[01:56] <EuaD1> i was trying to install owncloud on my xubuntu 14.04.2 install and apparently i messed up my apache2 install because now my MythTV web interface doesn't work
[01:57] <EuaD1> i now get The server encountered an internal error or misconfiguration and was unable to complete your request. when i try to connect mythweb
[02:03] <Jeremy26> Hi.  We're switching all our servers to Ubuntu-servers.  I'm setting up my 1st one -- currently in the middle of install.  It's a UEFI-mobo, with disks partitioned with a 300MB EFI partition in addition to the usual install locations.  Started the install, going OK, until it gets to the GRUB install.  It fails @: "[1] Install the GRUB bootloader on a hard disk" just telling me an "Installation Step Failed".  I can continue with no no bootloader
[02:03] <Jeremy26> apparently, but I figure there should be a way to do this right from the installer.
[02:03] <Jeremy26> How do I get the installer to install UEFI+Grub2 stuff?
[02:04] <sarnold> EuaD1: note that some apache config7uration diredctives changed bwtween the apache in 12.04 LTS and 14.04 LTS... ifyou were following a guide, make sure it's for the right version..
[02:07] <sarnold> Jeremy26: wild-guess time, maybe grab a terminal and run grub-install by hand? I can't recall if it just does the right thing based on your configs or not.. but it'd be one potential starting point
[02:08] <sarnold> Jeremy26: which install disc are you using? apparently 12.04.1 and earlier didn't do uefi so well...
[02:11] <Jeremy26> sarnold: Hi.  Using vivid-server 15/latest, atm
[02:11] <EuaD1> sarnold, i had mythtv and apache2 running for years and was running fine on 14.04 but as soon as i installed owncloud the mythweb web panel is no longer accieble
[02:12] <sarnold> Jeremy26: oh, that really ought to work :)
[02:12] <Jeremy26> sarnold: by "grab a terminal", from where I am @ installer step?  or continue without bootloader, and "do it" later?
[02:13] <sarnold> Jeremy26: I was thinking from where you are in the installer.. hopefully enough is installed at that point to have a useful shell available
[02:13] <sarnold> EuaD1: aha. try checking apache's logs for erro messages..
[02:13] <sarnold> there are probably more details there
[02:17] <EuaD1> sarnold, ok, according to the error.log, it says authtype digest configured without corresponding module
[02:20] <sarnold> EuaD1: auth changed the most in apache 2.4.. lookhere for starting point http://httpd.apache.org/docs/current/mod/mod_auth_digest.html
[02:21] <EuaD1> sarnold, ok, thanks. i just removed and purged owncloud and it still says internal error. so i'll read this over, thanks
[02:28] <Jeremy26> sarnold: ok.  i need to figure out what to DO to get the grub-uefi done right once I'm in the shell ...
[02:28] <sarnold> Jeremy26: I'm hoping  that a plain "grub-install" would do it :) but that's blatant hopefulism on my part
[02:30] <Jeremy26> sarnold: i'm sure you're nice to kittens and puppies and all, but I'm hoping for some docs ;-)
[02:31] <sarnold> Jeremy26: looks like you're right, it looks like it requires the destination device name and probably --uefi-secure-boot parameter http://manpages.ubuntu.com/manpages/utopic/man8/grub-install.8.html
[02:38] <Jeremy26> sarnold: more here, https://help.ubuntu.com/community/Grub2/Installing#UEFI, but still rather thin.  You'd think an installer would handle this for you.
[02:38] <Jeremy26>  Well, s|think|hope|g anyway.
[02:39] <sarnold> Jeremy26: yes, yes I would, and I suspect it often does :)
[02:41] <Jeremy26> sarnold: That's what has me a bit hesitant.  There's nothing weird here.  It's  a standard partition setup with the UEFI bits.  Wondering if something 'sick' here that's preventing the install.
[02:41] <Jeremy26> Oh, Arch's docs come thru a bit more: https://wiki.archlinux.org/index.php/GRUB#Alternative_install_method
[02:45] <Jeremy26> So, I'm at the "Installation step failed".  Do I *first* drop to shell, then install? or 'continue' past this without installing grub, and invoke the shell in a step or two?
[02:45] <sarnold> I'd try it now.. you know where you stand right now, hehe
[02:47] <Jeremy26> sarnold: You presume a lot with that "know" biz, but ok ...
[02:50] <sarnold> Jeremy26: who knows what the next screen will bring? a big failure of some sort that refuses to contuinue? or salvation? or a lengthy package install that's pointless because youy can't boot it anyway? :)
[02:52] <Jeremy26> sarnold: minor problem ... atm, at shell, there's no 'grub-install' available. hmmm .
[02:53] <sarnold> Jeremy26: grumble. sorry.
[02:53] <sarnold> Jeremy26: time for me to bail.. good luck :)
[02:54] <Jeremy26> re-grumble. thx.
