=== slank is now known as slank_away [00:15] in a preseed file, how do I force an unbuntu server install to configure multiple interfaces? [00:16] i am pxe-booting from eth0 and need eth1 configured by our corp dhcp server ro allow access to the package repos [01:01] escott, mdstat and details says everythings fine and reports no events. but something had to happen for it to go to md0 to md127 read-only. [01:17] SpinningWheels, could be that the arrray is functional but the filesystem itself is corrupted. you should fsck it === justizin_ is now known as justizin === justizin_ is now known as justizin [03:06] zul: wrt that quantum test fixes patch, i spent along time trying to figure out why the db plugin / subnet tests are failing. i think it points to an issue in the dependency chain. [03:08] zul: ...where data gets returned in a dict that isn't ordered correctly, or how the test expects. using a fresh venv from the pip-requires, its comes back fine [03:09] zul: i'd like to get that passing b/c it points to something subtle that may lead to other problems if we continue to just skip it. gonna look at it again tomorrow. cya [03:10] roaksoax: I just meant, it's currently New/undecided for both raring and precise, ISTM it should be fix released for raring [03:11] adam_g: ack [03:16] ive got an ubuntu 12.04 amazon ec2 instance that doesn't seem to have ssh running (connection refused on 22). i mounted the root disk on another instance, turned on bootlogging, put it back, and booted it then shut it down again and once again mounted it to see what the logs would say. i dont see anything about sshd in /var/log/boot.log. mostly i just see stuff about cloud-init. syslog also doesn't show sshd starting, but openssh-server is still [03:16] installed and /etc/init/ssh.conf looks reasonable. any idea how i can further debug what's going on? [04:39] make sure you have ssh/22 enabled in the AWS rules [04:39] it's in your console thing [04:41] http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/using-network-security.html [04:41] maco: ^ [04:41] hey jorge! i did check that [04:42] i don't see anything in syslog about sshd attempting to start on the last couple boots. last thing i see is when we did a "sudo service ssh restart" after modifying sshd_config [04:42] but then, i think it's pam logging that, not sshd itself, so i'm not convinced it would show up on boot [04:47] hi guys how do i block all income traffice from a particular port and allow specific pubci ip address only..any idea..? [05:19] ruben231: you can use ufw for that: https://help.ubuntu.com/12.10/serverguide/firewall.html === hatake is now known as daenglinuxx [07:09] is there anyone around who has worked with uefi? === smb` is now known as smb [08:29] having issues installing grub2 on a 12.10 system with linux raid on /dev/md0p1 /boot .. when running update-grub it returns grub-probe error cant find physical vol (NULL) .. did not have this issue with prev distros has any one else experienced this and is and found a solution ? [08:36] hi all [08:38] koolhead17, greetings [08:40] hello cfhowlett === yofel_ is now known as yofel [10:01] Hello. I can see that kernel 3.5.x exist in 12.04 LTS now days. How can I make sure my server always get's the latest kernel when it is released? Am still at 3.2.x after apt-get dist-upgrade [10:03] Hi! I have tomcat7 installed on my Ubuntu 12.04 64bit machine. It runs correcty if I start it manually (sudo /etc/init.d/tomcat7 start), and it is in the correct runlevels 2 to 5, but it doesn't get started on boot. I need some help here... === daenglinuxx is now known as terlalu === terlalu is now known as blackjack [11:25] join #c === acidflash_ is now known as acidflash [11:38] hi, is that the right source for mongo charm? http://jujucharms.com/charms/precise/mongodb [12:08] How can I find out why a certain init.d script is not called on boot, although it is linked from the correct runlevels? [12:09] Have SSH running on a server that's hooked up directly to the internet with an ip address and no local network. Guy there is able to ssh from local term to server's external ip OK. When we try and ssh from our machines we get login prompt but credentials are rejected. We have denyhosts installed and otherwise default ubuntu sshd settings. Any ideas? [12:26] kzar check the log file? [12:26] patdk-lap: Yes I want to, server is unfortuntely off no and in korea :( [12:26] now* [12:39] Hi, can anyone provide some assistance to a linux noob. Trying to send email from ubuntu server to a WS2K8R2 SMTP relay [12:47] joners1: what's the issue ? [12:50] Well I can send mail to the windows relay using ssmtp [12:50] but cant using postfix, afraid i know so little about it that im afraid im either running around in circles or missing something obvious! [12:53] ok, so just set relayhost [12:53] in the postfix configutation [12:53] http://www.postfix.org/postconf.5.html#relayhost [12:53] for example [12:56] Hi, tried that one didnt seem to work [12:57] sorry, should give you some more information [12:57] i also need to set the 'sender' address as well, the windows box will only accept email from one mail address [12:57] tried the relayhost option but didnt get any mail through [13:07] joners1: postfix only sends mail originating from a