/srv/irclogs.ubuntu.com/2013/07/12/#ubuntu-server.txt

=== kees_ is now known as kees
anepanaliptossarnold: hash -r00:10
sarnoldanepanaliptos: awesome, much better, thank you :D00:12
anepanaliptosstill the guide doesnt work, but that's a rtfm fix00:12
sarnoldwhat's wrong now?00:12
anepanaliptos(even though it did work before, not sure what changed..)00:12
anepanaliptosi did the make install00:12
anepanaliptosi run ffmpeg00:12
anepanaliptosand it tells me ffmpeg is not installed and i can pull it from the repos00:13
sarnoldanepanaliptos: is it in ~/bin or ~/ffmpeg_build/ somewhere?00:14
anepanaliptoshang on im recompiling (i did make -j9 so dont worry it wont be long)00:14
anepanaliptosok done00:14
sarnoldespecially fast if you changed nothing, hehe :)00:14
anepanaliptosyes it is.00:14
anepanaliptosthe exe is in the root dir00:15
anepanaliptoswell, "root" ~/ffmpeg_build/ffmpeg00:15
anepanaliptosroot@tesla:~/ffmpeg_sources/ffmpeg# ./ffmpeg00:15
anepanaliptoswas totally valid and ran ffmpeg00:15
anepanaliptosdoes anything have to do with the fact i am doing this as root and not sudo -s ?00:15
anepanaliptosi dont see why.00:16
anepanaliptosrebooting.. maybe....00:16
anepanaliptos(im doing that now..)00:16
sarnoldanepanaliptos: I thnk the make install step should copy it somewhere into the prefix's children (~/ffmpeg_build/)00:17
anepanaliptoswtf00:20
anepanaliptosit created the bin folder in my ~ dir00:20
anepanaliptosand in there are the binaries (or symlinks to em)00:20
anepanaliptossomewhere.. these instructions have gotten foobared00:20
sarnoldanepanaliptos: maybe that means make install uses the --bindir parameter?00:21
sarnoldanepanaliptos: you're right that sudo -s vs logging in as root, or doing the compile as a user and then sudo make install, could all fiddle with the $HOME variable..00:22
anepanaliptosno00:22
anepanaliptossomeone broke the document/howto00:22
anepanaliptosi went back using the wiki history thing00:23
anepanaliptosand im gonna try the instructioned that looked like the page from when it worked00:23
sarnoldah! :)00:23
anepanaliptossource ~/.profile00:23
anepanaliptosthat line is missing from the 'current' one00:24
anepanaliptosstil ldoesnt work00:24
anepanaliptosill rtfm and let yo uknow00:24
anepanaliptosi got this. :)00:24
i00nsunice00:43
i00nsuhello00:43
i00nsuhow to enable vhost's \ site-availables/*  ? is not working for me00:45
sarnoldi00nsu: did you try the a2ensite command as described here? https://help.ubuntu.com/lts/serverguide/httpd.html00:47
i00nsusarnold: yes. and service apache2 reload00:47
i00nsuthe DocumentRoot is at /home/user/Public_html/00:48
sarnoldi00nsu: did you get any error messages when you ran them? did you get errors in the logs?00:48
i00nsuand Directory too00:48
i00nsuis working but points to /var/www/ instead to /home/user/Public_html :o00:49
i00nsusarnold: no errors00:52
Nicolas_Leonidasafter a power outage my network stopped working, yesterday someone recommended I should remove some udev stuff00:53
Nicolas_Leonidasto make it work, what can I do to fix this problem?00:53
sarnoldNicolas_Leonidas: in my experience, loss of network connectivity after powerfail often means devices didn't come up in the correct order -- often solved by rebooting a modem or other upstream link, followed by rebooting a router, then perhaps a switch, and then the device that doesn't have network access...00:55
Nicolas_Leonidassarnold: that's a good point, but outage happened 3 days ago or so, and I've restarted everything many times00:56
sarnoldNicolas_Leonidas: oh :/00:57
Nicolas_Leonidasthe router works, all other computers have connectivity00:57
Nicolas_Leonidasdmseg says "eth0: link is not ready"00:57