[03:03] <Jeremy26> Well, the next step was, effectively, 'finish the installation', then reboot.  @reboot, I get dropped to a Buaybox/ash shell, at prompt "(initramfs)", with absolutley no keyboard responses.  I.e. hosed.
[03:03] <Jeremy26> Not exactly a stellar "first time"
[03:23] <Jeremy26> wow.  'rescue' mode is even worse :-(
[03:24] <Jeremy26> What to do now?  Do I have to start over and reinstall?
[04:43] <EuaD1> looking at synaptic, i don't see Apache Module mod_auth_digest anywhere. i would really like to get mythweb working again
[04:49] <ObrienDave> !info mod_auth_digest
[04:51] <EuaD1> why am i getting this error then? AuthType Digest configured without corresponding module
[07:31] <nivv> OpenTokix: FYI, when I did the dry run it stopped at the grub prompt, that was the reason I couldn't reach it by SSH
[07:55] <OpenTokix> nivv: ok, so its up now?
[07:55] <nivv> OpenTokix: Yup! Just connected it to a screen and pressed enter. And now on the productionserver it upgraded perfectly! I'm a happy camper
[07:55] <OpenTokix> nivv: awesome! ^5
[07:56] <nivv> Just forgot that custom PPA's need to be updated too
[08:05] <nivv> OpenTokix: do you know why that happens? Shouldn't it just bypass that since it's a headless server?
[08:08] <OpenTokix> nivv: There is an issue, that was discussed last week actually. https://bugs.launchpad.net/ubuntu/+source/grub2/+bug/797544 <-- this very strange default in 14.04 - I have changed it in my config managment now, so I always have a timeout set.
[08:09] <nivv> OpenTokix: ah I see. Did you change your /etc/grub.d/00_header? I'm a bit scared to reboot the server (haven't done so after the upgrade)
[08:10] <OpenTokix> nivv: no, I changed in /etc/defaults/grub and reloaded update-grub2
[08:35] <amarcolino> Hi, I am running a clean install of ubuntu 14.4 with bonding (bonding-mode 4), the system can update and access the internet via w3m, however, I can't seem to ping other computers on the local network and vice-versa, anyone can help me?
[08:37] <OpenTokix> amarcolino: your switch is configured to handle 802.3ad?
[08:37] <OpenTokix> amarcolino: Can you ping the hosts if you unbond the interfaces?
[08:38] <amarcolino> ... hmmm... haven't tried to unbound the nics and no since I ain't using a switch , it connects directly to the router, give me a sec.
[08:40] <OpenTokix> amarcolino: router is also a switch
[08:40] <OpenTokix> I assume its your home router? yes? Or is a commercial router?
[08:41] <amarcolino> plain home router and I wouldn't know whether it can handle 802.3ad, it works accessing the internet but the local network :s, that's a pain not being able to connect to it
[08:42] <OpenTokix> ok, doubt it support 802.3ad, and if it does you have to enable it probably.
[08:43] <amarcolino> nope just unbound the nic, so using only one and I still can't ping
[08:43] <OpenTokix> amarcolino: ok, - is the host you're trying to ping also a linux machine?
[08:44] <amarcolino> nope it is a windows machine, tried both ways and neither can ping, I suspect it is the linux instead of the windows
[08:45] <OpenTokix> ok, no firewall enabled?
[08:46] <amarcolino> nope
[08:46] <OpenTokix> so, run tcpdump on the interface when you ping, to see if you get anything back
[08:48] <amarcolino> how do i use tcpdump, i.e. tcpdump [ping address],?
[08:50] <OpenTokix> on your local machine
[08:50] <OpenTokix> if your interface you are ping:ing from is eth0
[08:51] <OpenTokix> as root: tcpdump -i eth0 icmp
[08:53] <amarcolino> OpenTokix: once I use tcpdump and ping on the windows machine it works
[08:53] <amarcolino> so there must be something blocking it
[08:53] <OpenTokix> yes
[08:53] <OpenTokix> probably the default firewall on the windows machine
[08:54] <amarcolino> weird now the windows machine can ping but the ubuntu cant ping the windows, any suggestions?
[08:54] <OpenTokix> windows firewall
[08:55] <OpenTokix> Enabled by default and blocks everything
[08:56] <amarcolino> going to try another laptop and see if the same happens, it is weird to not be able to ping a clean ubuntu install
[08:56] <OpenTokix> since for me at least, that folder is not handled by my git repo
[08:56] <OpenTokix> wrong window
[09:02] <amarcolino> OpenTokix: for some reason my windows 7 laptop is blocking ping connection, yet windows 8 laptop is fine, grrr. Need to find out why, I most likely did something when I setup this laptop since it is the one I use often.
[09:02] <amarcolino> OpenTokix: thanks for the help it was driving me insane
[09:02] <OpenTokix> amarcolino: depends on what you select for the network setup. Like "home network" or "public network" That will set other settings in the firewwall on windows.