relayhost [13:09] ok, any idea how i can send email through the relay at all? :) [13:09] joners1: the MUA sets the 'From' field, postfix relays the message and other MTAs (postfix, exim, exchange) carry the message usually only it came from the host the PTR record for the domain in 'From' specifies [13:09] joners1, http://www.dnsexit.com/support/mailrelay/postfix.html [13:10] also, exchange may not allow you to send mail "From" anyone except the user you're logging in as [13:11] thats right, thats what is happening in this case [13:11] if you have control of the exchange server, the better solution is to allow relay from your postfix server [13:11] we are just moving to office 365 and have a secure connection from the Windows Box to a 365 account (using TLS) [13:12] i wanted to create a white list of ips that can relay email through, for things like printers and scanners etc [13:12] is this postfix intended to work after the move? [13:12] if so you may not be able to do that [13:12] I don't think office 365 allows mail relaying [13:12] it does but only through a specific account [13:12] then what you need to do is configure your devices to send mail from that account [13:13] there may be a way to accept the messages in postfix and mangle them so they fit, but that's beyond my knowledge of postfix [13:13] I wouldn't say that's a good solution [13:14] lol maybe not [13:14] any other suggestions at all? I can get it working with ssmtp with no issues [13:14] that [13:14] points to the windows relay and the mail goes out to 365 and delivered with no problems [13:14] that's because ssmtp isn't a 'real' mail server, and doesn't relay [13:14] it can easily log in with your credentials and send from the right account [13:15] its not logging in though, just pushing email to that relay to send on [13:15] the windows relay is authenticated in to office 365 so anything else that points to the relay is unauthenicted, basically an open relay internally [13:16] sorry, hope that im not confusing the issue here [13:16] you're changing your story [13:16] also, if exchange was configured for open relay, postfix would work [13:16] ok, shall i start again and try and clarify? [13:17] what version of exchange? [13:17] because there's big differences between them [13:17] no exchange at all [13:17] Office 365 is a hosted email platform [13:17] yes, but what runs on your windows server? [13:18] just the SMTP relay feature, part of IIS [13:19] and it's configured to authenticate to office 365? === slank_away is now known as slank [13:19] thats correct [13:19] to relay mail [13:19] yes [13:19] ok [13:19] what's your ssmtp config? [13:19] what i need to figure out, is how to send email to that relay [13:19] two seconds [13:19] ill get it now [13:21] ssmtp.conf simply looks like..... [13:21] root=emailaddress@domain.com [13:21] ok [13:22] is that a valid email in your office 365 domain? [13:22] mailhub=ipaddress_of_windows_relay:25 [13:22] yes [13:22] and is there any auth configured there or just anonymous? [13:22] rewriteDomain=ourdomainname.com [13:22] ah, there you go [13:22] hostname=ubuntuserver name [13:22] that's the feature it has [13:22] the rewrite domain name? [13:23] postfix doesn't have that as far as I know [13:23] yes [13:23] at least not as a builtin [13:23] gotcha! [13:23] how about sendmail or something else? [13:23] your ssmtp is faking that all mail is comfing from your actual domain [13:23] ok, so faking the email details to the windows relay [13:23] sendmail is a disaster, I couldn't begin to tell you how [13:23] right, ssmtp is mangling outgoing email messages [13:23] thats what i noticed looking at the web [13:24] to change the "from" headers to your domain name [13:24] ok [13:24] do you know of any way to send attachments from ssmtp? [13:24] it should be possible with postfix but I don't know how, it's not a normal config [13:24] ok [13:24] well thank you for your help :) [13:24] ill see if i can do some more research, that narrows things down for me though [13:25] it's very extensible but how difficult, I have no idea [13:25] i shall have a read up on it and see what i can find [13:26] one possible workaround is to change the system mailname, but that'll get confusing in logs [13:26] your server will pretend to be yourdomain.com [13:26] and that might break other stuff [13:26] interesting.... [13:26] might be worth a go, ive got a couple of test boxes that i setup to play around with different configurations [13:28] the mail name is located in /etc/mailname right? [13:40] what would cause resolv.conf to have the header bit in twice along with my old DNS settings? [13:41] (12.04 server) [13:41] (with a static IP config) [13:44] hmm... for some strange reason /etc/resolvconf/resolv.conf.d/original and tail have my old settings in them [13:52] ubuntu server seems to completely not listening to my file permission, in sudo or otherwise [13:52] res22: explain [13:52] it's hard [13:52] what command are you running? [13:53] I have a group of users who I want to be able to write to a specific directory. The directory is chowned to that group, and the permission are recursively g+rw [13:54] I hope you included x also, for directories [13:54] they can't. Also sudoers simply doesn't work when I specify a group, but