,dc=com password in a RTC system? [12:04] is there any attribute (like the olcRootDN) that I can use to specify a password [12:04] ? === Blizzerand_ is now known as Blizzerand [12:15] My mysql server fails to start. I somehow messed up the database of the first install, purged it and cleaned all traces of it by hand. But now it still fails to start and even to install the package using aptitude. What can it be? [12:17] nyk2005: What kind of errors are you getting? [12:17] soren: just [fail] [12:18] soren: nothing in the logs [12:18] You said you can't even install it using aptitude. [12:18] How does that fail? [12:20] and during package install: "invoke-rc.d: initscript mysql, action "start" failed" [12:23] Ok, so the package install, but fails to run.. Hm. When you say you cleaned all traces of it... What exactly did you do? [12:24] find / | grep mysql [12:24] and remove all that [12:25] when I just run /usr/bin/mysqld_safe, it starts then says "mysqld_safe mysqld from pid file /var/run/mysqld/mysqld.pid ended" [12:25] and in /var/log/messages "Nov 11 13:23:18 biodata2 kernel: [46908.006040] type=1503 audit(1257942198.520:839): operation="open" pid=22153 parent=22046 profile="/usr/sbin/mysqld" requested_mask="r::" denied_mask="r::" fsuid=0 ouid=0 name="/sys/devices/system/cpu/"" [12:26] what kind of message is this? [12:26] Erk.. [12:26] Did "find / | grep mysql" show anything in /var/lib/dpkg/info ? [12:26] What version of Ubuntu is this? [12:26] nyk2005: that's probably apparmor [12:27] i had the same messages with bind [12:27] It is apparmour, yes. [12:27] ah because I also deleted the profile [12:27] Yeah. Now nothing will work. [12:28] shit... so just aptitude remove/install to get it... [12:28] purge. [12:28] Not just remove. [12:28] but I can't because it then tries to start mysql.. :) [12:28] And you still haven't answered my question. [12:28] 13:26:27 < soren> Did "find / | grep mysql" show anything in /var/lib/dpkg/info ? [12:28] ...and did you delete any of that? [12:29] ..although it trying to start mysql when you remove it suggests that at least some of that stuff is still around. [12:29] soren: Isn't "find /var/lib/dpkg/info | grep mysql" more efficient? :) [12:29] Jeeves_: Depends on what you're trying to do. [12:29] soren: no I didn't touch any of the OS stuff [12:29] nyk2005: Except apparmour. [12:29] how do i find out my device URI [12:30] soren: ah yes.. [12:30] soren: because it has "mysql" in it and resides in /etc... which made it susceptible.. [12:30] soren: so i just replace the file from here:soren [12:30] soren: I mean http://ubuntuforums.org/showthread.php?t=849274 [12:31] nyk2005: how do i find out my device URI [12:31] nyk2005: what device? [12:31] printer [12:32] printer has an uri? why not use cups? [12:32] asentaja: Why do you need it? [12:32] adding printer with lpadmin [12:32] nyk2005: I don't know. [12:32] nyk2005: 13:26:50 < soren> What version of Ubuntu is this? [12:33] lpadmin printer1 -v device URI [12:33] server 9.04 [12:33] or 9.10 [12:34] I am using cups [12:34] soren: actually /etc/apparmor.d/usr.sbin.mysqld exists, it just has a parsing error: "AppArmor parser error in /etc/apparmor.d/usr.sbin.mysqld at line 352: Could not open '(null)" [12:35] I have a question about sieve scripts: Is there an "#include xxxx" so I can organize the code in several files? [12:35] but that file never has that many lines.. [12:35] mok0: Which sieve implementation are we talking about? [12:36] nyk2005: I can't help you when you don't answer my questions. [12:36] soren: the one in Ubuntu's mail stack, I believe it's CMU's [12:36] mok0: Err.. [12:36] mok0: Dovecot, you mean? [12:36] asentaja: if your cups is configured from the webinterface it's enough (at least for me) to access it by name. [12:36] soren: yes [12:36] mok0: Which version? [12:37] soren: karmic :-) [12:37] it dosent have a web interface [12:37] im using command line [12:37] "lpadmin" command === mrchrisadams_ is now known as mrchrisadams [12:37] soren: I've been looking for a while, and I'm afraid everything goes into one huge file, which sux IMO [12:38] mok0: http://wiki.dovecot.org/LDA/Sieve#Include_scripts [12:38] soren: my hero! :-) [12:39] :) [12:39] dovecot