[09:02] <OpenTokix> Why do I even know this? I dont use windows.
[09:05] <amarcolino> it is setup as home, it is probably one of the services I disabled or something I enabled/disabled in the administrator config, sigh, this will take a while :'(
[09:06] <OpenTokix> amarcolino: Not like ping is crucial to the function of your ipv4 network
[09:09] <ReScO> hey peeps
[09:09] <ReScO> I'm having some trouble setting up my firewall...
[09:11] <ReScO> i have fail2ban, but i want to close pretty much every port except 22,80,443,587,993 and 3000, how can i do this in such a way, a portscanner will think it's offline?
[09:12] <ReScO> and i want to open port 22 for certain IP's only.
[09:16] <amarcolino> OpenTokix: so far I can ping from windows 7 to windows 8 but not the other way and if I use bonding ubuntu cant receive or send pings on the local
[09:20] <amarcolino> OpenTokix: this is strange just done a reboot with a single nic functioning and tried pinging the server from both laptops and nada, can you help?
[10:45] <Spyidonas> quick question , if a folder served by apache is on www-data group and user, and its file permissions are 777 shouldn't the folder be writeable by php?
[10:46] <rbasak> Depends on directory permissions all the way up - the php user must be able to get to the directory in question.
[10:46] <OpenTokix> Spyidonas: You should geneerally never have a folder set to 777, unless its a upload folder or such. - if www-data owns it. - 775 whould be enoigh
[10:46] <rbasak> Also, check for apparmor denials
[10:47] <Spyidonas> i set it to 777 because 775 doenst work
[10:47] <rbasak> (not sure what the default is there)
[10:47] <rbasak> Spyidonas: that's fine for debugging a test deployment but be sure to change it back before going to production
[10:48] <Spyidonas> i havent modify any apparmor, does the ubuntu 12.04 default denies write? the thing is folder above the one i want have read/write access
[10:48] <Spyidonas> for example /var/www/uploads is readable-writeable
[10:48] <Spyidonas> and the /var/www/upload/test isn't
[10:57] <Kartagis> Spyidonas: never ever use 777. change the directory's ownership
[11:02] <Spyidonas> Kartagis: i did it for test because nothing works.
[11:03] <Spyidonas> Kartagis: not even 777 apparently
[11:07] <Kartagis> Spyidonas: never ever use 777. change the directory's ownership
[11:08] <Spyidonas> Kartagis: to www-data , correct? Still doesn't work
[11:09] <Kartagis> Spyidonas: is this wordpress?
[11:09] <Kartagis> Spyidonas: is it also php-fpm?
[11:09] <Spyidonas> Kartagis: no, its an existing site and im adding a webmail
[11:09] <Spyidonas> Kartagis: the existing site has upload options and php reads/writes to folder
[11:10] <Kartagis> because php doesn't run under a different user
[11:10] <Kartagis> whereas php-fpm does
[11:10] <Spyidonas> Kartagis: i created a new folder throwed everything the webmail has and i have the error i cant read/write
[11:10] <Spyidonas> Kartagis: it's the standard php/apache that comes with ubuntu 12.04
[11:11] <Kartagis> have you read the <webmail> docs?
[11:11] <Kartagis> is it horde?
[11:11] <Spyidonas> Kartagis:no its the afterlogic
[11:12] <Kartagis> you are on ubuntu, right? check what user the webserver is running under with ps aux|grep apache
[11:12] <Spyidonas> Kartagis:their site http://www.afterlogic.com/support/faq-webmail-pro-php#3.1 says everything about permissions
[11:13] <Spyidonas> Kartagis:apache runs under root...
[11:14] <Kartagis> that is a Very Bad Idea™
[11:14] <rbasak> dannf, hallyn: what are you expecting from me on bug 1427406? Do we have a patch for Vivid that works on all architectures that we're ready to apply?
[11:14] <Spyidonas> Kartagis: :S I haven't configure that server, my task is to simply activate this webmail clinet :/
[11:14] <Kartagis> edit the apache configuration file for user
[11:15] <Kartagis> Spyidonas: /etc/apache2/envvars
[11:16] <Kartagis> export APACHE_RUN_USER=www-data
[11:16] <Kartagis> export APACHE_RUN_GROUP=www-data
[11:16] <Kartagis> should be that
[11:16] <Spyidonas> Kartagis: will it mess up the existing site ?
[11:16] <Kartagis> oh, and you actually should have a user www-data
[11:16] <Spyidonas> Kartagis: i have a user www-date
[11:17] <Kartagis> I don't think so, but shit happens
[11:17] <Kartagis> then nope
[11:18] <Spyidonas> Kartagis: the exports are as you typed them
[11:19] <Spyidonas> Kartagis: without modification
[11:19] <Kartagis> one sec
[11:19] <EuaD1> is anyone in here familar with znc server?
