[00:09] why does it seem like sometimes after kernel/grub updates that the default kernel selection no longer works and i have to press enter? sucks for remote systems [00:25] That happens if the GRUB 'recordfail' flag is set due to an unclean OS shutdown [00:25] TJ-: thanks, i will check [00:26] You can disable that functionality in he grub configuration [00:27] It's a relatively new feature that's a pain in the rear [00:28] recordfail has been around many years :) [00:35] odd, all the suggestions im finding are old and seem like hacks === markthomas is now known as markthomas|away [01:49] i need help [01:51] roseysdad: Just ask, people will answer if they are active and have a helpful answer. [02:40] does it make sense to run mail server on VPS with just 256M of ram? 14.04 LTS 32-bit, postfix+dovecot+something to filter spam [02:46] maybe if it's ssd backed or osmethinng... [02:46] otherwise feels a bit itght [02:54] how can i see what services are running in systemd? [03:02] sarnold, pure SSD [03:03] and super low volume (never over 50 mails a day, 99% of them under 10 kilobytes) [03:05] ksx4system: It might work, but I'm still pretty nervous about that, especially if it is to be used for something in production/mission-critical. [03:07] adun153, what exactly is that *insanely* resource hungry? the spam filtering part? [03:08] (of course I'm going to run postfix/dovecot combo operating without SQL database) [03:09] Ubuntu has never exactly been a "lean" distro, if you know what I mean. [03:09] You can probably run it, but it might have to swap quite a bit, I thin. [03:09] *think. [03:09] adun153, if you take some time it can be as light as properly tuned Debian :) [03:10] but Ubuntu has longer support cycles (5 years for LTS) [03:11] what about 512M? would it help? [03:11] ksx4system: I'd be a lot more comfortable with 512. [03:12] this is insane [03:12] ksx4system: But it seems like you have the time, you can set up the 256 instance, try it out, see if your hypothesis is correct. [03:13] You asked us for our opinions, you got it from sarnold and myself. Eventually, it's entirely up to you. :) [03:13] adun153, well I have around 3 months until date I have to move [03:13] so I can do some proper testing :) [03:13] Sounds good. [03:14] webmail is not an issue (hardcore tuned lighttpd + minimal php5 + squirrelmail) I think [03:14] this should eat at most 20M [03:15] if not less... [03:15] what about backup MX? do I need a big (256-512M) VPS too? [03:18] If just running Postfix, I think 256 should be fine. [03:19] will 128 do? [03:19] (yes, I always try to squeeze stuff as much as possible) [03:22] 128? I don't know. Leaning towards "no". [03:25] challenge accepted lol [03:25] lol [03:25] good luck [03:29] thanks :) [03:29] (I was actually surprised that mail server needs anything over 64M) === Jalen_ is now known as Jalen === Jalen is now known as Guest46060 === cpaelzer_ is now known as cpaelzer [08:16] Good morning. [08:51] hey guys - anyone here successfully install Ubuntu 12.04 or 14.04 (either LTS) successfully on a Dell C6100 blade server? It will just not install - and the installer is quite slow when it does try and work. [08:52] mripguru: "It will just not install", do you get errors or something? [08:53] lordievader: nope - but, most of the operations, etc. take on average a year and a day -- and when it did install earlier on (another blade in the same cluster) - it would boot to a black screen [08:54] so no - it doesn't technically error [08:54] mripguru: Hmm, the installer does run? [08:54] but it doesn't work - either. [08:55] lordievader: it looks like it's going to install successfully - but, usually just hangs up for awhile [08:55] like the partitioning stage [08:55] From the limited shell the installer gives you I'd test the disks and the overall performance of the machine. [08:55] can take 20-30 min [08:56] Since what you describe make me think that there is some trouble accessing the disks. [08:58] bad disks? [08:58] mripguru: Could be, like I said, test them ;) [09:05] yup - let's see what another node