it does when I specify a user [13:54] yes [13:54] and they have x permission all the way to root? [13:55] they can access the directory, they can't write to the files [13:55] can they create a dir or 'touch' a new file? [13:56] permission denied [13:56] brb [13:57] Daviey: any idea when the iso started being oversized? [13:57] may 20/01? [13:59] jamespage: NFI, sorry [14:03] Daviey: well I think I have the answer [14:03] Daviey: quantal: apt-get install kvm -> Need to get 11.3 MB of archives. [14:04] Daviey: raring: apt-get install qemu-kvm -> Need to get 50.4 MB of archives. [14:04] ugh [14:05] Daviey: maybe hallyn can comment on the increase :-) [14:05] jamespage: Possibly recommends which could be suggests [14:05] hallyn: Can we look at this ASAP please? [14:05] (it's pushing our cd images over limit) [14:09] Daviey: guh - also noticed that qemu drops the BD on librbd-dev - so no ceph RBD support [14:09] * jamespage crys a bit [14:09] Daviey: I'll raise bugs [14:10] jamespage: super. I know hallyn is aware of some mismatches... He tried to align with the debian package closer. [14:11] but having bugs to track this stuff is optimal :) [14:13] hallyn: bug 1118406 for rbd support [14:13] Launchpad bug 1118406 in qemu "qemu lacks support for Ceph RBD in raring" [Undecided,New] https://launchpad.net/bugs/1118406 [14:13] Daviey: this highlights the gap we have in openstack testing on current dev series... [14:14] ugh, yeah [14:17] hallyn: bug 1118407 for the bloat-out [14:17] Launchpad bug 1118407 in qemu "qemu-kvm/system has a large number of dependencies" [High,Triaged] https://launchpad.net/bugs/1118407 [14:23] jamespage: all right, i'm going to finish the merge of lastest debian qemu stuff this afternoon. will fix rbd then. for the size increase, i'll take a close look. I'm not sure it's dependencies - it may be due to pulling in of all arches [14:23] if it's due to all arches, then next upload should help (as it splits qemu-system-common-x86 into separate package) [14:23] but i'll take a look,thanks! [14:24] hallyn: in a general finger pointy kinda way - "libgl1-mesa-glx | libgl1" looks like it might pull in quite a bit [14:25] will look [14:28] hallyn: ta [14:35] Hi all. How can I tell if my computer will run ubuntu server 12.04? Right now it is running 10.04 and I would like to upgrade it. [14:42] xpistos: You could check first if it's on the list here: http://www.ubuntu.com/certification/server/ [14:46] genii-around: Thanks. I will check that out nw. [14:49] genii-around: This isn't a "server" persay. It is Dell Dimension 4400 (circa 2001). Is there a doc with the hw reqs somewhere? [14:50] !requirements [14:50] Hardware requirements to install, boot and comfortably use Ubuntu are listed at https://help.ubuntu.com/community/Installation/SystemRequirements - For a !flavor with lower requirements, see !Xubuntu or !Lubuntu [14:50] hrm [14:51] <_ruben> heh, nice .. no clue what key combo i just accidentally hit (area of arrow keys / enter / ins/del / etc) .. but it backgrounded my irssi .. though i was no way near ctrl-z [14:52] Pici: The requirements there look pretty old... [15:02] Pici: Thanks. I checked and I am good. I have 1.3 Ghz proc and 2 GB of Ram [15:02] xpistos: I'm running it here on a VPS with 512mb of ram, so you should be just fine. [15:02] and I use it headless so, looks like I have a project on Saturday === security is now known as megha [15:14] smoser, hallyn: hey, got that while testing the fix for bug 1045955 on precise: [15:14] Launchpad bug 1045955 in lxc "lxc-create -t ubuntu-cloud broken on quantal images (no ubuntu user)" [Medium,Fix released] https://launchpad.net/bugs/1045955 [15:14] ++ ubuntu-cloudimg-query quantal released i386 --format '%{url}\n' [15:14] confused by argument: quantal [15:15] it looks like the cloudimg query tool doesn't even know what quantal is, so even with the fix in lxc, it's not likely to work [15:15] stgraber, apt-get install distro-info [15:17] stgraber, for places that i've had to know a name like that, i've hard coded one, but if distro-info is available, i'll use it [15:17] and at least currently, cloud-utils is Recommend'ing distro-info [15:19] smoser: any chance you can SRU that Recommends? [15:20] win 26 [15:20] oops [15:20] i'm not opposed to that. but it would/might change package content in a cloud-image (not sure if distro-info is there on precise) [15:20] ie, cloud-utils is seeded [15:21] Could somebody comment on bug 1081489 please? I'm not sure about the status of sssd, or what to suggest to the reporter. Is an MIR even on the table, and what would that involve? [15:21] Launchpad bug 1081489 in autofs "autofs package is missing the lookup_sss.so module(!); negating autofs-enabled sssd functionality (automount fails)" [Medium,New] https://launchpad.net/bugs/1081489 [15:22] stgraber, i think you could have just done 'q' to that and it would maybe work. [15:23] ie, single letter arguments 'hklmnopqrstuvwxyz' are then checked to see if they might be a release. [15:28] yolanda: https://code.launchpad.net/~zulcss/nova/nova-config/+merge/147147 [15:28] ok [15:36] zul, done [15:36] yolanda: thanks === slank is now known as wedgwood [17:35] hello masters! [17:36] i would like to compile openldap source with openssl ubuntu package. But when i