[11:20] <Kartagis> User ${APACHE_RUN_USER}
[11:20] <Kartagis> Group ${APACHE_RUN_GROUP}
[11:20] <Kartagis> Spyidonas: put these in /etc/apache2/apache2.conf
[11:20] <Kartagis> EuaD1: you mean the BNC?
[11:22] <EuaD1> Kartagis, it's called ZNC
[11:22] <EuaD1> I have a linode server running a ZNC server
[11:22] <Kartagis> EuaD1: the bouncer, right?
[11:22] <EuaD1> yes
[11:22] <Kartagis> EuaD1: what do you need help with?
[11:22] <Kartagis> I am on znc
[11:24] <EuaD1> well, it's sort of a funny issue but im logged into my znc server, looking at the web admin panel my user has 3 networks. i seem to have forgotten how to resume as my user because when I try to switch to that nick it says that user is already logged in
[11:25] <Kartagis> logged in as connected to a network?
[11:27] <EuaD1> Kartagis, yes. the web panel shows i'm connected to only 1 network
[11:27] <Kartagis> EuaD1: did you change usernames?
[11:27] <Kartagis> on your irc client*
[11:27] <EuaD1> Kartagis, yes
[11:28] <EuaD1> for example. if you hop into #linuxtechandgaming on freenode you can see me logged in multiple times
[11:29] <Kartagis> EuaD1: can you kill all your users and start anew?
[11:30] <Kartagis> aka fresh start
[11:30] <EuaD1> Kartagis, i'm not entirely sure how to be honest.
[11:30] <EuaD1> i wanted to set up a bouncer so that I would never miss a conversation
[11:30] <Kartagis> /msg *status help
[11:32] <EuaD1> shit, i just realized what time it was. Kartagis i need to get going to get to work on time. sorry i need to run
[11:32] <EuaD1> Kartagis, i sent you a direct message showing what it returned. maybe i'll talk to you later for help. thanks
[11:32] <Kartagis> EuaD1: /msg *status restart should do it
[11:36] <Kartagis> Spyidonas: are you all good? I need to go to lunch if you are
[11:37] <Fun> hi folks
[11:38] <Fun> who here knows decent uk vps provider?
[11:38] <Fun> :D
[11:39] <Kartagis> Fun: DO provides UK locations IIRC
[11:40] <Fun> true
[11:40] <Fun> any else? :D
[11:40] <Fun> I like ovh however they dont offer UK
[11:54] <Spyidonas> Kartagis: are you still there?
[11:58]  * Patrickdk has ovh blacklisted on all his servers
[11:59] <Patrickdk> never seen anything ligit from that network
[13:20] <Jeremy26> can anyone verify whether ubuntu-server installation disk (vivid) is supposed to be bootable as UEFI, and that it should be able to install grub2+UEFI to a target system?  I can't manage to do either one.
[14:30] <hallyn> rbasak: we're looking for guidance as to who to contact who would be able to write the patch we need
[14:30] <hallyn> (else we'll just have ot wait until next week and i may have time to look at it in depth)
[14:30] <hallyn> we can discuss it during team meeting
[14:30] <hallyn> the bug is listed as an action so it'll come up
[14:31] <Jeremy26> fyi, the ubuntu-server vivid/current installer fails to correctly detect/use DHCP setting, requiring manual network setup.  other-OS installers and rescue disks have no DHCP problems.
[14:41] <dannf> rbasak: no known fix yet :(
[14:42] <dannf> rbasak: i told hallyn i'd look at it again, but i need to finish another task first
[15:35] <designbybeck__> I'm trying to update one of my test servers and I'm getting this: https://gist.github.com/anonymous/d2d0fe04998b82521505
[15:36] <designbybeck__> sudo: unable to open /var/lib/sudo/beck/3: No such file or directory
[15:37] <designbybeck__> my main user is what I ssh'd in with, it should be a sudoer
[15:37] <designbybeck__> just tried as root and got the same thing
[15:37] <designbybeck__> odd
[15:40] <jamespage> zul, I'm seeing quite a few python3 isms creeping into openstack - assertRegex instead of assertMatchesRegex for example
[15:40] <jamespage> something in the dependency chain is dealing with that upstream, but not in Ubuntu
[15:40] <jamespage> any ideas?
[15:40] <zul> not off the top of my head
[15:41] <zul> jamespage:  oslotest maybe
[15:42] <jamespage> zul, hmmm we are quite out-of-date there
[15:44] <jamespage> zul, nope not that
[15:44] <zul> jamespage:  then not off the top of my head
[15:44] <jamespage> np
[15:45] <jamespage> zul, can I steal you ironic branch for testing? did you look at trove?