in Karmic is 1.1.11. Bear that in mind when reading that page. [12:39] soren: you mean, it's the newest version I guess [12:40] morning [12:40] mok0: Well, there's a 1.2 dev version. [12:40] soren: ah I see [12:40] mok0: If that page talks about "CMU sieve" vs. "Dovecot Sieve", "CMU sieve" is the one you should pay attention to. [12:40] soren: ok, thanks [12:41] Dovecot uses CMU Sieve in 1.0 and 1.1. [12:41] 1.2 can use either. [12:41] I dont know how it works adding a printer threw web interface, but i need to use the command line tool lpadmin for that, and it needs the URI for it [12:42] * mok0 hurries off to translate old procmail recipes to sieve-format :-) [12:43] * soren makes a coffee run [12:43] asentaja goes fuc***g mad [12:46] asentaja: it's very comforable in the webinterface, all done by wizards.. :) [12:46] just cupsserver:631 [12:51] do you need GUI for the web interface i think yes? [12:51] see ubuntu server is only command line [12:52] I think the URI is something like /dev/usb* [12:52] but not really sure [12:53] soren, can you use variables inside a sieve script? [12:53] soren: n/m [12:55] * MenZa slides soren a steaming hot mug of coffee. [12:55] * atomic_1 intercepts it [12:55] :O [13:02] :( [13:04] please why wont anyone help me [13:06] Because we have no clue about printing, I suppose. [13:11] New bug: #479900 in nut (universe) "Nut upgrade overwrites /etc/nut/nut.conf" [Medium,Triaged] https://launchpad.net/bugs/479900 === RoyK is now known as RoyK_oink [13:15] * MenZa virtualises soren [13:24] <\sh> soren, just a reminder, all ideas of keybuk weren't successful [13:26] How about mine? [13:28] <\sh> I'll test it just now... [13:30] <\sh> soren, you meant just exec ifup -a in /etc/init/network-interface.conf [13:30] Yes. [13:30] It will have a side effect, though. [13:30] <\sh> I think this is also too early for that... [13:31] How do I incorporate spam filtering in Ubuntu's mail stack? [13:31] \sh: What do you mean? [13:31] <\sh> soren, when I read the file correctly, it will be always triggered when udev adds an interface... [13:32] Exactly. [13:32] <\sh> soren, which means, an ifup -a will fail the same way, because not all interfaces are added at some time... [13:32] Right.. [13:32] So? [13:32] It will also be called when all the necessary interfaces are called, at which point it will succeed. [13:32] <\sh> the bonding e.g. will be available, but without any interfaces added to the bond...(this happens as well with keybuk ideas) [13:32] Err... when all the necessary interfaces have been added, I mean. [13:33] The bond interface doesn't fail to be configured if the nic's are not available? [13:33] Oh, ffs.. [13:33] <\sh> nope...the bond0 interface is there... [13:33] <\sh> but without the needed NICs it's useless, and e.g. an vlan interface which is using the vlan_raw_device bond0 doesn't work [13:34] I understand. [13:34] I just expected ifenslave's ifup script to bail out if the listed interfaces were unavailable.. [13:34] but it doesn't. How "clever". [13:35] <\sh> well, tbh, bonding interfaces are somewhat sick...it needs to be auto bond0 and you need to set something like 0.0.0.1 as address to be used as e.g. vlan_raw_device [13:36] Oh! [13:36] I think I may have the solution. [13:38] \sh: Try this: /usr/share/doc/ifenslave-2.6/examples/two_hotplug_ethernet [13:39] <\sh> oh [13:39] <\sh> there is a bug in this example ;) [13:39] What? [13:39] ah, i see. [13:39] <\sh> ;) [13:40] There may actually be two.. [13:40] Let me check. [13:40] Ah, no, it's probably only the one. [13:43] <\sh> soren: this will be fun with a setup like this: http://www.sourcecode.de/content/network-setup-freaks-me-out [13:46] Ok anyone help me out? I need to View and Edit PDF files in my GUI. I can view new files fine. As soon as I save them they become unviewable? What is my solution? [13:47] \sh: What does eth0 and eth1 do all by themselves? [13:47] <\sh> soren, nothing :) [13:48] garymc: Wrong channel. [13:48] <\sh> soren, they are just the way out of the server ... [13:48] yeah there is no right channel [13:48] <\sh> garymc, #ubuntu is the right one [13:48] garymc: #ubuntu [13:48] No right channel does not mean that any channel is