does [09:06] ? Why? [09:06] Test where the problem lies not somewhere else. [09:09] Hey does anybody here tried the openstack-installer on ubuntu server...I'm behind a proxy and the creating of the container works but the initialization of the container fails...I'm 100% sure that there has to be some additional proxy-setting that I might have missed [09:10] lordievader: I want to see if another node has the same behavior [09:10] for comparison. [09:11] the command.log says error.py:35] showing error view for: Top-level container OS did not initialize correctly. [09:13] while initializing it seems like that the deb repositories are not available but I check them manually they are up. [09:44] lordievader: I think you may be onto something - but, I don't think it's the actual disks. [09:44] mripguru: What did you find? [09:45] lordievader: just for the hell of it - I'm installing XenServer - which does not support FakeRAID [09:45] and it seems to be installing just fine. [09:45] mripguru: Why not start with some basic dd tests to check the disk speeds? [09:45] so looks like the issue may actually be in the FakeRAID [09:46] lordievader: can do that - but, I know the disks are fine (more or less) [09:46] Oh that can be, but perhaps there is something blocking Ubuntu from achieving the full disk speed (then you know why it is so slow ;) ) [09:48] lordievader: sounds like I should try turning off the FakeRAID and just building a true software RAID setup? [09:49] (after running some disk tests) [09:49] That is not what I am saying, I am saying you need to form a hypothesis then gather some data to see if it is correct. [09:49] Only then you can think of a solution. [09:50] Just doing things without reason is rather futile. [09:51] lordievader: agreed [09:55] lordievader: at this point - all roads lead either to the FakeRAID or the disks themselves - need to isolate each [09:55] mripguru: Exactly ;) === lordieva1er is now known as lordievader === ming is now known as Guest66148 === ReScO_ is now known as ReScO === peterrus- is now known as peterrus === mukhbiir is now known as neoark === spinza_ is now known as spinza === rsalveti_ is now known as rsalveti === broder_ is now known as broder === neurotus is now known as Guest56754 === _fortis_ is now known as _fortis [10:42] lordievader: I suspect part of the slowness may actually be the fact that I'm trying to install over IPMI/DRAC [10:43] SOL IPMI? [10:43] lordievader: no - some bastardized version of Supermicro IPMI [10:44] SOL = Serial Over Lan ;) [10:44] not serial [10:44] full graphical UI [10:44] Hmm.. Not sure. What kind of speeds does dd show? [10:47] Hello, I am having some trouble with mysql? I can't seem to connect to it from my client on my computer, I use the IP and the 3306 port and it just errors, it says communication link failure (I am using DBBeaver), any help would be awesome, it says in the /var/log/mysql/error.log that it is ready to be connected to... but nothing? its enabled in my firewall so what can I do? [10:48] dannysmc95: Is your mysql server listening on a port or a socket? [10:48] bekks how would I find out? [10:49] By looking at your config file. [10:49] [mysqld] ? [10:50] bekks: it says socket = /var.run.mysqld.sock and port = 3306 under that [10:50] dannysmc95: But there is no IP address configured to listen on? [10:51] It uses bind-address = 127.0.0.1 [10:52] So it is listening on localhost Port 3306 [10:52] Is 127.0.0.1 the IP you are trying to connect to? [10:52] No I am external so using the servers IP [10:52] Your mysql server isnt listening on that IP. [10:53] Okay but if I remove that IP would it not stop localhost access? [10:54] ? [10:55] 0.0.0.0 will make it listen on all address, a.b.c.d (your external IP) will make it listen on that IP only, 127.0.0.1 makes it listen on localhost only. [10:56] Okay so I should set bind-address to 0.0.0.0 [10:56] Only if you need to access it from localhost, too. [10:57] Yeah I would do, so I can use php? [10:57] you can use php using the socket, too. [10:58] But 0.0.0.0 should be fine. [10:58] bekks: access denied? [10:58] dannysmc95: Is that a question? [10:59] bekks: yeah its the correct user and password...? [10:59] Did you restart the mysql server after changing the config? [10:59] yeah [10:59] Then your credentials arent correct. [10:59] It works via CLI [11:00] same password and same user, works my doing mysql -u -p and then typing password? [11:00] Nope. [11:00] mysql -u user -p password -h host [11:01] Eh? I am saying when connected to SSH to my server it works by doing the command above and works fine? but when I am trying to connect from a client on my pc here it doesn;t? do I need to allow the host? [11:03] Sounds like a firewall is doing its job. You don't want you sql server exposed to the internet. [11:03] So if it works on your host, thats fine. Now you need to configure your server to allow connections from your client. [11:04] bekks: how exactly do I do that? I am very new to having a server? [11:05] dannysmc95: I dont know how/what you did on your server to inhibit connections from your client. [11:05] bekks: neither do I... [11:07] I did the grant all on command in mysql via ssh but it says 0 rows affected? [11:08] GOT IT [11:09] Thanks for the walkthrough, I needed to permit my host name [11:09] allow* [11:16] Anyone can help me with git? I am trying to let other people connect to my server securely [11:16] I think best with keys? === agumon is now known as Uptime === popey_ is now known as popey === Caveman is now known as Guest18416 [13:00] xnox: A friend from uni just messaged me with a question about OpenERP. /o\ [13:00] ERP :P [13:02] Odd_Bloke: =)))))))))) [13:02] ccccccegjekvhffrubfiujluhejujredutkbfbfbdrvi [13:02] sigh [13:03] * mripguru hates being the local 'IT guy' [13:03] Odd_Bloke: did you say: Oh dooo ask someone else? [13:03] =) [13:03] https://www.odoo.com/ [13:04] xnox: <3 === MrPPS is now known as Valfor [13:05] mripguru: In their defence, I (and xnox) did work for a company that developed OpenERP plugins about 5 years ago. [13:05] gosh, has it been that long already?! [13:05] that's not a defense - that's an excuse. [13:05] xnox: Maybe... 4? [13:05] *checks CV* [13:05] Odd_Bloke: i bet tinyerp modules are still valid in odoo [13:06] Odd_Bloke: i started in 2011 i think, and you must have been there before. [13:06] xnox: I can only assume you're trying to make me angry. :p [13:06] so in february it will be 5 years for me. [13:06] xnox: I left July 2011, I think. === arcsky_ is now known as arcsky [13:38] hi. anyone using packer to provision ubuntu 15.10 machines? [13:39] are you using 15.10 on servers? [13:39] I'm trying to provision a server machine, yes. [13:40] I hope this is not for production - it was released yesterday... [13:40] indeed :-) [13:40] and isn't LTS [13:40] and ... [13:40] oh well [13:40] This is for a buildbot / build farm. We're providing nginx.org packages built for different OSes. [13:41] ic [13:41] thresh: let me guess - the nginx.org repo? ;) [13:41] :P [13:41] yes [13:41] * teward|live whistles nonchalantly then throws the idea of sbuild out there [13:42] teward|live: last time I checked it could not build stuff on centos and sles :-) [13:42] thresh: no, it can't, i meant specifically for your Ubuntu/Debian builds [13:42] thresh: granted, though, if you're doing builds on individual 'machines', even virtual, i [13:42] i'm left scratching my head as to why [13:43] but that's my opinion :) [13:43] the idea is to have prebuilt images for all OSes we support and do builds in a clean env every time [13:43] (that's what sbuild does *cough cough* [13:43] ) [13:43] and on the real kernel [13:43] ah [13:44] 'kay then :) [13:44] * teward|live silently disappears back into the shadows [13:44] anyhow, there is a problem with packer and ubuntu 15.10 (while the exactly same config works for 15.04), so I'm kinda puzzled here. === rattking_ is now known as rattking === klaas_ is now known as klaas [14:39] uh, looks like whenever I launch ubuntu 15.10 server cd install, providing options in the console UI (clicking escape, escape, enter from main grub UI) with "boot: " gets