[15:45] <zul> jamespage:  i pushed to ~ubuntu-server-dev/ironic/kilo already
[15:45] <zul> lemme push my trove stuff somewhere
[15:46] <zul> jamespage:  https://code.launchpad.net/~zulcss/trove/kilo-3
[17:12] <yossarianuk> hi - I have an upstart issue
[17:13] <yossarianuk> I have created an upstart file -(for starting a meteor app) - it works fine via 'service meteor start/stop'
[17:14] <yossarianuk> but doesn't start on boot
[17:14] <yossarianuk> I can 'cheat' and add the line 'service meteor start'  to rc.local and that starts on boot
[17:15] <yossarianuk> but how can I work out why the upstart job /etc/init/meteor.conf doesn't start on boot ?
[17:16] <yossarianuk> do you have to tell upstart somehow taht the service should start on boot ?  i.e the equivalent on 'update-rc.d ... ... '
[17:16] <rbasak> yossarianuk: are you aware of http://upstart.ubuntu.com/cookbook/?
[17:17] <rbasak> You need to tell upstart what other event to start your job on.
[17:17] <rbasak> For example:
[17:17] <rbasak> start on runlevel [2345]
[17:17] <rbasak> stop on runlevel [!2345]
[17:17] <rbasak> Then it'll get started on when the runlevel changes to 2.
[17:23] <rbasak> hallyn: so I've got to task switch to Docker for now, and probably won't be able to look again at apparmor-profile-load until next week. I hope that's OK.
[17:26] <an3k> I'm running Ubuntu Server 14.04 and want to set display resolution and screen resolution for grub as well as console to 1024x768. With the default settings I get a display and screen resolution of 800x600 for grub and a display resolution of 1024x768 and a screen resolution of 800x600 for console.
[17:26] <an3k> I tried setting GRUB_GFXMODE=1024x768 but that didn't changed anything. I added GRUB_GFXPAYLOAD=1024x768 but that didn't changed anything too.
[17:26] <an3k> I also modified /boot/grub/grub.cfg manually and set gfxmode=1024x768 and set gfxpayload=1024x768 but that also didn't worked.
[17:26] <an3k> I also modified /etc/grub.d/00_header (based on http://askubuntu.com/questions/18444/how-do-i-increase-console-mode-resolution/18463#18463) but even that didn't worked.
[17:27] <genii> Did you do sudo update-grub after you made those changes?
[17:27] <an3k> yes
[17:28] <an3k> after all of that didn't worked I undid all modifications and enabled the nvidiafb and set GRUB_CMDLINE_LINUX_DEFAULT="video=nvidiafb:mode_option=1024x768-16"
[17:28] <an3k> nothing changed.
[17:29] <an3k> After I manually ran fbset -xres 1024 -yres 768 -match the console was finally in 1024x768 but after the next reboot it's again in 800x600
[17:30] <genii> an3k: If the fbset worked manually you could just add that command to your rc.local file
[17:30] <hallyn> rbasak: ah, ok.  you don't mind if i step in then?  oh wait, i'm out the next two days, so i woudn't be speeding it up much...
[17:30] <hallyn> rbasak: are you around on friday?  can i ping you then to get the status of your debdiffs, and I'll take it fro mthere?
[17:31] <rbasak> hallyn: sure, I'll be around.
[17:31] <hallyn> rbasak: thanks!  i appreciate it.
[17:31] <an3k> genii: thanks, I'll do that. Howver, shouldn't be th there a configuration for setting the resolution?
[17:31] <hallyn> i should be albe to finish it up fri/sat then
[17:32] <yossarianuk> rbasak: thanks I already have  'start on (started mongodb and started nginx and runlevel [2345])'
[17:32] <an3k> like GRUB_GFXMODE= in /etc/default/grub
[17:32] <yossarianuk> off now anyway - i'll look tomorrow...
[17:32] <genii> an3k: Instead of the GFXMODE or GFXPAYLOAD methos you could try the old way of VGA=xxx in the GRUB_CMDLINE_LINUX_DEFAULT
[17:33] <genii> an3k: You'd need to look up the 3 numbers to use for the mode you want
[17:33] <an3k> genii: tried that. All it did was keeping grub in 800x600 and console in 800x600 (1024x768) but kernel log in 720x400 ...
[17:34] <an3k> 791 which is 1024x768x16
[17:35] <genii> an3k: I'm out of immediate ideas on the issue.
[18:04] <rbasak> arosales: did you mean to publish the last minutes on ubuntuserver.org? I see your post but it's still listed as a draft.
[18:04] <arosales> rbask, I may have forgetting to press publish.  Let me check the fomatting and click publish
[18:08] <Jeremy26> Has anyone successfully installed ubuntu-server with Grub2+UEFI?  Every time fails here -- simply get to the "Install step failed" telling me no bootloader's installed.  End up with an unbootable system if I continue.
[18:17] <pmatulis> Jeremy26: it depends on the EFI implementation
[18:18] <arosales> rbasak, thanks for the reminder, http://www.ubuntuserver.org/?p=1243 posted
[18:19] <Jeremy26> pmatulis: Hi.  Can you explain that?