the right channel. [13:48] no help there [13:48] <\sh> soren, the hotplug didn't work [13:49] garymc: That does not mean that this magically becomes the right channel. [13:49] \sh: You changed /etc/init/network-interface.conf back to the way it was to begin with, right? [13:49] <\sh> soren, yepp [13:50] <\sh> soren, /etc/init.d/networking restart after login works as expected... [13:50] <\sh> just not during boot up time [13:50] Can you work out why it fails? [13:51] <\sh> soren, can I somehow pass --debug to upstart and tell upstart to log into a file? just working from an ILO so I need some logfile [13:51] <\sh> soren, (--debug via kernel-commandline is known) [13:52] Hey, i don't think i can make today's meeting. However, i've posted some comments regarding spamassassin which was ne of my action'sfrom last week. [13:53] (to ubuntu-server ml) [13:53] \sh: Don't know. [13:54] smoser: ping [13:54] <\sh> let's see if someone from #u-d has a clue [13:55] \sh: surely you can direct the verbosity to the ttyS of iLo? [13:57] <\sh> Daviey, you mean direct the output of kernel + upstart through ttyS ilo port 3002? ;-) [13:58] \sh: i mean, the fact it's iLo makes no differnet, surely. It's just a (serial) console. [13:58] difference* [13:59] I assume you have ipmi aswell? [13:59] <\sh> oh well...playing with our firewall.... [14:00] well if you cna get shell into the network, just run the ipmitools on a box within the network- and ssh to that :) [14:00] * Daviey has to run, ttfn [14:00] meeting -> #ubuntu-meeting [14:01] <\sh> Daviey, this is hard...:) because our ilo network is not reachable from within the production network... [14:03] ttx, here [14:04] smoser: after the meeting [14:16] <\sh> grmpf [14:22] Greetings and salutations! [14:57] Hi all, is it possible to run a cluster head as a node too? I.e. is it possible to run a cloud on one machine (I'm testing) [14:58] Jeez these sieve scripts ignore everything I try to do... :-( [15:01] ttx: so the samba sru === aubre is now known as aubre_afk [15:02] ttx: i was going to do an samba karmic SRU tomorrow but I was going to add that libsmbclient nautilus bug as well and go through the changelog and see if we can cherrypick any others === RoyK_oink is now known as RoyK [15:11] zul: ok [15:35] mathiaz, kirkland: see https://bugs.launchpad.net/ubuntu-release-notes/+bug/458904/comments/32 [15:35] Launchpad bug 458904 in eucalyptus "When installing a node, euca_find_cluster fails to locate the cluster controller if instances are running" [High,Fix committed] [15:37] ttx: hm - right [15:37] ttx: has the avahi-daemon update been already published? [15:37] mathiaz: yes [15:37] ttx: :( [15:38] (in -proposed) [15:38] mathiaz: though it should be fixed in eucalyptus, no ? [15:39] ttx: ha - in -proposed that's good then [15:39] ttx: well - technically yes [15:39] ttx: however if the avahi update is pushed at the same time as eucalyptus to -updates [15:39] ttx: it will be restarted [15:40] hmmm [15:40] mathiaz: I don't know how acceptable that would be :) [15:40] mathiaz: could you comment on the bug ? [15:40] ttx: right - I agree that's is a bit stretched [15:42] mathiaz: but it's true I tested a 7.1 -> 7.2 eucalyptus update alone [15:42] since avahi update was picked up already [15:43] ttx: right - I've added a comment [15:43] ttx: and I'll do the test [15:44] ttx: with both avahi and eucalyptus published at the same time [15:44] mathiaz: maybe that's what kirkland tested. [15:45] ttx: may be. [15:49] mathiaz: not sure how we can validate the fix for bug 454405 [15:49] Launchpad bug 454405 in eucalyptus "the CC is returning incorrect networkIndex values on describeInstances" [Critical,Fix committed] https://launchpad.net/bugs/454405 [15:59] ttx: yeah - me neither [15:59] ttx: upstream said so? [16:00] mathiaz: well, they could help in validating the fix [16:01] since we lack a clear testcase. [16:22] anyone there? [16:22] can I ask a question === dendro-afk is now known as dendrobates [16:22] !ask | googa [16:22] googa: Please don't ask to ask a question, simply ask the question (all on ONE line and in the channel, so that others can read and follow it easily). If anyone knows the answer they will most likely