compile openldap, it doesn't see openssl....how can i do that? [17:36] thanks in advance! [17:37] paco1: is there a ./configure switch that you need to add to add ssl support? [17:38] yes, i do that. [17:39] paco1: is it still set during the build? :) === railsraider_ is now known as railsraider [17:45] i found the right package > libssl-dev [17:46] paco1: did you forget it as a build-dep in the package? (sorry for not thinking itsooner..) === megha is now known as baba [17:57] yes, and i found also "libsasl2-dev" to give --with-cyrus-sasl === baba is now known as gia === gia is now known as baba [18:01] it's perfect. thanks! [18:02] :D [18:09] hallyn: lxc sru uploaded to precise-proposed. Will now poke at quantal [18:12] ok, so if I have mysql-common installed why isn't the mysql client installed (anymore) it should be in /usr/bin right? [18:13] feisar_: did you install mysql-client? [18:14] ah, that'll be why then... although I'm sure I used the client before... [18:14] sarnold: thanks, that's installing now [18:14] woo :) [18:16] something really odd happened today my webserver can no longer talk to my MySQL server (wordpress just says error connecting to db) and I do not understand why (yet)... i did some updates [18:18] feisar_: anything interesting in the logs? [18:19] I am just trying to find where mysql (mariadb) logs it's connections [18:19] is mysql even running? [18:20] hmmm... a manual login from the webserver to the db server tells me I'm not allowed to log in but my IP has not changed [18:41] adam_g: https://code.launchpad.net/~zulcss/cinder/cinder-ftbfs-version/+merge/147205 [19:01] I have created a custom upstart job. When starting the job with the service command everything works well [19:02] the minor problem is that the command does not autocomplete in bash. For example, service networking start, autocompletes the word networking, but with my custom upstart script, "myname" myname does not autocomplete [19:02] anyone know how to fix this annoyance? === matsubara is now known as matsubara-afk [19:13] integrate the path of your command in the system [19:14] longdays: perhaps start reading /usr/share/bash-completion/bash_completion -- there's a lot in there for services handling.. [19:15] longdays: oh. forget that. look in /etc/bash_completion.d/upstart [19:17] thanks, will look there [19:25] Grrr [19:25] Package: python-pyclamd [19:25] Depends: python, python-support (>= 0.90.0), clamav-daemon [19:25] WHY!? [19:25] Why should a python-lib depend on the daemon :( [19:27] perhaps someone slipped [19:27] use easy-install [19:28] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=645872 [19:28] Debian bug 645872 in python-pyclamd "python-pyclamd package installs dependencies no necessaries" [Normal,Open] [19:28] Open since october 2011 :( [19:31] hallyn: and quantal equivalent taken care of [19:36] stgraber: woohoo :) thanks [19:36] i'm hunkering down for a massive qemu update [19:36] i'll peek my head out in a few hours :) [19:37] hallyn: unless we messed up bug triaging, those two uploads should close all outstanding sru bugs [20:17] Hi fine people. I got a fresh isntall of ubuntu server 12.10. I wan to make it a fileserver as the server has a usb hdd. i read in a guide that the usb hdd woould mount automaticaly in /media, but i cannot find it there... my fdisk -l here: http://pastebin.com/SDDAMXtn [20:18] Shogoot: first, I wouldn't use neither non-lts nor a single usb drive for a file server [20:18] secondly, a server install won't automount anything [20:19] im aware of the non lts and siingle usb hdd. i take the risk, all data is backuped [20:20] oh, and using ntfs for a fileserver won't perform very well [20:20] since it'll be handled in userspace [20:21] but of course, you can always add that to /etc/fstab and automount it and share it with samba [20:22] argh. isee its ntfs. [20:22] is it possible to see if the hdd has some data on it from linux even beeing ntfs? [20:23] its suposed to be empty, now im in doubt [20:23] just mount it [20:24] apt-get install ntfs-3g [20:24] (if that's not installed already) [20:25] sudo mount /dev/sdb1 /media/newhd [20:25] like that , right? [20:25] yeah [20:26] if there's no data on it, make another filesystem like ext4 or xfs or something on it [20:28] hello, does anyone know how to fix byobu disable f-keys under 12.04 [20:28] lfreeway: what's the problem? [20:29] smn@neoguaton:/media/newhd/newhd$ ls -a [20:29] . .. [20:29] smn@neoguaton:/media/newhd/newhd$ [20:29] kirkland: using 12.04 with tmux as backend and Shift + F12 is not working... I googled quite a lot and counld find a resolution. I'm using it without anything else.. Only thing it works is if i Ctrl+a and manually write :source /usr/share/byobu/keybindings/f-keys.tmux.disable [20:29] seems its empty [20:30] Shogoot: better unmount it, then and create a good filesystem on that dev [20:31] lfreeway: I can't reproduce on latest upstream ... can you try the latest upstream PPA, "sudo apt-add-repository ppa:byobu/ppa && sudo apt-get update && sudo apt-get install byobu" [20:34] sudo umount sdb1? [20:35] hi, where does mysql log failed login attempts? [20:35] nowhere [20:35] Hi all. I was asking for help earlier with ubuntu server ignoring /etc/group permissions, or seemingly atleast. I have