[18:19] <an3k> genii: looks like it's because of nouveau ...
[18:25] <an3k> funfact: grub runs in 1280x1024 with a 15 years old graphics card but only in 800x600 with a 5 years old nvidia card ...
[18:56] <an3k> genii: I'm now using a ATI MACH64 with 2 MB RAM and PCI ... all problems solved.
[18:57] <genii> an3k: Cool.
[18:57] <hxm> hello, my ubuntu works slow, for login ssh needs 1 minute and for see the pages i host some of times it needs 2 minutes, i dont see any process in top that eats cpu or memory and logs says nothing, what can i debug?
[19:00] <genii> hackemate: Check /var/log/auth.log for brute-force password attempts, and /var/log/apache2/error.log for people trying to reach admin pages for applications. You might want to install fail2ban which will limit attempts
[19:03] <sarnold> hackemate: wouldn't hurt to try some ping or mtr or smokeping to find out your latency, packetloss, etc
[19:04] <hackemate> i see many lines like this Mar 24 17:43:09 datatec CRON[9587]: pam_unix(cron:session): session closed for user root
[19:04] <hackemate> Mar 24 17:43:56 datatec CRON[9615]: pam_unix(cron:session): session closed for user root
[19:04] <hackemate> fail2ban is installed
[19:04] <hackemate> but seems does nothing really
[19:04] <hackemate> pam_unix is ssh?
[19:04] <hackemate> i did, average is ime=178.513 ms top was 300
[19:05] <sarnold> the important bit there is that that's a cronjob, probably not a big deal
[19:06] <genii> hackemate: For fail2ban maybe check here: https://www.digitalocean.com/community/tutorials/how-to-install-and-use-fail2ban-on-ubuntu-14-04
[19:06] <sarnold> 130 ms to where?
[19:06] <sarnold> I can get from west coast of the united states to japan in 100ms
[19:07] <hackemate> ping 37.187.42.205
[19:07] <sarnold> west coast US to UK is 190 ms
[19:07] <hackemate> right now is under 100ms
[19:07] <hackemate> from spain to france
[19:07] <genii> hackemate: Is the ping time from the remote machine to the server or from the server to some other machine?
[19:08] <hackemate> from remote machine to the server, this is what i see in localhost 64 bytes from localhost.localdomain (127.0.0.1): icmp_seq=1 ttl=64 time=0.099 ms
[19:08] <genii> ping time to localhost is meaningless
[19:10] <sarnold> yeah, but from spain to france, 100ms seems tolerable
[19:10] <sarnold> I got 163 from west coast us, it's closer than uk for me :)
[19:11] <genii> hackemate: From ssh session on the server, if you do ping -c 10 google.com       ... what sort of lag on that?
[19:13] <hackemate> wow i got stuck 5 minutes
[19:13] <hackemate> i mean 4 minutes
[19:14] <hackemate> now works fine, again
[19:14] <hackemate> but sometimes is like cpu is 100%
[19:15] <hackemate> the google ping is this 64 bytes from muc03s07-in-f6.1e100.net (173.194.44.6): icmp_seq=1 ttl=54 time=17.1 ms
[19:19] <sarnold> how about all ten pings?
[19:20] <hxm> 64 bytes from muc03s07-in-f5.1e100.net (173.194.44.5): icmp_seq=1 ttl=54 time=17.4 ms
[19:20] <hxm> all like this
[19:20] <hxm> i think is not networking problem, looks like cpu 100%
[19:20] <hxm> i have 2 ssh connected and sometimes i cant write on them
[19:20] <hxm> and suddenly it starts to respond again
[19:21] <hxm> auth.log is filled of pam_unix  sessions from cron
[19:23] <genii> hxm: How far apart is your most often cron job? For instance, sometimes you see a problem like a cronjob running every 5 minutes that actually takes more than 5 minutes to do
[19:24] <hxm> i go check that
[19:24] <sarnold> "filled"? that;s not good.. they shuold happen periodically...
[19:25] <hxm> im talking about 5-10 per second,
[19:25] <sarnold> WHAT??
[19:25] <hxm> dont scare me
[19:25] <hxm> i removed all cronjobs
[19:26] <sarnold> and did you look through the /etc/cron* files and directories?
[19:27] <hxm> man type a cd takes eternal, im getting paranoid
[19:29] <hxm> there are some files in cron.d, nothing in else cron.*
[19:30] <hxm> in cron.d i have apache2 aptitude and logrotate, roundcube updatenotifier
[19:31] <genii> Hm
[19:31] <hxm> norfix.es is still getting minutes to load
[19:32] <sarnold> hxm: how about /etc/crontab?