reply. :-) [16:23] How do clients use printers on a print server lets say working CUPS? === rgreening_ is now known as rgreening [16:48] * soren goes to dinner === mrchrisadams_ is now known as mrchrisadams === robbiew is now known as robbiew_ [17:28] how can I select a specific mirror at install time? [17:28] or specify one local (apt-proxy) mirror [17:29] <\sh> anyone online with similar problems as described in #446031 ? [17:31] \sh: I have almost the same setup, only with dhcp, and I have no problems [17:32] <\sh> cemc, running karmic? [17:32] \sh: yes [17:33] <\sh> cemc, can you paste your /e/n/i ? [17:33] http://pastebin.ubuntu.com/316127/ [17:34] looks similar, only with dhcp [17:35] <\sh> cemc, can you try with a static ip ? [17:37] \sh: not right now, gf is working on the comp. but I have a laptop, also with Karmic, and I can try there. I don't have bridge there right now, but I had a couple of days ago, and it also worked fine, just gimme a sec [17:38] <\sh> cemc, no NM on it? [17:42] \sh: http://pastebin.ubuntu.com/316135/ - this is the config right now, just copy-pasted from the bug and changed the IPs [17:42] <\sh> and it works? [17:42] \sh: I have NM on it, but it doesn't manage anything because of the auto stuff in /e/n/i [17:42] yes, it works, I'm writing from it right now [17:43] it's a Karmic upgraded from Jaunty. but the other one is a clean Karmic install [17:43] both 64bit [17:43] <\sh> cemc, I don't use bridging but bonding I had the very same problem as lars, does bridging needs a manual eth0 device ? [17:44] \sh: that's what I have in /e/n/i, it says manual for eth0. and if I do ifconfig, I can see eth0, it's up, with no IP address [17:45] I guess it needs to be up for the network to work, even if through bridge [17:47] <\sh> cemc, hmm..strange... [17:49] \sh: did you see the fix in the bug ? [17:50] <\sh> cemc, it was me with the fix ;) [17:50] ah, ok [17:50] <\sh> I debugged now for 2 days with soren because of this...and this works as expected. [17:51] what created /var/run/network anyway? [17:51] I have that on both my machines [17:51] <\sh> I'm just trying to test this out, or if something else might be happening [17:51] <\sh> /var/run is a tmpfs...and ifup checks in /var/run/network/ the interface states === robbiew is now known as robbiew_ [17:52] I see, but what created /var/run/network ? [17:52] <\sh> /etc/init/network-interface.conf creates this directory inside the pre-start script, but somehow when /etc/init/networking.conf is processed, this directory is not there and the call to ifup -a failes [17:52] <\sh> cemc: in your case, eventually /etc/init/network-interface.conf or some magic network-manager script [17:55] \sh: I see... anything else I can do? [17:56] <\sh> cemc, nope...thx for your help :) I'm doing some more testing tomorrow, when I inject the new ifupdown package into my auto-dist-upgrading puppet recipe [17:58] \sh: sure, np [18:12] New bug: #480820 in mysql-dfsg-5.0 (main) "mysql-server-5.0 package upgrade problem" [Undecided,New] https://launchpad.net/bugs/480820 [18:19] <\sh> hmm...puppetd is still not started correctly, doesn't write the pid file, so stopping won't work either === lamont is now known as lamont` === lamont` is now known as lamont === dendrobates is now known as dendro-afk [18:54] i [18:58] if I want alias rm='rm -i' system-wide, where do I put it? I want it to be set for everbody, local user, login through ssh, etc [18:59] yo yo, i'm running libvirt-qemu and I'm trying to get usb passthru working, I've edited my apparmor to enable it and tried everything else i can find on google but it's still not working, after enable with apparmor and adding a USB physical device with virt-manager, I get the error "Error starting domain: internal error Timed out while reading monitor startup output".... anyone have any ideas how to get this to work? [19:00] generaldisarray: you might want to try in #ubuntu-virt [19:01] doh thanks cemc i didnt realize there was a diff. group will do [19:01] np [19:03] cemc: /etc/bashrc [19:03] /etc/.bashrc [19:04] thx [19:06] err I thought /home//.bashrc [19:06] didn't realize you could put on in /etc/ [19:06] so /etc/bashrc is called each time someone logs