sarnoldNicolas_Leonidas: plug/unplug the wire from the NIC and the switch?00:57
sarnoldtry other ports on the switch?00:57
Nicolas_Leonidassarnold: tried that too, when I connect the cable to another computer the other computer works00:58
sarnoldplug it into a different switch? or plug it into an auto-sensing NIC somewhere else?00:58
sarnoldNicolas_Leonidas: aha, so not that port on the switch then..00:58
Nicolas_Leonidastoday I bought a new NIC, same problem, now I'm almost convinced it's a software issue with ubuntu00:58
Nicolas_Leonidasjust found something interesting on syslog00:59
Nicolas_LeonidasDHCPDISCOVER on eth0 to 255.255.255.25501:00
Nicolas_Leonidasport 67 interval 1801:00
i00nsusarnold: I restart the server and is working.01:01
sarnoldNicolas_Leonidas: did you find any 'link becomes ready' messages for your NIC?01:01
sarnoldi00nsu: hunh. odd.01:01
i00nsu:)01:02
Nicolas_Leonidassarnold: in where?01:03
sarnoldNicolas_Leonidas: dmesg ..01:03
sarnoldNicolas_Leonidas: does 'ip link' show you your NIC?01:03
Nicolas_Leonidasno01:04
Nicolas_Leonidasshould I reset the router to factory settings?01:04
sarnoldNicolas_Leonidas: probably not. _maybe_ the router has some MAC firewall rules or something, but that seems a stretch01:05
sarnoldNicolas_Leonidas: are there errors in the router's logs?01:05
anepanaliptosgia sou re ellhna.01:05
Nicolas_LeonidasI just saw the dhcp log on router for the first time let me see if there is anyting in there01:06
Nicolas_Leonidasreset the router to factory settings01:16
Nicolas_Leonidasthat didn't help either01:16
Nicolas_Leonidas"booting the system without full network configuration"01:16
adam_groaksoax, jamespage:  once that openstack tempating stuff lands, this will help sync the templates /w https://code.launchpad.net/~gandelman-a/charm-helpers/sync_include_hints/+merge/174320 and any other non-py assets from charm-helpers01:32
adam_gcurrently it does not, and will delete those that exist in a local charm during sync01:33
=== dir is now known as guampa
=== acrocity_ is now known as acrocity
=== holymacaroons is now known as holymac
koolhead17Daviey: around?07:08
=== smb` is now known as smb
jamespageChocobo, havana-updates for cloud archive should contain those two missing packages in the next ~1hr or so09:07
yolandajamespage, zul , having this problem: http://paste.ubuntu.com/5867778/10:50
yolandaseems that isn't finding python-setuptools although it's added as build-depends?10:50
yolandajamespage, in case of python-cliff, do you think it makes sense to request a sync, not a merge? the only difference i see, apart from new things, is the pydist-overrides10:56
xnoxyolanda: can you pastebin full log?10:58
yolandaxnox, i found that the problem is dh $@ --with python2,sphinxdoc --buildsystem=python_distutils11:02
yolandaif i set the buildsystem to python_distutils it fails11:03
yolandaif i remove that it works11:03
yolandabut it's a merge from debian, and their rules file uses it11:03
xnoxyolanda: do you have a link to dsc? or a full build-log for me to look what's happening?11:05
yolandaxnox, let me reproduce again and show the log11:05
yolandahttp://paste.ubuntu.com/5867812/11:10
=== tjaalton_ is now known as tjaalton
yolandaxnox ^11:18
xnoxyolanda: you are building locally, not in sbuild / pbuilder? i thought a full log with how all deps are getting installed as well.11:20
yolandaxnox, it just fails on debuild11:20
yolandai can't arrive to sbuild step11:20
xnoxyolanda: can you push a branch with python-cliff-1.4-1ubuntu1 somewhere?11:20
yolandaok11:20
xnoxyolanda: or tar it up =) and scp to like people.canonical.com =)11:20
yolandaxnox: https://code.launchpad.net/~yolanda.robla/+junk/python_cliff_1.4-1ubuntu111:23
xnoxyolanda: cool, let me look.11:25