[19:32] <hxm> 47 * * * * root cd / && run-parts --report /etc/cron.hourly and  12 30 1 * * * root test -x /usr/sbin/anacron || ( cd / && run-parts --report /etc/cron.daily ) same for weekly and monthly
[19:33] <hxm> i candisable them too
[19:33] <sarnold> hxm: those are fine..
[19:34] <sarnold> I suggest reading this.. it might be jumping to conclusions, but .. https://www.fireeye.com/blog/threat-research/2015/02/anatomy_of_a_brutef.html
[19:34] <genii> Is there anything abnormal like i/o errors if you do: dmesg | tail -n 50
[19:35] <hxm> dmesg is empty
[19:35] <hxm> https://www.dropbox.com/s/09fpcuswhef4flj/Captura%20de%20pantalla%202015-03-24%2020.35.36.png?dl=0
[19:36] <sarnold> .. _nothing_ in dmesg? odd
[19:36] <genii> Waht?
[19:36] <hxm> i said sme: waht
[19:36] <genii> That is highly unusual
[19:36] <hxm> screenshot shows it
[19:36] <hxm> it is a virtualmachine of a dedicated server, maybe that
[19:36] <hxm> about bruteforce fail2ban could just ban them no?
[19:37] <genii> hxm: It's better to deny them for a period of time than to ban because some failed login attempts may be from valid ssh attempts
[19:37] <sarnold> if this is a container of some sort maybe dmesg is not available, but this is feeling like it may be a compromised machine with a rootkit installed
[19:37] <hxm> not visible in top?
[19:38] <sarnold> yeah
[19:39] <hxm> lsof could give a clue no?
[19:39] <sarnold> depends on the rootkit
[19:40] <teward> can someone look at https://bugs.launchpad.net/ubuntu/+source/php5/+bug/1336713 and see if they can test/replicate?  (I wasn't able to on a production instance that has `php5-fpm` installed but not the `php5` metapackage)
[19:41] <teward> (I ask here because server package :P)
[19:43] <hxm> what can i do now
[19:44] <sarnold> hxm: if it were me, I'd set ujpa replacement, restore from backups, etc., and do forensics on the system later
[19:44] <hxm> ujpa?
[19:45] <sarnold> up
[19:45] <sarnold> this is a new keyboard, I'm hot entirely used to it yet
[19:46] <genii> s/hot/not   ;)
[19:46] <sarnold> figures :)
[19:56] <teward> sarnold: we should replace your keyboard
[19:56] <teward> :P
[19:57] <hxm> do you know chkrootkit.org?
[19:57] <sarnold> teward: the transition is far harder than I expected
[19:57] <sarnold> hxm: I believe it's wildly unmaintained.
[19:57] <sarnold> hxm: .. and fighting a problem thats nearly impossible to solve anyway.
[19:57] <hxm> i know, but i wont get this tomorrow again
[19:58] <teward> sarnold: the alternative is radical implantation of cybernetic technologies which facilitate data entry via neural signals
[19:59]  * teward is a crazy, yes.
[19:59] <teward> or, go get your old keyboard :)
[19:59] <sarnold> teward: heh, I switched to this one because my old one caused significant pain in my hands and arms
[20:00] <sarnold> teward: or, well, I suspected it was involved. three weeks with a different keyboard have helped immensely
[20:03] <teward> rbasak: ping - have you seen anything else similar to https://bugs.launchpad.net/ubuntu/+source/php5/+bug/1336713 that affects the `php5` packages?  (I can only find that they have the `php5` metapackage on there which could potentially cause this if `apt` or some other system is not detecting `php5-fpm`, but...)
[20:21] <rbasak> teward: I've not seen it apart from in that bug IIRC. Need steps to reproduce to do anything about it :-/
[20:22] <teward> rbasak: yeah, I set it back to 'incomplete' - even tested a dist-upgrade on a production server I know for a fact hasn't been updated since January cause I"m insanely busy.
[20:22] <teward> rbasak: wasn't able to replicate with *just* php5-fpm (and there's been no updates to the nginx package that'd trigger this)
[20:22] <teward> rbasak: the only thing i can think that would do this is a case where `apt` derps and ignores php5-fpm being installed, and then installs Apache
[20:23] <teward> rbasak: since `php5` meta package
[20:23] <teward> (it was said in the description that it was present, I believe, but again, we need more info)
[20:23] <rbasak> teward: thanks. I updated the description to make it completely clear.
[20:25] <teward> rbasak: the only reason i'm on the bug is 'cause bdmurray pinged me (see -bugs for that discussion)
[20:26] <teward> rbasak: bdmurray said "Hey, they're using the PPA" maybe on the assumption that causes it (because nginx was mentioned on the bug), but I"m 99.99% certain that nginx, PPA or repository or otherwise, would not be the cause
[20:27] <Jeremy26> Is there detailed documentation for *installing* ubuntu-server on UEFI?  Every attempt I've made fails -- whether allowing the installer to try to auto-figure-it-out, or dropping to apt shell and manually installing the grub-efi packages.  Fails 100% of the time.  As a test of my h/w and partitioning, no issues installing Fedora ...
