[00:56] question about mdadm.conf. If I modify the current create statement from: "CREATE owner=root group=disk mode=0660 auto=yes" to "CREATE owner=MYUSER group=users mode=0660 auto=yes" and restart the computer, will it break anything? I believe the answer is no, I just want to be sure. I will then chown the current folder/array [00:56] or should I just leave that alone and chown it === markthomas|away is now known as markthomas === markthomas is now known as markthomas|away [04:29] anyone around? === mgagne is now known as Guest75711 [08:21] Good morning. [08:22] hey guys... ive got an apache virtual host... and was wondering if there was anyway to specify in the conf file that i want apache to serve the index.html file when i go to the site (ie instead of site/index.html) [08:23] ProbsNotCheraphy: It should automatically check the dir for an index file. [08:23] hmmm its not for some reason [08:24] You get a directory listing? [08:24] no it goes to some cgi thing [08:25] you can see it at omega.dtscode.io [08:26] ProbsNotCheraphy: I get an 'Hello, World! [08:26] what browser? [08:26] Firefox. [08:26] thats weird [08:27] On Chromium too. [08:27] ProbsNotCheraphy: Caching? [08:27] thats what im thinking now [08:28] yeah that was the issue [08:28] thanks lordievader [08:28] idk why it happened in the first place [08:29] ProbsNotCheraphy: ;) [09:09] Hi - is this 'freak' SSL bug anything to worry about in terms of servers or is it a client only issue ? [09:12] ie - http://it.slashdot.org/story/15/03/03/2036241/freak-attack-threatens-ssl-clients [09:30] hello [09:32] i have a dedicated server with mail and http services, i make backups often but i want to professionalize this service in some way, i wonder what is the standard to do this, -- rent an other server and use it as backup server when the main server fails? if so, how to make the on-the-fly copy data === kickinz1|afk is now known as kickinz1 === kickinz1 is now known as kickinz1|afk === kickinz1|afk is now known as kickinz1 [10:24] Hello experts, Installed Ubuntu 12.04.5 LTS on a SAS drive. NVMe drivers are detected with unknown model.Drives are not not showing using fdisk command .please advise [10:31] Hello experts, sorry to repost. Installed Ubuntu 12.04.5 LTS on a SAS drive. NVMe drivers are detected with unknown model.Drives are not not showing using fdisk command .please advise [10:47] Reddy: Anything in dmesg about the drives or the SAS controller? [10:48] lordievader: Parted –l result Model: DELL PERC H310 (scsi) Disk /dev/sda: 199GB Sector size (logical/physical): 512B/512B Partition Table: msdos [10:49] Model: Unknown (unknown) Disk /dev/nvme0n1: 1600GB Sector size (logical/physical): 512B/512B Partition Table: loop [10:49] Model: Unknown (unknown) Disk /dev/nvme1n1: 1600GB Sector size (logical/physical): 512B/512B Partition Table: loop [10:52] checking dmesg [10:54] I have no experience with SAS, but I'd say they are recognized. Should there be data on there? [10:57] installed OS on SAS drives, I want data to be placed on PCIe SSD for performance === Lcawte|Away is now known as Lcawte === Lcawte is now known as Lcawte|Away === cipi is now known as CiPi [13:01] Guys, is there a way/a tool that can port my linux system (ubuntu 12.04) from one server to another. I'm currently on digitalocean which has this 'droplet' containers and I wanted to move from (move the droplets) there to my own root server [13:03] hi there, anyone itc who can comment on Advantage (if you have it, do you feel like it's worth the price, etc)? [13:03] my company is thinking about purchasing it [13:23] adac: rsync? [13:25] RoyK, do you probably know a good howto on how to port linux with rsync? [13:31] adac: "port" usually refers to moving a system from one architecture to another. If you want to copy an entire installation, you can use rsync -avPHXA --delete / target:/ and then run grub-install on the target. If you haven't done this before, then it may be less hassle to just install a new system and setup the services again there [13:32] RoyK, ok I see [13:32] thank you very much [13:33] RoyK, does --delete really delete / ? [13:33] it deletes whatever file on the *target* that isn't present on the *source* (from where you run it) and thus creates a copy of it all [13:34] see man rsync === bilde2910|away is now known as bilde2910 [13:35] holy crap that will break stuff =) [13:35] good luck =) [13:36] OpenTokix: what? I've cloned systems that way several times... [13:37] ok - with configured services? [13:38] say you want to move system A to system B and don't want anything left of what's configured on system B (which can be any distro), yes, that works. [13:38] ok [13:38] it probably doesn't work too well if you're running selinux, but that's not common on debian/ubuntu, though [13:38] no, it runs the even more useless apparmor [13:39] I don't think apparmor will stop this. anyway - it doesn't touch the source system, so it can't do much harm [13:39] and again - I've used this to virtualise (p2v) systems several times [13:49] apparmor is nice, works great for me [13:49] you just run rsync unrestricted by apparmor, pretty simple [13:50] personally though, I'll normally just lvm snapshot, and dd the disk [13:50] Personally I am using a modern approach to my sysadmin and use configuration management [13:52] that actually helps you migrate the data? [13:52] if this was only about migrating the config, that is simple [14:03] Where should I go to get a hold of the openstack charms for Havana ? [14:04] need it to diagnose some issue [14:05] is 'old-stable' a correct pick ? === Lcawte|Away is now known as Lcawte === Guest75711 is now known as mgagne === mgagne is now known as Guest48046 === Guest48046 is now known as mgagne_ [14:10] hello there! any report of a libc recent update that breaks kerberos SSO in firefox/chromium? Namely, trying to sso on a webpage crashes the browsers. === mgagne_ is now known as Guest44965 [14:14] spidernik: had you rebooted the machine after libc update ? [14:15] (you may also find it apparmor related?_ === Guest44965 is now known as mgagne [14:19] yossarianuk: yes, freshly rebooted. I'll look at apparmor, thanks for the suggestion. [14:25] if we strace firefox we get: [14:25] writev(2, [{"/usr/lib/firefox/firefox", 24}, {": ", 2}, {"relocation error", 16}, {": ", 2}, {"/usr/lib/x86_64-linux-gnu/libgss"..., 45}, {": ", 2}, {"symbol krb5_cc_select, version k"..., 99}, {"", 0}, {"", 0}, {"\n", 1}], 10/usr/lib/firefox/firefox: relocation error: /usr/lib/x86_64-linux-gnu/libgssapi_krb5.so.2: symbol krb5_cc_select, version krb5_3_MIT not defined in file libkrb5.so.3 with link time reference [14:25] spidernik: you should possibly be able to see app armor messages in dmesg ot syslog [14:25] regarding spiderniks issue === beisner- is now known as beisner [14:28] /13 [14:30] Hello! [14:31] I'm trying to install the sql virtual user function of the mail server by seeing https://www.exratione.com/2014/05/a-mailserver-on-ubuntu-1404-postfix-dovecot-mysql/ [14:32] I want to keep my system account, too. But when I uncomment #!include auth-sql.conf.ext of /etc/dovecot/conf.d/10-auth.conf [14:32] , [14:33] >Mar 4 23:23:08 gateway dovecot: auth-worker: Fatal: master: service(auth-worker): child 14791 killed with signal 11 (core dumped) [14:33] >Mar 4 23:23:08 gateway dovecot: auth: Error: auth worker: Aborted request: Worker process died unexpectedly [14:33] happens... [14:43] Is there any good mangment tool for make daily mangment work with my ubuntu servers ~10-15 servers ? [15:00] arcsky: what kind of work? examples? [15:00] I would suggest puppet for changes to configuration/packages/etc [15:02] nevermind my silly question; I used a config file :-/ [15:08] Sling: config/packages/Security updates i want a simple mgmt system i have look for Landscape, ist bad? Puppet seems advance ? [15:08] hello, im here trying to make one pppd connection.. this is the thing.. i already made the connection, now i need to route the adress's 10.12.12.x for that pppd (ppp0) hwo can i do that ? [15:08] i dont understund well the foruns [15:09] arcsky: puppet is the future, landscape isn't ;) [15:09] so if you care about being relevant in IT in the future, I strongly suggest learning about puppet [15:09] even though it's a new way of thinking about server management [15:10] as for security updates: unattended-upgrades [15:10] Sling: wrong, - Config managment is the future. - puppet is a tool among many. - puppet dont scale well [15:10] lol [15:11] Learn about config managment, and use whatever tool [15:11] puppet is nearly a decade old ... i wonder whats "new" about it [15:11] landscape is very basic [15:12] OpenTokix: sure but if you can use puppet, you can use most other tools like it [15:12] anyway I don't want to turn this into the puppet vs chef vs discussion :) [15:12] use whatever you feel comfortable using, as long as its not manually logging into servers anymore [15:12] if you want to be relevant for the future you better look into cloud mgmt tools like juju ... :) [15:13] And version control === zz_DenBeiren is now known as DenBeiren [15:22] OpenTokix: what config mangment tool do you recommend? i looking for a simple [15:23] arcsky: the learning curve for it, is quite high - but it will make youre life so much easier in the long run. [15:23] I work with network and security dont wanna be a server guy ;) [15:24] arcsky: I would check out puppet, chef, cfengine, saltstack, ansible today. - I personally use cfengine3 since it suits me perfectly. [15:24] arcsky: I work with everything. [15:24] OpenTokix: ok thanks i will check them out [15:24] arcsky: puppet is the one with the most online resouces, and copy/paste-able stuff. [15:24] coolers [15:24] arcsky: however, it does not scale past 5000 hosts. - But if that isnt an issue you can ignore the "scaling issues" [15:25] not an issue === Lcawte is now known as Lcawte|Away === sonne_ is now known as sonne [15:45] Is it realistic for me to boot the vivid 3.19 kernel on trusty ARM 64? [15:52] frobware: you may want to ask this question at #hyperscale [15:53] Or possibly #ubuntu-arm, if anyone's currently awake in there [15:53] strikov, genii: ok & thx. [15:54] frobware: genii is right, #ubuntu-arm is the best place [15:54] strikov, heading there now. :) [16:11] HI,using ubuntu-server, according to this link " http://mpcabd.igeex.biz/python-arabic-text-reshaper/ " When I try to instal " l pip install https://github.com/mpcabd/python-arabic-reshaper/archive/master.zip " so it give this error http://pastebin.com/H1KtwfaY HELP please. [16:38] sarthor: as the error message says, there is no setup.py in that zip file; just unpack the zip in the directory you want it to be [16:45] i am uisn gubuntu 14.04.2 as host and i have 2 guest ,i am using openvswitch with bridge mode but many time openvswitch/brctl doesnt startup then i have to add eth1 to bridge and after that it work but not always [16:50] JanC, I unpacked and there is setup.py file. === martins-afk is now known as martinst === Lcawte|Away is now known as Lcawte === markthomas|away is now known as markthomas === Joel is now known as Guest10450 === Lcawte is now known as Lcawte|Away [19:42] hey, i want to use one package from debian-experimental. whats the best way to do this? [19:43] thor77: read about pinning, in apt_preferences(5) [19:44] thor77: What package? [19:44] genii: munin [19:44] thor77: I'd be tempted to just wget + dpkg -i the thing, but that would make future updates more difficult than it should be. might as well do it right... [19:44] genii: i want to use version 2.1 [19:44] genii: but ubuntu provides only 2.0 [19:47] thor77: a PPA maybe - https://launchpad.net/ubuntu/+ppas?name_filter=munin [19:47] but careful, anyone can bake one [19:49] pmatulis: cant find munin 2.1 there, only 2.0 or older :/ [19:49] thor77: bummer. you can make one yourself i suppose. it's not too hard. makes it complicated if the package has a lot of dependencies [19:50] http://packages.ubuntu.com/vivid/munin looks like it has a lot of dependencies [19:51] Yup. So that's why using a Debian repo to get it is probably asking for dependency hell [19:52] thor77: better compile munin .1 from scratch [19:52] thor77: better compile munin 2.1 from scratch [19:53] thor77: munin 2.1 is a development track [19:53] RoyK: munin-dev said to me i should use deb-repo because its a pain^^ [19:54] thor77: the do that in a vm [19:54] thor77: don't use deb-testing repos on a prod box [19:54] or deb-unstable - it breaks things [19:55] Interestingly 2.1.6 released March 9 2014, 2.0.25 released November 24 2014 ...so 2.025 is actually a later release [19:55] not very interesting - 2.0 is the stable track - 2.1 is the development track [19:55] but if their newest 'dev' is nearly a year old.. [19:56] sarnold: then it's probably better to use the code from git ;) [19:56] doesn't look like the munin guys are releasing much devel versions [19:57] RoyK: yeah, if it's any newer.. [19:57] http://demo.munin-monitoring.org/ <- 2.1 2.0 -> http://munin.crapwa.re [19:57] i like the new design [19:58] crapwa.re, hehe [19:58] i like my domain ;) [19:59] thor77: oh, heh, i missed that :) nice [19:59] sarnold: there are daily patches submitted to git [20:07] thor77: You might try apt-get build-dep munin for the version that's currently in Ubuntu, then wget the debian package and attempt dpkg -i on it and see if it