me back to grub... [14:45] nice, and 15.04 does exactly the right thing when I type in "/install/vmlinuz auto initrd=/install/initrd.gz" -- it tries to boot up the installer.. [14:45] can anyone confirm that? [14:45] that's using http://releases.ubuntu.com/15.10/ubuntu-15.10-server-amd64.iso [15:09] thresh: sounds like grub wasn't installed correctly [15:10] RoyK: "installed" on a CD you mean? [15:10] this is when I boot the CD up. [15:10] oh [15:10] no, haven't seen that [15:12] thresh: actually, I don't think I've _ever_ seen that from a cd boot, but then, I've only used linux for 20 years or so ;) [15:13] well me neither and that's what's happening with 15.10 server isos (both i386 and amd64) [15:13] thresh: I'm install 15.10 in a VM as we speak [15:18] RoyK: can you try this: https://gist.github.com/thresheek/82a67847b5308be5b9df ? [15:18] those the instructions to reproduce what I have [15:25] thresh: why do you try to leave the installer in the installer? === markthomas|away is now known as markthomas [15:41] RoyK: I want to provide a preseed configuration file. [15:41] thresh: can you edit the existing command-line using F6, where a preseed is already listed? [15:47] TJ-: that works even if I remove all the parameters except "initrd=/install/initrd.gz" there and hit enter there. === Lcawte|Away is now known as Lcawte [16:04] thresh: I see the same thing here in a VM; what is weird is "boot: xxxxx" then Enter shows "Loading image...". I cannot find where it gets the word 'image' from; the linux.c32 code uses "Loading %s\n", kernel_name which should show xxxxx [16:04] thresh: also, if i repeat it several times iso/syslinux reports "Out of memory" apparently indicating it does manage to load the kernel image into RAM (using 1GB VM) [16:06] interesting. :/ [16:39] Samba user authentication seems to be broken after upgrading to 15.10 === Lcawte is now known as Lcawte|Away [17:25] vertago1_: what sort of authentication are you using? [18:22] hi can anyone please help. I'm trying to setup ubuntu 14.04 with a raid 10 using the ubuntu installer - so far i've just made a right mess of it each attempt! I followed the UrbanPenguin youtube video but he only does raid one and he uses an older version of ubuntu too so the guide just hasn't worked for me [18:22] any help really appreciated. [18:22] got a bootable usb with ubuntu server 14.04.3 on [18:22] and also got a bootable gparted usb [18:22] I'd hoped between the two it would work [18:23] I'm running ubuntu server 15.04 on my home server. After doing apt-get dist-upgrade, this happens when I try to boot - http://imgur.com/a/20zFG [18:23] It seems to be failing to boot with Ubuntu Kernel 3.19.0-31 and if I interupt grub during boot and select 3.19.0.15, my system boots just fine without the above errors. Anyone have any ideas where I should even start with this? Thanks! [18:23] jonah_: there's some information on the installer's raid stuff here https://help.ubuntu.com/14.04/serverguide/advanced-installation.html [18:25] sarnold: thanks this looks very similar to the youtube I followed. both mention the option of Primary drives being selectable but I had no option for Primary in 14.04... [18:26] sarnold: i'll try follow the guide you've linked and just see if it is any different or works better than the youtube one. I get all the system installed etc but it just fails at the end and can't install grub, also I don't know what path i'm supposed to enter for grub... [18:27] sarnold: the other problem I had when partitioning with the installer live usb is it wouldn't let me set the partitions as 'bootable' when I pressed enter to change it to bootable nothing happens. very weird [18:28] lownin: you might try #ubuntu-kernel, or file a bug against the 'linux' package [18:29] lownin: that second photo showing failing to bring the cpus online is really something else.. [18:29] sarnold: thanks, I'll repost in #ubuntu-kernel [18:29] sarnold: I agree. [18:29] jonah_: iirc you're only allowed four primary partitions with old-style formatting, not sure about the new GPT format.. [18:32] sarnold: i only have four drives, so i'd just made 4 swaps partitions (1 on each drive) and 4 root partitions. then just installed at that but it didn't like it! [18:35] the question is, how should you use those 4 :) [18:35] raid0? raid1? [18:36] jonah_, did you use mbr or gpt? [18:37] gpt requires you make additional boot areas for grub [18:39] sarnold: #ubuntu-kernel isn't very active, is it? [18:40] lownin: it may not be, I have to admit I've never used it.. but the topic looks kept up to date... [18:41] sarnold: I was reading through their channel logs, and there are entire days where no one says anything. [18:42] must be nice :) haha [18:42] =p [18:55] patdk-wk: sorry trying to do raid10 [19:01] just trying now with a 3rd raid volume which I've got a efi boot area set for each sda3 sdb3 sdc3 sdd3 etc... [19:01] no idea if that will work! [19:01] but at least it gives that a B for bootable label... [19:07] I'm having issues updating my server, not sure what to do - http://dpaste.com/0JFQG68 if anyone has time to take a look, ty [19:08] hey guys, could someone confirm if ubuntu supports automatic daylight savings time when using NTPD? [19:09] ntpd is unrelated to daylight savings. you need to have proper TZ file installed. === markthomas is now known as markthomas|away [19:10] jge: i can confirm. [19:11] maserati: try again after an apt-get update? [19:12] sarnold: i did =[ same result [19:15] jrwren: thanks [19:15] maserati: hmm, I wonder if something changed that might require apt-get dist-upgrade? [19:15] maserati: it's been ages since I've seen this kind of thing.. If it were mine, I think I'd delete the package lists and packages from /var/cache/apt/ and maybe clear any apt proxies you might be using, too, re-do the apt-get update, then try the apt-get upgrade again [19:16] maserati: another approach is to try to apt-get purge those two packages, making sure you don't actually delete the installed kernel packages, then try re-installing them.. [19:18] sarnold: Just tried option 1 there, same thing. trying option 2 === AppRoyale is now known as approyale === markthomas|away is now known as markthomas [21:04] Anyone want to help a guy figure out what the errors I am getting mean? [21:06] bernierunns: it always helps to just state what problems you're seeing and wait for someone to help [21:06] rather than just ask if anyone wants to help [21:06] heh === NegativeFlare_ is now known as NegativeFlare [21:08] ok Errors were encountered while processing: [21:08] linux-image-generic-lts-utopic === maserati is now known as FSMonster [21:09] among five other errors [21:20] What can I do about missing partition tables? I get an error when I boot the server that I have missing partition tables and it then asks to skip the mount or fix manually. [21:20] I just skip it because i have no idea how to fix this issue. === FSMonster is now known as maserati [21:21] if the partition table is suppose to be missing then you should remove that entry from /etc/fstab.. this will remove that prompt [21:24] OK, how do i display drive info so I can see whats going on with my drives? I'm fairly new to command line stuff. [21:25] 'blkid' 'cat /proc/partitions' and 'df' will give some ideas [21:25] ok, I'm checking that out now [21:27] but it should have complained about the missing device at boot [21:27] Could it be that I don't have an external drive configured to mount at boot? [21:28] huh, https://bugs.launchpad.net/ubuntu/+source/debian-installer/+bug/1505839 [21:28] Launchpad bug 1505839 in debian-installer (Ubuntu) "Unable to install from text mode interface" [Undecided,Confirmed] [21:29] its more likely there is a external drive configured to mount at boot that is either not attached or has been formatted so the UUID doesnt match whats in /etc/fstab [21:31] OK, that makes sense, I can edit that info in vi right? [21:31] yes, with sudo or as root === shadeslayer_ is now known as shadeslayer === keithzg__ is now known as keithzg === wolsen_ is now known as wolsen === Lcawte|Away is now known as Lcawte === Lcawte is now known as Lcawte|Away