[20:31] <teward> rbasak: the *only* way it could happen is if `php5`'s dependencies list was in a "php5-fpm not present with regards to compatibility and depends" state - that's the only real way it'd trigger this, 'cause php5-fpm doesn't depend on apache components, and nor does nginx
[20:41] <mdeslaur> coreycb: is trusty getting nova 2014.1.4 too? doesn't look like it's been uploaded...
[20:52] <coreycb> mdeslaur, yes and it should more or less be done.  let me kick tests off and soon should be able to mark the bug as verified.
[20:58] <mdeslaur> coreycb: cool, thanks!
[22:01] <dmsimard> Hi. Any idea why installing the linux-generic-lts-utopic package in pkg_include (preseed) yields a kernel panic ? Is this a bug ? It looks like the initramfs doesn't run - the initrd for 3.16 isn't generated
[22:01] <dmsimard> Someone else's reported the issue and hacked together a update-initramfs in a late_command but that doesn't sound like an ideal solution
[22:01] <dmsimard> http://askubuntu.com/questions/526193/preseeding-3-16-hardware-enablement-kernel-requires-manual-update-initramfs
[22:02] <Jeremy26> One last try -- anyone successffully installed ubuntu-server to UEFI?  preferably Vivid?
[22:02] <dmsimard> We have precise installs that run fine when installing the trusty lts kernel in preseed, without having to resort manually running update-initramfs
[22:02] <teward> !vivid
[22:03] <teward> Jeremy26: ^
[22:03] <teward> you should probably start in #ubuntu+1 ( just sayin )
[22:03] <Jeremy26> teward: Been there on & off for 2 days .  The only response to any question I got was to come back here.
[22:04] <rbasak> Jeremy26: maybe try the ubuntu-server mailing list? UEFI is still not common enough that many people who happening to be watching IRC at the time you ask will have experience with it.
[22:04] <teward> Jeremy26: well then I suggest patience - but as with all QA and iso testing, you should consider filing an issue against the image if it's not working as it should be with efi installations
[22:04] <teward> i also agree that the ubuntu-server mailing list will yield better results
[22:05] <teward> (I force my systems to use legacy if efi is an option)
[22:05] <rbasak> dmsimard: that sounds worthy of a bug report to me.
[22:05] <dmsimard> rbasak: Thought so too.. Will file one. Thanks.
[22:05] <rbasak> dmsimard: maybe against the initramfs-tools package in the first instance?
[22:05] <rbasak> I'm not sure where else would be relevant.
[22:05] <dmsimard> Yeah I was about to ask where to file it
[22:06] <dmsimard> I haven't exactly pinpointed where the problem occurs
[22:06] <rbasak> I think a missing initramfs makes initramfs-tools a reasonable first guess. It can always be reassigned later.
[22:08] <sarnold> Jeremy26: dang, still no luck? :( please file a bug against ubiquity, I think that's the most logical next step
[22:10] <Jeremy26> sarnold: Hi.  Nope.  Terrible installler experience.  Grub's just non-functional.  UEFI's barely usable without manual intervention.  DHCP doesn't manage to configure the network at all -- all other installers do.  A USB-key install still insists on a CD being attached.
[22:10] <Jeremy26> Yeah, really terribly frustrated atm.  Looking at Fedora.
[22:11] <sarnold> Jeremy26: damn. I haven't needed a cd for installing for.. uhh.. many years. :/
[22:12] <Patrickdk> what is a cd?
[22:12] <Jeremy26> Patrickdk: it's what ubuntu-server apparently requires to install itself
[22:12] <Patrickdk> since when?
[22:13] <Patrickdk> I've been using pxe boot install since 2007
[22:13] <Jeremy26> Tried a Vivid Server install on UEFI from a USB key?
[22:14] <Patrickdk> can't claim I ever used usb install, and never used a computer with uefi boot
[22:33] <JanC> Patrickdk: if you used an x86 computer that was manufactured in the last several years, then it was most likely using (U)EFI
[22:34] <Patrickdk> not true, they still have the, do not use uefi option
[22:34] <JanC> they don't
[22:34] <Patrickdk> I don't own any of them
[22:34] <Patrickdk> all of my servers, even recent ones still have the don't use uefi
[22:35] <Patrickdk> I don't have any of the hp bl460c g9's yet though, just the g8's
[22:35] <JanC> they have a "run a BIOS emulation layer on top of UEFI" option
[22:42] <Patrickdk> no, no uefi support at all
[22:42] <Patrickdk> that is brand new, in the g9 servers that just came out
[22:45] <dmsimard> rbasak: FWIW: https://bugs.launchpad.net/initramfs-tools/+bug/1436098
[22:54] <Jeremy26> Patrickdk: I've supermicro mobos. All have full UEFI support.