groans [20:08] thor77: doesn't look like you've configured fastcgi correctly ;) [20:09] RoyK: yes, its a pain with nginx and i dont need it [20:09] http://munin.crapwa.re/static/dynazoom.html?cgiurl_graph=/munin-cgi/munin-cgi-graph&plugin_name=crapwa.re/fire.crapwa.re/diskstats_iops/xvda&size_x=800&size_y=400&start_epoch=1424808308&stop_epoch=1425499508 [20:09] zooming doesn't work without it [20:09] i know [20:09] but i dont need zooming [20:09] I use that regularly [20:10] or better: i tried to configure it a few times with no success and no i gave up because i dont really need this [20:11] but if you have a working configuration with nginx, i would be happy if you would send it to me :) [20:11] * RoyK has 100+ machines in munin and tend to need it some times [20:12] do you have one with nginx? :P [20:12] just using apache for this - it's dead easy to configure and with 100+ machines monitored, it won't make a difference performance wise [20:12] nope - I rarely use nginx - just for small things - better try to ask on #munin @OFTC [20:14] thor77: why do you need 2.1? [20:14] RoyK: new beautiful design [20:16] a little more flashy design isn't really a lot compared to the dependency hell you might be having by adding experimental repos [20:17] RoyK: maybe you're right. i think i will wait until the update is available in the official repos :) [20:18] and maybe i will setup apache, php is much less pain with it [20:18] last time I tested php with nginx, it was rather easy [20:19] used it with wordpress [20:19] but ditched nginx because of some modules I use with apache [20:25] until know i found nothing that doesnt work with nginx but does with apache and i like the configuration from nginx [20:28] * RoyK sets up nginx vm for testing on 1404 === bilde2910 is now known as bilde2910|away === bilde2910|away is now known as bilde2910 === kickinz1 is now known as kickinz1|afk === martinst is now known as martins-afk === Lcawte|Away is now known as Lcawte [22:17] hey all, does anyone in here happen to know if there exists an nrpe plugin to monitor hdd temperatures? === martins-afk is now known as martinst [22:29] sudormrf: there is an lmsensors nrpe module [22:30] ikonia, can lmsensors read hdd temps? I installed lmsensors and it was only looking at stuff on the mobo [22:30] the hd temps are fed to the system over the motherboard [22:30] hmm. when I run "sensors" and do the initial setup, then run the command again the individual drive temps are not returned [22:31] not all motherboards monitor hd temps [22:32] depends on the controller chipset if the hard disk can report back [22:32] gotcha [22:32] well it looks like mine doesn't. [22:32] hddtemp returns the data (as does smartmontools), would prefer the data from smartmontool [22:36] trying to install this one: https://github.com/vint21h/nagios-check-hddtemp#installation but the installation instructions are lacking. I ran the steps listed, but am not seeing what I think should be right [23:17] ok, next question. smartctl works locally on the machine, however nagios is saying this" CRITICAL - SMART_ENABLE: Inappropriate ioctl for device " [23:18] this drive IS part of an mdadm array. [23:18] however it is also saying that for a drive that is not part of that array [23:18] sudormrf: smartctl does not work on mdadm devices. [23:18] It works on physical devices only. [23:19] bekks, the command is checking /dev/sda [23:20] instead of /dev/md0 [23:20] running smartctl -a /dev/sda on the device it works [23:20] And is that exactly the command which is run by nagios? [23:20] good question [23:21] let me check [23:22] hmm. looks slightly different. [23:22] maybe? not sure [23:27] running the plugin against the drive manually, not with nrpe, it returns reslts [23:27] results [23:27] odd [23:52] is there a way to set up a mailing list / alias feature in Postfix, or do i have to set up mailman and dovecot and postfix [23:54] teward: http://www.postfix.org/aliases.5.html [23:55] sarnold: any other setup needed, such as DNS changes? [23:55] * teward is not sure what else he needs :/ [23:56] teward: depends, this is very primitive, but doesn't require much. [23:56] mmm [23:56] mail to foo@example gets exp0loded... that's it. [23:57] oh, dear... [23:57] welp, that explains that... [23:57] "Network is unreachable" [23:57] i wonder if my rules are too restrictive [23:58] ooo also timeout [23:58] s [23:58] no wonder its not working === markthomas is now known as markthomas|away