set up groups with users assigned in the file, but sudoer entries that specify the groups permissions are ignored, and users with group read/write permissions to files aren't given these permissions. This has been very annoying to diagnose, any help? [20:37] Shogoot: umount it and mkfs -t ext4 ... [20:37] res22: probably a configuration issue - post your configs [20:39] which ones? [20:40] RoyK, partitioning is done [20:40] res22: ls -la on those files/dirs, "groups" of the user etc [20:40] res22, if memory serves sudo follows the first match that it finds. is it possible that there are multiple matches [20:40] Shogoot: partitioning or making the filesystem? ;) [20:40] RoyK, hehe i guess its the making the filesystem [20:40] ah! mysql uses reverse lookups... [20:41] Shogoot: then use blkid to get the uuid and dd that to to the fstab [20:41] Shogoot: with usb the device name may change, so better use the uuod [20:41] well, with anything the device name may change, really [20:42] what in all heavens is that :) [20:43] kirkland: brb [20:43] Shogoot: what what? [20:43] i have no idea how to use uuid [20:43] oh wait sudo vol_id --uuid /dev/sdb2 ? [20:44] just run blkid [20:44] or sudo blkid if you're not root [20:44] then add something like this to your fstab [20:44] UUID=03987de0-d66a-4dc8-922e-83745dffd9d1 /boot ext2 noatime 0 2 [20:44] oops [20:45] UUID=03987de0-d66a-4dc8-922e-83745dffd9d1 /boot ext2 noatime 0 2 [20:45] just change that to ext4 [20:45] http://pastebin.com/efEhxKFg [20:46] UUID="03987de0-d66a-4dc8-922e-83745dffd9d1" /whereever ext4 noatime,errors=remount-ro 0 2 [20:46] something like that [20:47] perhaps add ,nofail to the options column in case you haven't plugged in the usb drive [20:47] otherwise it'll fail booting [20:47] its plugged in [20:47] well, if you rebooted the server without it, would you like the box to come up cleanly? [20:48] its been rebooted with the usb hdd pluggged in. [20:48] im a bit unsure on the line i shall add to fstab [20:48] just saying it might help [20:48] pastebin it [20:49] erm [20:49] http://pastebin.com/hqB8KCcZ [20:49] don't use my UUID [20:49] no [20:50] what if he wants to boot your disk...? :) [20:50] again... this is my blkid output: http://pastebin.com/efEhxKFg [20:50] Shogoot: more like UUID=81bd9741-f527-4454-a008-68b4d168619d /mountpoint ext4 noatime,errors=remount-ro 0 2 [20:51] kirkland: updated to lastest upstream (5.31) via the ppa and still with tmux as backend shift + F12 is not working... I'm doing something wrong? [20:51] sarnold: heh [20:51] lfreeway: what terminal are you using? [20:52] kirkland: ctrl+a ! gives me "Can't break with single pane" [20:52] ok . line added. shal i reboot ? [20:52] lfreeway: this is almost certainly a problem with your terminal [20:53] Shogoot: given you have console access, yes [20:53] yes [20:53] kirkland: the default one [20:53] lfreeway: you're on an Ubuntu desktop, running gnome-terminal? [20:53] kirkland: no, I'm server [20:54] lfreeway: screen? [20:54] kirkland: on the desktop with gnome-terminal it's working fine [20:55] lfreeway: okay, so you're on a tty then where its broken? [20:57] RoyK, where im suposed to find the hdd? [20:58] Shogoot: where you mounted it? [20:58] kirkland: don't know, that's the weirdo - both with tmux and screen as backend it's not disabling the f-keys... I also did a apt-get purge before installing the from the PPA [20:58] Shogoot: pastebin your fstab again [20:58] also I don't have a thing in .buobu === kkszysiu_ is now known as kkszysiu [20:59] ah.. i see it /mountpoint [20:59] lfreeway: okay, I did reproduce it on a tty [20:59] lfreeway: it seems Ubuntu's tty doesn't handle shift-F* correctly [21:00] lfreeway: and it looks like tmux eats ctrl-a ! to break the current pane [21:00] kirkland: exactly [21:00] (which is a pickle, I admit) [21:00] lfreeway: can you please file a bug? [21:00] lfreeway: thanks! [21:00] kirkland: sure [21:01] kirkland: may I use the occassion to thank you for your work, it really makes working in terminal enviroment much much easier [21:01] RoyK, its mounted where i want and this part is OK! now..... i got to find out how i make that hdd shared on my network [21:01] !samba | Shogoot [21:01] Shogoot: Samba is the way to cooperate with Windows environments. Links with more info: https://wiki.ubuntu.com/MountWindowsSharesPermanently and https://help.ubuntu.com/12.04/serverguide/C/windows-networking.html - Samba can be administered via the web with SWAT. [21:01] RoyK, thanks btw, great help [21:02] i got samba already, i gotto find out of config [21:02] Shogoot: btw, don't bother using swat - it's better to do the config manually [21:02] it's not hard [21:02] heeeeelp ;) [21:02] no, just edit smb.conf [21:02] there's a lot of help in there [21:02] or man smb.conf [21:03] * RoyK slaps Shogoot with a small herring [21:04] http://fish.dnr.cornell.edu/nyfish/Clupeidae/blueback_herring.jpg [21:07] perhaps a small perch or a roach [21:08] or a northen pike, if I feel evil :D [21:09] ha-ha [21:10] Shogoot: well, never mind the fish, server running? [21:11] best fish for slappnig is : http://en.wikipedia.org/wiki/Salmo_trutta_trutta tasty too [21:11] running, yes. security= user smb.conf working [21:12] well, a bit heavy, better stick with something smaller