=== N0DE` is now known as CripperZ- === deny_ is now known as misterpink === Sachiru is now known as Guest15651 === Sachiru_ is now known as Sachiru [06:55] Are the nosuid, nodev and noexec permissions generally safe to use for the /tmp filesystem on a server? I have found some daemons which write executable files to /tmp... === Beliq_ is now known as Beliq === a1berto_ is now known as a1berto [07:48] is there a way to set upstart log size limits? === Lcawte|Away is now known as Lcawte [08:04] jodh, do you know the answer to yoyo09kh3's question? [08:04] ^^ [11:16] morning [11:32] I tried installing Ubuntu server 14.04, and got the same error as http://ubuntuforums.org/showthread.php?t=1981020 [11:32] I couldn't install grub, so I installed LILO during installation [11:46] jeffreylevesque: try putting a UUID in there.... it's stupid to use ambiguous device names [11:46] blkid lists the UUIDs [11:46] put a UUID in fstab and run update-grub, and it will probably do it for you [11:46] (for boot and root) [11:46] sometimes "root" in grub files means /boot [11:48] peetaur2: What iare UUID? [11:49] universal unique id [11:50] the ones listed by blkid are in the filesystem header stuff [11:50] How so I get those? [11:50] blkid .... already said that [11:50] peetaur2: http://tinypic.com/r/256bbj4/8 [11:52] jeffreylevesque: that has no POSIX file systems listed [11:52] eg. ext4 [12:03] peetaur2: How do I get that information, sorry I get disconnected on my iPhone (since laptop in inoperable) [12:08] run "blkid" and it prints it on screen [12:12] peetaur2: Got it [12:12] I see the uuid's [12:14] peetaur2: where do I assign these UUID's? [12:17] put them in fstab without the " [12:17] UUID=... / ext4 blah blah [12:17] just replace /dev/... with UUID=... without any quotes [12:18] and then run: update-initramfs -u and then run update-grub [12:18] to do that properly, you should be in a chroot or in the root system, not just mounted in /mnt/ or whatever [12:19] peetaur2: The terminal is prefixed with (initramfs). How do I get I to chroot? [12:20] well you can't chroot from there... but it is possible to get it to finish booting (but not easy) with some manual commands [12:20] you should use a LiveCD or rescue cd [12:20] cd /root? [12:21] the Ubuntu installer rescue mode should offer to chroot for you already [12:21] no... you have to get root mounted, then mount dev, proc, sys in there, then "chroot /whereveryoumountedit/" [12:21] and on initramfs you are missing all the commands to do that [12:22] I put the usb installer in. Do i go in rescue mode? [12:23] yes [12:23] peetaur2: Where is the fstab file located [12:23] I'm doing the rescue mode [12:24] fstab is /etc/fstab ... and if you have to ask, yo7u should back it up first. cp /etc/fstab /etc/fstab.bak [12:24] what I do when there's no mouse copy & paste is to run: blkid >> /etc/fstab (>> not > or you delete the file) [12:24] and then edit it with vim, and remove the unwanted lines, and copy and paste [12:24] and if you don't know vim, use another editor [12:28] peetaur2: I can cd into '/etc' without doing any unmounting in 'initramfs'. But, there is no fstab [12:39] /etc and everything in your initramfs are inside the initramfs archive... not the real root [12:52] peetaur2: I booted from USB and I did ls, I can see the file now fstab. But, if I try to use pico or vim, it says /bin/sh: vim: not found [12:54] okay so use nano or vi [12:54] or get in the proper rescue that does the chroot... if you get that one, it'll have whatever your broken OS had installed (and it should come with vi even if not vim and nano and pico) [12:55] That worked [12:55] Nano [12:58] also your chroot should probably have apt-get working... you can even install things [12:59] Im at a library. So i dont have internet access with their firewall thing [13:00] peetaur2: I hope fstab changes will work, otherwise I have to go home [13:03] jeffreylevesque: you also ran these, right? pdate-initramfs -u ; update-grub [13:03] first puts the info in the initramfs, and 2nd puts it in grub.cfg [13:03] In the shell before modifying fstab? [13:05] after [13:05] those commands read fstab [13:08] Fstab has no defined UUID's [13:09] peetaur2: There are four lines the first column for each is "none" [13:09] My phone is about to die [13:09] Have to go home [13:10] I took notes of what you said [13:10] Second column reads /dev/pts, /run, /proc, /sys [13:11] Third column: devpts, tmpfs, proc, sysfs [13:12] Fourth, defaults; nosuid,size=10%,mode=755; 0; 0 [13:12] The rest of the columns are 0s or blank [13:13] only modify the / and /boot lines [13:13] just change the first column, nothing more [13:13] and back up the files [13:13] if you are unsure, always back up anything you edit, especially things required for booting [13:14] The EC2 apt repositories for 12.10 were pulled down immediately when 12.10 went EOL. Does anyone know if 13.10 EC2 repos will remain online? Right now they are (which is a good thing for me at the moment). [13:15] ex: http://us-east-1.ec2.archive.ubuntu.com repositories used by EC2 images [13:15] Ok, created a backup [13:15] stmiller don't plan on it. end of life = end of support. suggest you upgrade to long term support version : 14.04 [13:16] cfhowlett I know - upgrade is in the works. I am looking for any dates or indications for 13.10 status and those repositories. Otherwise, EC2 automation with 13.10 images is out of luck. [13:26] HI we are developing linux boxes which should have identical system configuration and Os (ubuntu 14.04). This boxes will boot in UEFI mode. I like to know whether the grub file /boot/efi/EFI/ubuntu/grubx86.efi will be identical for all system, or has some dependencies to installed system? thx [13:43] zartoosh: identical on all, the boot images are secure-boot compatible and therefore using signing just-in-case Secure Boot is enabled [13:47] I found IRC as a webservice [13:48] peetaur2, https://github.com/jeff1evesque/audio-analyzer/issues/351#issuecomment-49870612 [13:56] jeffreylevesque: ooookay that makes no sense. I said to enter the UUIDs for boot and root, but you did it for /dev/pts. And you set it to an exfat partition ... which won't work. And blkid doesn't list any POSIX file systems, so I don't think you have any bootable Linux system there. [13:57] service networking restart doesn't seem to work for me ... this is what it reports: stop: Job failed while stopping & start: Job is already running: networking [13:57] maybe when you edited fstab, it was not you root, but the live system. Check that with "df" and see which device is mounted on / [13:57] peetaur2, what should the partition be, ntfs? [13:57] jeffreylevesque: you need to find your Linux install .... and it shoudl be a POSIX system: ext4, ext3, reiserfs, xfs, btrfs (don't use that), etc. [13:58] I used to be able to use /etc/init.d/networking restart successfully in Ubuntu, but now that too doesn't work [13:58] not exfat, vfat, fat32, ntfs [13:58] and don't break your old ntfs partitions unless you know what you are doing [13:58] back them up [13:59] make sure to format something that Linux can install to (I suggest ext4, the default choice) [13:59] peetaur2, i created the exfat partition yesterday so i could tell which partition i was going to use (stand out against the current ntfs). I thought ubuntu installer would allow me to erase the exfat partition, and use it accordingly for the Ubuntu Server [13:59] zetheroo1: in 14.04, that script is disabled without telling you. It's a feature. You have to use "ifdown $device" and "ifup $device" now [14:00] How do I remove the partition entirely? [14:00] so if I change something in /etc/network/interfaces how do I get all the changes applied in one go? [14:00] zetheroo1: ifdown is a stupid non-equivalent though.. for example it won't take down an interface that you removed from /etc/network/interfaces ... for that use: ip a del ipgoeshere/bits dev $device (CIDR notation) [14:01] peetaur2: that is extremely annoying [14:01] it is almost sane.... but makes no sense to just make the script return 0 and print no warnings. yes... extremely annoying. [14:01] jeffreylevesque: select one in that partition editor you showed me in a screenshot earlier, and say to remove it [14:01] peetaur2: I am configuring bonding and bridging in the interfaces file and I want it all implemented like in 12.04 ... :P [14:02] jeffreylevesque: but make sure! if you delete your windows partitino, it won't be a dual boot.... it'll be Linux only [14:02] peetaur2, my windows partition is important. I can't delete that [14:02] so the command "service networking restart" is also useless!? [14:03] peetaur2, I'm going to record the results of "df" [14:06] restarting the machine is almost simpler then using this ifdown/up stuff :P [14:07] peetaur2, https://github.com/jeff1evesque/audio-analyzer/issues/351#issuecomment-49877903 [14:07] jeffreylevesque: make sure you back it up ... if you're not very familiar with this stuff, it is at risk if you continue without a backup [14:08] peetaur, I cant' access either partitions at the moment [14:08] jeffreylevesque: okay so from that df, I'd say you don't have your linux filesystem mounted. [14:09] jeffreylevesque: well you should find some way to back it up ... one way is to use dd and copy the image to another equal sized disk (in this case the other disk will even boot windows) [14:09] if it's a different sized disk, it will still copy but it is harder to restore and won't boot [14:10] peetaur2, how can you tell that linux is not mounted, and how do I remove the exfat partition. I guess I'd like the laptop to be able to boot into windows again. Then, I will try to setup this ubuntu installation again [14:11] if it was mounted, it should show a real disk in the leftmost column, like /dev/sda3 and on the mountpoint column, it would say / [14:12] seeing nothing mounted on / means it's some weird hack like a rescue with some weird root that isn't mounted normally [14:13] maybe you would be best with a gparted live cd for removing the exfat partition [14:13] there are lots of ways, but the gparted gui would be easiest [14:13] gparted can also work in any GUI LiveCD; don't need a special CD [14:13] but the installer has no GUI [14:16] peetaur2, is the best way to remove this phantom ubuntu so my machine boots back up to windows 7, is to use gpart? Or can I restore my machine another way? [14:18] to make windows boot, you need a proper GRUB installation or some other bootloader. The easiest would be to use the windows cd to reinstall the windows bootloader maybe. I don't know the details on that, but you can ask in ##windows [14:19] I think you should (1) back up everything (2) plan your partitioning.... reserve some space for windows, and for linux, and try to use fewer partitions (3) change your partitions and leave an empty one for Linux (4) install Linux there [14:19] and for dual boot there are more details that I'm not to familar with ... [14:28] hi [14:28] so I have an outdated server running ubuntu 11.10, and I'm willing to do an upgrade to 12.04 LTS. the server is in production, so i wish i could do this upgrade at night unattended. [14:28] I have read about the DistUpgradeViewNonInteractive flag which should agree to every default prompt [14:29] I have many services, like dovecot, postfix, and so on with custom configuration (virtual aliases...), how can I know the upgrade will not break everything ? [14:29] where can I read about the changes of every packages/services ? [14:30] do I really have to read all the changes? or should it work right away (backward compatibility in the /etc config?) === cripperz is now known as CripperZ === CripperZ is now known as N0DE` [14:38] micmac: the update will break everything... back it up [14:38] micmac: ... is a good assumption to keep you safe. ;) [14:39] I do backups regularly [14:39] micmac: ubuntu release update deleted my /etc/openvpn once, including private keys. In such cases, only the backup will save you. [14:39] micmac assume breakage. have plans to restore and reconfigure your system and pull the trigger [14:40] ok [14:42] hmm I could try to clone the server into a VM, then launch the do-release-upgrade, and then check if the services are ok [14:42] then apply to the original [14:43] (still with backups) [14:43] I guess some of the services could have major changes, others would not, depends on the ones I'm using [14:44] testing like that can be good. Might take some time, but might save you a headache. [14:44] might be better than reading all the doc too.. [14:45] micmac actually, peetaur2 brings up an excellent suggestion. As you have a production server and need to minimized downtime, you might consider mirroring your present setup on a different box. then upgrade that and see how it behaves before you go to working box [14:47] will do that [14:47] thanks [14:47] bye === Solution-X is now known as Solution-X|AFK === chmurifree is now known as chmuri [15:32] TJ-, thank you so much you did understand me main reason I was asking this question, it is of Secure Boot. Many thanks again for your answer. [15:40] Good afternoon. === TDog_ is now known as TDog [16:11] what is a chroot? [16:12] is it a template for clients on a server? [16:12] CooLBALL1: see https://help.ubuntu.com/community/BasicChroot [16:13] neither [16:16] i want to set up a server that runs applications that fat clients can use [16:17] are openSSH LAMP and postgreSQL required for this type of server? [16:21] how should we know? [16:21] that would depend on what the *applications* are [16:21] and how you configure them === Ursinha is now known as Ursinha-afk [17:27] I'm getting the following error when running a test on my config for SSL, and through Googling as far as I can tell everything seems to be in line. Anybody have any idea what's wrong, or anything I should be looking into? [17:27] Error: http://p.ngx.cc/c7 - Config: http://p.ngx.cc/d8 [17:32] DeltaHeavy: why are you trying to use a certificate signing request instead of a certificate or a keyfile? [17:36] sarnold: Yeah, I just realized I can download a set of certs. I forgot I needed one signed. Thanks === Ursinha-afk is now known as Ursinha [17:48] hallyn: http://www.openwall.com/lists/oss-security/2014/07/23/10 [17:49] sigh, every time you post a link from there i have to worry [17:51] hallyn: sorry :) [17:51] sarnold: that shoulda been easy to spot. hope i didn't review the patch that went in through [17:51] sarnold: what about this one specially interests you? [17:52] suppose this coudl explain why rootfs sometiems doesnt' cleanly umount [17:52] hallyn: heh, that's always my fear too, "please don't let this one be one I've overlooked' [17:52] heck this could explain why the installer hasn't been cleanly umounting for psivaa and plars. maybe :) [17:52] hallyn: mostly that root-in-container seems to be enough to able to trigger it [17:52] do you need that? [17:52] i thought it said an unpriv user can do it [17:53] that -might- just be an artifact of how it affected RHEL? [17:53] yeah [18:17] I put these lines at the bottom of /etc/ssh/sshd_config and now I can't connect via my SFTP client. I have this identical config on another server so I'm unsure why it's doing this. Users NOT in the group 'sftp' work though - https://www.refheap.com/88518 [18:18] DeltaHeavy: check error messages in the logs? [18:18] Bah, true. [18:23] sarnold: Got it working, home directory of the SFTP account had to be owned by root [18:24] could someone assist me. I messed up by ubuntu server installation [18:24] DeltaHeavy: nice. :) [18:25] jeffreylevesque: What's wrong? [18:25] I decreased my C: drive to allow a large enough partition for a dualboot into Ubuntu. I couldn't decifer which partition I created for ubuntu during install. So I booted back in windows 7, and made the extra space exFat partition, this way i could tell when I attempted to install ubuntu, hoping ubuntu would reformat the partition as needed. [18:25] jeffreylevesque: irc tends to work best if you just ask questions :) see DeltaHeavy's question for a good example :) [18:25] After install, ubuntu starts first, but doesn't load properly, and I can't boot into windows now [18:27] jeffreylevesque: in what way does ubuntu not load properly? [18:27] sarnold, https://github.com/jeff1evesque/audio-analyzer/issues/351#issuecomment-49867492 [18:28] LILO?? [18:29] grub wouldn't install [18:29] gave me an error [18:30] jeffreylevesque: sorry, this one is way beyond me; it looks like you've got the four primary partitions filled (sda1, sda2, sda3, sda4) then the extended partition (sda5), which seems wrong to me.. and using unetbootin, something I've never tried before, and using LILO on top of everything else... [18:31] jeffreylevesque: Why did you post an issue on a totally unrelated peice of software? [18:31] That's not what that's for :p [18:31] jeffreylevesque: you may have better luck in #ubuntu. I know it's a madhouse in there but this channel is mostly populated with folks who have simple server systems that boot ubuntu and only ubuntu. :) [18:31] They're for legitimate bugs, not misuse. [18:31] sarnold, thank you! === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha [20:18] hello. I am updating the IPs on my servers in /etc/network/interfaces. Is there a command to get Ubuntu to update all the interfaces so that it is the same as that which is in /etc/network/interfaces ( like service networking restart, but doesn't horribly break the server ) ? [20:34] toyotapie, reboot? [20:35] the correct way is to ifdown, edit interfaces, then ifup [20:58] anyone here know about dual boot with ubuntu server? === thesheff17_ is now known as thesheff17 === TDog_ is now known as TDog