in? [19:32] I have good download but terrible (less than 3kb/s) upload speed from my server in my local network, any tips? For more enfo please check this thread in the Ubuntu Forums http://ubuntuforums.org/showthread.php?t=1320471 thanks! [19:41] what app are you uploading with [19:41] and have you tried uploading with another computer on your network? [19:52] cemc: if you're still there the file you want is /etc/profile [19:53] cemc: /etc/bashrc doesn't affect login shells [19:53] cemc: unless its referenced by each users ~/.bashrc === NCommander is now known as Guest28444 [19:56] New bug: #480885 in eucalyptus (main) "Eucalyptus avahi publication jobs failed to restart if avahi-daemon is restarted" [Low,In progress] https://launchpad.net/bugs/480885 [20:00] Hypnoz: I see [20:00] (was a netsplit) [20:03] hi.. im trying to setup a mailserver and need to know how to let users add email accounts without having shell access [20:05] smackd: you could try installing postfix with virtual users and postfixadmin (web interface). [20:05] cool. i have postfix installed. is virtual users a config option or another package? [20:06] config option..to provide mail to non-system's user account [20:08] smackd: well, that's a little bit more complicated, you may need to read up on some stuff. try this: https://help.ubuntu.com/community/PostfixCompleteVirtualMailSystemHowto [20:09] but this is not a quick "i need it now" thing :) [20:19] how would i ssh into a remote server, and run a command and be able to close the tunnel? is this what a job is? === dendro-afk is now known as dendrobates === Guest28444 is now known as NCommander === NCommander is now known as Guest52407 === Guest52407 is now known as NCommander [20:28] <\sh> marks256, ssh user@server "/bin/ls" ? [20:28] \sh, what does the material in quotes do? [20:29] <\sh> marks256, it's example to remotely get a directory listing (ls) actually you don't need the quotes [20:30] what i want to do is to ssh into a remote server to do some video rendering, but it's going to take about 16 hours, so i'd like to be able to turn my computer off, and not have to leave the ssh tunnel open on my computer [20:30] <\sh> marks256, oh...for that "screen" is the right tool for you [20:30] !screen [20:30] Screen is a window manager for terminal sessions, also useful over SSH. The 'screen-profiles' package provides very useful additional utilities. See https://help.ubuntu.com/community/Screen [20:31] Hm, /me wonders if we should update that for byobu [20:31] <\sh> marks256, ssh to the server, start screen , start your job , do ctrl-A D for detaching screen, logout and come back after 16 hours, ssh to the server, start "screen -r" et voila === stgraber_ is now known as stgraber [20:31] <\sh> ok.../me is gone for today...cu tomorrow [20:32] oh [20:32] ok thanks! [20:32] i'll give that a shot [20:37] * JanC <3 screen & byobu ☺ === dendrobates is now known as dendro-afk [20:56] hi [20:56] hi [20:58] Which sendmail is faster in ubuntu server? [20:59] in desktop I used exim, but idk if sendmail is faster in server [21:03] hello to all. i have a strange problem with apache vhosting. this here is my default file in the folder sites-availble http://pastebin.com/d1d584f0 [21:04] xperia2: ill look at mine now [21:04] from all vhosts only the last one "gallery" works ! all other give me 403 Forbidden [21:04] You don't have permission to access / on this server. [21:05] Mack_: thanks ! [21:05] xperia2: do you have index.html in them all? [21:05] yes ! [21:06] cuz thats the error my server gets w/o one === dendro-afk is now known as dendrobates [21:06] because the directory listings are off [21:06] Mack_: We tend to prefer Postfix. [21:06] hmmm [21:06] ScottK: hmm, hows the speed on that? Because using desktop, the built in sendmail took about 10 minutes to send an email [21:07] but wouldnt let the page load until it was sent [21:07] normally it works if i have only one vhost ! [21:07] Mack_: If it's that slow that's a sign of some other problem. [21:08] Mack_: It looks like Postfix on my laptop usually takes between half a second and a second to send a mail. [21:09] i will backup now the default file and leave only one vhost for prooving that