and more spiky like a perca fluviatilis [21:13] * mandoguit thinks that something fishy is going on here and those links are red herrings [21:17] mandoguit: heh - esox lucius is a bit fishier ;) [21:18] or perhaps nastier [21:18] doesn't taste very well either [21:18] but I guess that's {{oftopic}} somehow [21:18] s/oft/off/ [21:19] bah [21:22] soooo ok... ive configed and i can actually see the stuff on the hdd... goodie. [21:23] but where do i find the permitions to add directories from my win7 machine? [21:23] cant see it [21:24] Shogoot, win7? win7 is not posix it doesn't have permissions as we understand them [21:25] ok this is samba [21:25] RoyK, [21:25] yes escott_ samba is installed [21:25] Shogoot, [21:25] i was under the impression samba would fix the win vs linux filesharing thing [21:26] well, obviously you need to allow the users to write to this place [21:26] if you're using authenticated users, you need a common group or a very wide umask [21:26] otherwise, if the system is closed, you can turn off authentication [21:27] security = user [21:27] aka none [21:27] i got security = user [21:27] then just chown the whole tree to the samba user [21:27] errrrrr [21:28] and samba has only one user wich is myself [21:28] sec [21:28] rgr [21:29] sorry [21:29] security = share [21:29] that'll disable auth [21:29] aka "win 95 security" [21:30] perhaps add "guest only = yes" [21:30] but don't let anyone near this system unless you want to restrict data [21:30] s/unless/if/ [21:31] not working and i am restarting smbd and nmbd [21:32] i still cant make a directory from my win7 box on the shared hdd [21:32] tried settiing security = share? [21:32] if you don't want free access, use smbpasswd 'username' to set a password [21:33] the unix password hashes aren't compatible with those windows uses [21:34] if you use security = user, don't use guest only = yes [21:34] making my way through the ubuntu server guide here > remote administration > puppet any particular good reason to install/implement this software over just using ssh connection and cli commands? tnks [21:36] puppet is a large system for mass-administering systems. if you don't understand linux, don't even try [21:36] RoyK, http://pastebin.com/MQPBxFNT [21:37] Shogoot: that's just comments [21:37] bah sorry it didnt cut out all the teks [21:37] !pastebinit | Shogoot [21:37] Shogoot: pastebinit is the command-line equivalent of !pastebin - Command output, or other text can be redirected to pastebinit, which then reports an URL containing the output - To use pastebinit, install the « pastebinit » package from a package manager - Simple usage: command | pastebinit -b http://paste.ubuntu.com [21:38] RoyK: just a home system setup right now....sounds like it would be overboard for my general purposes then. thnks [21:38] mandoguit: no reason for puppet for a home system unless you're trying to learn it [21:39] Royk: maybe later on then... again tnks. :) [21:40] exit [21:40] mandoguit: learn the system, use the commandline, edit the files and you'll learn how to work things out [21:42] patebininit coolest thing i must say. http://paste.ubuntu.com/1622288/ [21:42] looka t line 280 [21:42] most of the juicy stuff is tehre [21:43] so, can you connect to the box? [21:45] * RoyK needs to sleep soon [21:45] yes [21:46] i can se it on explorer.exe too [21:46] well, then auth isn't a problem [21:46] just cant make a directory or file on it [21:46] try to chmod 777 that dir [21:46] what is taht with makein smb user chmod [21:47] then create a dir [21:47] then see what user created it [21:47] fixt [21:47] how do isee who made it? [21:48] you can't if you don't use autentication [21:48] security = share == no auth [21:48] im going to change that to = user [21:49] then you'll have to use smbpasswd to set passwords, or reconfigure samba to use pam auth [21:49] and better get some cheap drives for a small raid while you're at it [21:50] a single drive will fill up with all sorts of crap and won't be backed up until it dies [21:50] all data has backup [21:50] how do i see who made the directory? [21:50] at which time people scream in a solitary havoc [21:50] i changed security to user again [21:50] restart samba [21:50] not reload [21:51] and log in [21:51] and your username will be on the files [21:51] rgr [21:51] grrbprpprpbb [21:51] who is the local boot here in channel=? [21:51] !samba [21:51] Samba is the way to cooperate with Windows environments. Links with more info: https://wiki.ubuntu.com/MountWindowsSharesPermanently and https://help.ubuntu.com/12.04/serverguide/C/windows-networking.html - Samba can be administered via the web with SWAT. [21:52] !ubottu sleep [21:52] cant remember these commands [21:52] i guess ill need voice anyways [21:52] well, my time to /sleep ;) [21:52] Shogoot: just rtfm [21:52] it's not that hard [21:53] RoyK, thank again. [21:53] im gonna set my hdd to the test now ;) [21:53] smartctl -t long is a good start [21:54] or perhaps forget about that and just get a few cheap drives to setup in a raid [21:55] Shogoot: I've been working with storage for 10+ years, and what I've learned is that all drives fail [21:56] :) [21:56] perhaps tomorrow, perhaps in five years, but you never know [21:56] most drives fail the first 3 months, if