xnoxyolanda: so it succeeds for me fine here. Do you have python-setuptools installed locally?11:27
xnoxas it's needed to build the source package.11:27
xnoxyolanda: or you can use11:27
xnox$ debuild -S -sa -nc11:27
xnoxwhich will build the source package without running the "clean" step.11:27
xnoxdoes that work?11:27
yolandaInstalled: 0.6.37-1ubuntu111:28
yolandai'll remove and install again11:28
yolandai'm not sure if requesting a sync instead of a merge, the only significant delta is the pydist-override11:29
xnoxyolanda: you will not be able to sync the package as something is very odd:11:33
xnoxxnox@sochi:/tmp$ md5sum ubuntu/*orig* debian/*orig*11:33
xnoxca4b80452f73b58a01e18f8ebcfb4504  ubuntu/python-cliff_1.4.orig.tar.gz11:33
xnox3d5017c362d9b34eddf88aa43099411e  debian/python-cliff_1.4.orig.tar.gz11:33
xnoxxnox@sochi:/tmp$ du -a ubuntu/*orig* debian/*orig*11:33
xnox124ubuntu/python-cliff_1.4.orig.tar.gz11:33
xnox36debian/python-cliff_1.4.orig.tar.gz11:33
xnoxyolanda: debian original tarball is so much smaller and has different checksum, so launchpad will reject the sync. How come they are so different is a mystery though.11:33
yolandai'll send the debdiffs for the merge then11:34
xnoxso you could merge the packaging, but continue to use ubuntu's tarball for ubuntu uploads.11:34
xnox(or note that in sponsorship / debdiff request)11:34
xnox$ diff -r -U 4 debian/cliff-1.4/ ubuntu/cliff-1.4/11:35
xnoxOnly in ubuntu/cliff-1.4/: cliff.egg-info11:35
xnoxOnly in ubuntu/cliff-1.4/demoapp: build11:35
xnoxOnly in ubuntu/cliff-1.4/docs: build11:35
xnoxOnly in ubuntu/cliff-1.4/: PKG-INFO11:35
xnoxOnly in ubuntu/cliff-1.4/: setup.cfg11:35
xnoxso ubuntu tarball is not "clean" and has build artifacts in it11:36
yolandadebdiff is complaining for that also11:37
davegarathHi all I'm trying to export a nfs share with no_root_squash option but when I mount it on my client and write a file with root this file is squashed to nobody:nogroup.11:46
mardraumnfsv4?11:47
davegarathIs this a bug ?11:47
davegarathnfs411:47
mardraumtry 311:47
davegarathok]11:47
mardraumnfsv4 is a cruel mistress.11:47
davegarathbut In nfs4 no_root_squash doesn't work  ?11:48
mardraumI haven't tested, but see if it works for you with 3 first11:48
mardraum4 requires other stuff configured like user mapping11:48
davegarathty mardraum in vers3 it's all ok11:52
mardraumcool :D v4 changed a lot, so if you want to try implement it there, I think you'll need to do a bit of reading :D11:56
mardraumit's also a relatively immature implementation compared to v311:57
=== asac` is now known as asac
hxmi was configuring a samba server and shared some directories, but when i go to \\server.name it says i have no permissions and i should access to the shared directory directly12:15
hxmhow to configure samba for list the shared directories?12:15
=== raijin is now known as raininja
=== andreas__ is now known as ahasenack
=== cmagina-away is now known as cmagina
Chocobojamespage: thanks, I just used pinned the raring release and use those.  Not sure if that was a great idea or not...13:22
zulroaksoax:  i will need your reviewing skills today13:53
roaksoaxzul sure!13:54
zulroaksoax:  https://code.launchpad.net/~zulcss/python-swiftclient/mock-fbtfs/+merge/17441614:00
roaksoaxgive me 514:01
zulroaksoax:  noooooo14:03
raininjahow the heck do you get a version for the package in apt14:08
samba35i have 1 strange problem ,one of my ubuntu server on vmware (as a guest ) doesnt boot when there is unexpected problem (mostly power failure ) how do i fix it (normally it auto boot )14:53
patdk-wksamba35, look into GRUB_RECORDFAIL_TIMEOUT15:06
samba35patdk-lap, what i suppose to look for grub_recordfail_timeout15:14