it works ! [21:09] ScottK: well this was on my serverusing ubuntu desktop, and I'm switching that soon [21:09] eh, postfix should be able to send mails in a fraction of a second ☺ [21:10] JanC: Yes, usually DNS lookups are the major bottleneck [21:10] I was just going to say: "maybe a second if you do all sort of DNS lookups " [21:10] ;) [21:10] certainly not a minute [21:11] would be bad for a real ISP if they get 100 mails / second ;) [21:12] xperia2: have u tried all the vhosts independently? [21:12] JanC: yeah, thats why I had a problem with it :P [21:23] whats the argument for password in mysql? [21:23] -u is user [21:23] but whats for password? [21:23] -p [21:23] see mysql --help ;) [21:30] Mack_: just wanted to say you that my vhosting works now for at least three websites. you was right about the empty folders. [21:30] the only thing I need to figure out is why directory indexing doesnt always enable like it should [21:31] what me makes scary a litlle is that my selfwritten php scripts in this folders have disapeared ! [21:32] the websites in this folder worked great. i am asking me if the wordpress blog software coudl be responisble for this erase ? [21:35] kirkland may I ask who is doing UDS session scheduling for you guys? I see that some of the sessions are not on the server track. [21:35] I could fix that for you if you want [21:36] rickspencer3: mdz is doing it [21:42] mathiaz shall I track them for you guys? [21:43] * rickspencer3 does [21:43] mdz ^ [21:43] rickspencer3: sure - if mdz is fine with that [21:43] rickspencer3: neither ttx nor kirkland nor me can schedule anything [21:48] hello all [21:48] can someone tell me is there a wiki or a guide to install dtc on ubuntu server [21:49] hi guys...im looking for some help with file premissions [21:50] Im setting up a secure samba share, im using a 700 file permission, however on login i dont have the correct permissions [21:50] any one know what i may be doing wrong ? [21:55] does samba need to have a certain file permission set to be able to log in ? [21:55] that would seem rather illogical [21:58] porn http://upload.wikimedia.org/wikipedia/commons/d/df/Body_painting.JPG [21:58] hey thats suse [21:58] not ubuntu [21:58] :) [21:58] * dru shakes his fits like a mad man [21:59] kane_: hmm, I tried -p and it didn't work :/ [21:59] mack: 'mysql -u username -p database' and then you'll be prompted for the db password [21:59] OH its -p= [21:59] i promise that works :) [21:59] yeah [21:59] I did -p pass [21:59] well I'm using mysqldump [22:00] but its the same syntax for password [22:02] rickspencer3: if you're looking into scheduling session for the server track, there a couple of them related to puppet [22:02] rickspencer3: they have puppet in their name [22:02] rickspencer3: upstream will only be at UDS on Monday and Tuesday [22:03] rickspencer3: so if the schedule could take this into account that would be very helpful [22:03] rickspencer3: upstream has been subscribed and marked as essential [22:03] rickspencer3_: I'm not sure that he has entered his attendance dates yet [22:09] mathiaz, I'm not scheduling [22:10] (for server) [22:10] but if you need some help adjusting a session at any time, please feel free to ask me [22:10] rickspencer3_: ok [22:11] mathiaz, is there one I can help with now? [22:12] rickspencer3_: not really - AFAICT no sessions have been scheduled up to now [22:12] rickspencer3_: hm [22:12] rickspencer3_: actually no [22:13] using a real client now [22:19] can anyone recommend a good (affordable) model server? i've been looking at HP Proliant DL585's lately. They look pretty good. When i say affordable i mean less than $400. I need to do some pretty CPU/Memory intensive rendering [22:19] i would recommend building one from scratch with that budget [22:20] that's what i was thinking. I figured i could build a quad dual core system with 16gb of ram for less than a grand [22:29] I'm trying to run postfix with virtual users using mysql but I keep getting this error [22:29] Nov 10 21:43:08 Directories postfix/trivial-rewrite[12406]: warning: connect to mysql server localhost: Can't connect to local MySQL server through socket '/var/run/mysqld/mysqld.sock' (2) [22:30] I