they survive that, they can live a long time [21:57] that is, not "most drive fails the first three months", but "most drives that fail are either < 3 months or > 3 years" [21:57] ive got this hdd for more then 5 [21:57] :S [21:57] what does smartctl say about it? [21:57] but im not worried its gonna be a clone of an newer hdd [21:58] havent run it [21:58] sec [21:58] start with smartctl -H [21:58] then smartctl -t short [21:59] usb drives don't always support checking, though [22:01] errr [22:01] http://pastebin.com/FGgAinGS [22:02] yes [22:02] you have to give it the device name [22:02] obviously ;) [22:02] /dev/sdb or something [22:02] !man | Shogoot [22:02] Shogoot: The "man" command brings up the Linux manual pages for the command you're interested in. Try "man intro" at the command line, or see https://help.ubuntu.com/community/UsingTheTerminal | Manpages online: http://manpages.ubuntu.com/ [22:03] smn@neoguaton:/dev$ smartctl -H /dev/sdb1 [22:03] smartctl 5.43 2012-06-30 r3573 [x86_64-linux-3.5.0-17-generic] (local build) [22:03] Copyright (C) 2002-12 by Bruce Allen, http://smartmontools.sourceforge.net [22:03] Smartctl open device: /dev/sdb1 failed: Permission denied [22:03] smn@neoguaton:/dev$ [22:03] woops [22:04] !pastebin | Shogoot [22:04] Shogoot: For posting multi-line texts into the channel, please use http://paste.ubuntu.com | To post !screenshots use http://imagebin.org/?page=add | !pastebinit to paste directly from command line | Make sure you give us the URL for your paste - see also the channel topic. [22:04] i know i know :P [22:04] http://pastebin.com/Dw5A4yDf [22:04] you need to be root to run that - put sudo in front [22:05] SMART Health Status: OK === ErtanERBEK is now known as Guest57736 [22:06] well, good [22:06] still, it'll eventually die [22:06] http://paste.ubuntu.com/1622345/ [22:06] so beware of hardware failure [22:07] typical usb thing [22:07] rgr [22:07] better get a cheap sata card [22:07] install internally [22:07] add a mirror [22:07] do it right [22:07] that usb drive will eventually die [22:07] get two smallish drives, create a raid-5 on them [22:08] copy the data over [22:08] add another if you get out of space [22:08] grow it to raid-6 if you get a lot of drives [22:09] remove the smaller drives during time, when all drives are at the same size, the raid will grow [22:09] or you can grow it to all what's left [22:09] so, Shogoot, learn linux, learn md raid, preferably - start off with a vm where you can play around a bit with it [22:10] that way you can avoid the errors most newbies do [22:10] your single drive server will probably work for a while, but that lone drive will eventually die [22:10] so better get a raid running [22:11] just my 2c [22:11] i will :) [22:11] nite [22:11] good night [22:13] if off too. bye bye people and boots === e-diablo_ is now known as Ertan [22:27] I asked before, what is the equivalent of chkconfig --list on unbutu-server? [22:28] Hello. I have a nice PXE setup for automatically deploying servers. Everything is over gigabit LAN with local mirrors. However, even a minimal installation seems to be slow -- mayb 30-60 minutes. Anyone have thoughts on how to speed this up or better rapid delpoy solutions? [22:29] sweettea: service --status-all ? [22:29] http://www.ubuntu.com/business/server mentions that there are great new rapid deployment tools, but fails to mention what they are. [22:31] chrisd___: I believe that is referring to juju, which is the cloud/maas deployment thing. [22:32] Pici: I want to get a list of all the services running that I might want to turn off [22:32] like networkmanager [22:36] Pici: It mentioned that there are a bunch of things for deployment, and then mentioned JuJu as a secondary thing for cloud deployment. [22:37] chrisd___: perhaps maas is what you're after? [22:37] http://www.ubuntu.com/cloud/orchestration/deployment [22:37] http://maas.ubuntu.com/ too [22:37] juju is only going to make sense on a pile of machines if yo'uve got maas on them, or install openstack on them manually [22:38] juju won't get you to installed machines more quickly [22:39] Yeah, those things just seem like the new Ubuntu-y way of doing Cobbler or Puppet or such. Which isn't a bad thing, but not going to help speed up the Ubuntu Installer itelf. I'm seriously an hour deep on this automated preseed install for a minimal KVM server. I can do the same in RHEL in 10 minutes. [22:40] Just trying to figure out if I'm doing something wrong/there's a faster way to do this, or if the installer and preseed are just inherently slow? [22:41] chrisd___: I've never done installs over the network, but on a local vm, it only takes a few minutes :/ [22:41] Pici: sysv-rc-conf --list [22:41] that sound about right? [22:41] sweettea: ps auxw would be more useful [22:41] does anyone know how to stress test a ton of memory thats not using memtest [22:41] sarnold: thats what I have been using :) [22:42] sarnold: Are you using preseed files to automate the local vm install, or is that with doing things manually it's still faster? [22:43] I'm trying to figure out the bottleneck here, but it just seems like the Ubuntu Installer itself. The slow parts are installing all the packages and what not. [22:43] sweettea: compiling qt4-x11 is pretty