can get results when I use postmap to run the config file but when postfix is actually handling email it can't connect [22:30] ninnypants: is your mysql running, and is it configured to listen on a local socket? [22:31] it's running but I'm not sure about listening to a local socket how do I check that? [22:32] let me check the precise config directive. in the meantime, check if the file /var/run/mysqld/mysqld.sock even exists [22:32] Also postfix is, by default, in a chroot. The socket needs to exist inside the chroot [22:33] ah, very good point [22:33] the sock exists I've already checked that [22:33] then what ScottK said ^ [22:35] ninnypants: this may be topical for you as well: http://ubuntuforums.org/showthread.php?t=251119 [22:36] I'll check that thanks. also how do I know if something exists inside chroot? [22:38] chroot to that dir, or cd to the dir and prefix it to the path. ie if your chroot is in /my/chroot, then it's /my/chroot/var/run/mysql/mysqld.sock [22:38] ninnypants: however, looking at the forum post, it looks like the easy solution is to switch postifx to connect to mysql over tcp rather than unix socket [22:38] and then all this headache goes away [22:39] That's generally the solution. [22:41] ok thanks [23:30] Anyone know NFS permissions well> [23:30] ? [23:33] has anyone installed 9.10 server on a dell poweredge R805? it has a dell sas 6 controller in raid1. i can install onto the disk but cant boot off it (grub it says no device found). [23:33] I am having difficulty in understanding how to share a folder with two clients [23:34] maybe i need to install a kernel mod in grub? [23:34] cant seem to find anything online about it, only that REL5 and suse are supported [23:35] How do I make a server update packages that were held back after I do a sudo apt-get upgrade? [23:36] PaliPals: sudo apt-get dist-upgrade usually does it [23:36] I am on LTS and don't want to distro upgrade? [23:37] That doesn't change the release you are running. It just allows new packages to be pulled in if needed [23:37] Ah ok. That makes sense. Thanks === dendrobates is now known as dendro-afk [23:38] So if I use sudo apt-get dist-update and apt-get dist-upgrade I shouldn't get held back packages anymore? [23:40] Also, with NFS I have the server working and it mounts the shares fine. The issue is that when one client puts files on the server, the other client does not have perms to them. [23:40] And vice versa. [23:40] So, how do I get the perms right so they both can access? [23:43] anyone have had problems with apache after upgrade ? mine just don't respond anymore, it seems to be running, but it don't do jack no more except using alot of cycles. apache2ctl fullstatus etc just hangs. it's just wierd. [23:44] PaliPals: No such thing as dist-update. Just update. [23:44] K [23:44] Thanks ScottK [23:45] Any advice on my NFS post at all. I know it has something to do with groups but I am just missing the idea or something. [23:46] i can't find a newer version of rtorrent for 8.04 in backports, but this is my first time trying to use backports. am I missing something? I don't see it in there at all. [23:48] ok, new development. [23:49] when i boot the OS, it says it cant find the root device and there's no /dev/mapper/mybox etc.. and kicks me to busybox, but if exit busybox, then everything works. i've had some .. timing issues before. this seems like one of those [23:50] verbal: Add rootdelay=90 to the boot parameters and see if that works. [23:50] If it does, then you can try smaller values. [23:51] ScottK: thanks i'll try that [23:53] i can't find a newer version of rtorrent for 8.04 in backports, but this is my first time trying to use backports. am I missing something? I don't see it in there at all. [23:55] Lownin: Looks like it was never backported, but there is very little different in newer releases anyway [23:55] I'm having some trouble with watch directories. I don't know if it's a bug so I was trying to rule that out. [23:58] Thanks for the info, ScottK [23:59] mathiaz: Do you know if we have a spec for server based calendaring solution? [23:59] ScottK: hi! I think zul registered one [23:59] OK. Thanks. [23:59] zul: ^^ what's it called?