stressful.. [22:43] sweettea: start as mnay of those as it takes to push you into swap.. :) [22:45] chrisd___: preseed to automate the local vm install [22:45] chrisd___: (I don't actually know how the tool does it, but preseeding the isos is one of the steps :) [22:46] sarnold: Cool, I'll have to try that and check the speed difference. [22:47] chrisd___: there's also the FAI tool, in universe, that might do the job. At least they claim "minutes" too.. [22:48] chrisd___: but .. I've got to wonder if it is something more basic, like wrong NIC speeds, or not writing to the disk correctly, or .. meh. it really shouldn't take long. [22:56] sarnold: I've seen it with multiple systems in different network segments. [23:00] chrisd___: hrm. :/ if you don't come up with a good reason soon, perhaps filing a bug report or askubuntu.com question would be the way to go :/ === kentb is now known as kentb-out [23:18] sarnold: damn I think I have bad memory [23:18] any experience setting up mcelog? I need to find out whih dimm is bad :( [23:19] sweettea: sudo apt-get install mcelog [23:20] sweettea: then watch file /var/log/mcelog [23:20] sweettea: compiling qt4-x11 is how I discovered I had a bad motherboard or cpu :( [23:20] (and, like you, memtest86 never showed any problems) [23:21] does it log to mcelog by default? I thought syslog was still default [23:22] RoyK: /var/log/mcelog worked for me on 12.04 lts.. [23:22] sarnold: do I have to do anything aftr I install mcelog? [23:22] newer or older might be different :) but it had its own files and put very little effort into setting it up. [23:22] ok, last I had issues was with 10.04 [23:22] like run it? [23:22] i ran memtestr and the machine froze [23:22] memtester [23:23] sweettea: iirc, it should start automatically... [23:23] sweettea: memtest86 or the one you have in the boot menu should work [23:23] sweettea: if not, probably "service mcelog start" would be sufficient [23:23] RoyK: that doesn't spot everything though :( [23:23] neither does mcelog [23:24] i heard memtest sucks [23:24] yea [23:24] RoyK: I had a clean bill of health from memtest86, but mcelog routinely would log cache coherency problems, and once in a while, the machine would lock up -- when an irrecoverable fault would happen. [23:24] sarnold: but if he had a hang during memtest, well [23:24] RoyK: true enough, true enough... [23:24] RoyK: heh, that weas memtestr, I assumed that was a linux program... [23:24] RoyK: it won't be able to help find the dimm to replac,e if any [23:25] sweettea: oh, I asked a buddy, apprently the "banks" reported via mcelog have nothing to do with the physical banks of memory. [23:25] sweettea: desktop system or server? [23:25] server [23:25] ecc memory [23:25] amd AD [23:25] then call support [23:25] huh [23:25] i built it myself [23:25] budget [23:25] oh [23:25] ok [23:25] sweettea: ah, /etc/init.d/mcelog start :) [23:26] enable mcelog by default, as sarnold suggests [23:26] mcelog is running. [23:26] yay [23:26] good luck :) [23:26] its running [23:26] mcelog file is empty [23:27] dmesg has stack errors [23:27] sweettea: was it running last time things crashed? [23:27] memory errors can block i/o [23:27] so better use a network console or a serial one if you're an old timer [23:27] .. I don't think it'll tell you anything about the past. it has to be running to log anything. [23:27] ohhh [23:27] so it cant write to disk? [23:27] errors? [23:27] blarg [23:27] often not [23:28] and when I tried netconsole, I got nearly nothing :( it wasn't worth the hassle. :( [23:28] serial console always work [23:28] netconsole should too [23:28] I dont have that hooked up, I already racked the crap [23:28] I'd have more faith in serial console.. [23:28] I might as well just buy memory and replace [23:29] sweettea: iff it is memory [23:29] im pretty sure its not hd [23:29] I was writing dd overnight [23:29] sarnold: tried to do a good old memtest86? [23:29] it could be chip :/ [23:29] to the hd [23:29] RoyK: yes, it never found anything. I assumed it was because the faults weren't on the boot cpu. [23:29] Ive never had a bad chip in my entire life [23:29] (well, by 'never', I mean "not in 30-something hours..) [23:29] if it is [23:30] ill be sad [23:30] sarnold: memtest86 tests everyhing [23:30] memtest wont tell me which dimm is bad [23:30] sweettea: if it is just a bad dimm, it ought to .. [23:30] it'll tell you the address [23:30] but I guess it would tell me if its indeed mem vs cpu [23:30] sweettea: or, at least, I've had success with taking bad addresses from memtest86 and blocking them out with mem=.... [23:31] sweettea: the address is absolute in a small system [23:31] its 64G of ram [23:31] not big [23:31] will take at least 1 day right [23:31] something like that, yes [23:31] for a full run [23:32] the memory errors usually show up before stage 3-4, though [23:32] early random writes etc [23:32] okay going to the DC :( [23:32] good luck sweettea :) [23:32] thanks for your time guys [23:45] i want to use the command line only to listen to radio and songza.com problem is these are flash and java based. i know i can use mplayer with an ip to steam, but i am having problems finding the ips i want. if songza.com streams to a gui browser doesnt there have to be an ip somewhere to get?