samba35patdk-wk, sorry can you please tell me where i suppose to look15:16
samba35patdk-lap, sorry15:16
patdk-wkI just did15:16
=== dlloyd_ is now known as dlloyd
samba35 set timeout=${GRUB_RECORDFAIL_TIMEOUT:--1}15:19
samba35 what it suppose to be15:19
patdk-wkwhat?15:22
patdk-wkyou do know how to use google?15:23
samba35i supposw to change value ? or what15:23
patdk-wkhttp://askubuntu.com/questions/211797/how-do-i-make-ubuntu-power-loss-proof15:23
samba35ok15:23
=== dannf` is now known as dannf
=== Jikan is now known as Jikai
davegarathis there a way to force nfs-server to export nfs3 only ?16:43
davegarathI have no way to force my application to  mount nfs with nfsvers=3 and I have to use no_root_squash in order to write root files16:44
=== acrocity_ is now known as acrocity
=== Jikai is now known as Jikan
=== hatch_ is now known as hatch
=== whaley_ is now known as whaley__
=== whaley__ is now known as webron
=== matsubara is now known as matsubara-lunch
=== andreas__ is now known as ahasenack
kearneykidhey i am trying to install ubuntu server on a old hp server that ran windows 2000. whe i try to install normal i get a blank screen and when i use nomodeset and vga=normal i can see the error in the photo attached. also sometimes the cap and scroll lock leds flash on the keyboard.17:24
kearneykidhttp://img577.imageshack.us/img577/96/tl0i.jpg17:24
patdk-wkkearneykid, likely you need to use an older ubuntu, as the old hp server is probably way too old17:27
patdk-wktry 10.04 on it17:27
kearneykidwill do, thanks17:28
patdk-wkheh, older, newer ubuntu, that only supports newer cpu's17:28
patdk-wkwhere the definition of newer cpu, is old, but not that old17:29
sarnoldwe did (or soon will) lose support for 386 chips..17:29
patdk-wk386, that was done long ago17:29
patdk-wk486 was lost and 586 also17:29
sarnoldpatdk-wk: you might be thinking of the optimized compiles..17:29
patdk-wkmaybe, but I thought 12.04 needed p4 or better17:30
sarnoldegadget anyway says it was 3.8.. :) http://www.engadget.com/2012/12/15/linux-to-drop-i386-support-in-the-3-8-kernel/17:30
kearneykidby any chance do you know much about overclocking haswell?17:30
patdk-wksarnold that is linux though, I know ubuntu dropped it sooner17:32
patdk-wkya, ubuntu 10.10 drops pre i686 support17:33
sarnoldpatdk-wk: no kidding :) hehe17:33
sarnoldpatdk-wk: thanks :)17:33
patdk-wkand 12.04 dropped non-pae supported cpu's17:33
patdk-wkI used to know where that was offically stated, having a hell of a time locating it17:34
patdk-wkhttps://help.ubuntu.com/lts/installation-guide/i386/hardware-supported.html17:35
patdk-wk"However, Ubuntu precise will not run on i586 or earlier processors. Despite the architecture name "i386""17:35
patdk-wk"Support for i586 and lower processors, as well as for i686 processors without the cmov instruction, was dropped in Ubuntu 10.10. Most i686 and later processors are still supported"17:35
sarnoldpatdk-wk++17:37
=== webron is now known as whaley
=== matsubara-lunch is now known as matsubara
m0nk3yjoeAnyone know powerdns well?18:23
* RoyK doesn't18:24
* sarnold waves to m0nk3yjoe :)18:28
m0nk3yjoesarnold, Hello there!  You are all over these "Intertubes"18:31
m0nk3yjoe=)18:32
sarnoldm0nk3yjoe: I try :)18:32
mgriffin_can someone confirm for me that mysql server for precise does not have a dependency on libaio1 but quantal does?18:36
mgriffin_oh, precise filed this as won't fix.. https://bugs.launchpad.net/ubuntu/+source/mysql-5.5/+bug/99976618:47
uvirtbotLaunchpad bug 999766 in mysql-5.5 "MySQL 5.5 not compiled with native AIO" [Medium,Fix released]18:47
mgriffin_:(18:47
StathisAhello, i'm looking for a ubuntu server administration blog/webpage having tutorials,guides,examples etc...any suggestions?18:48
sarnoldStathisA: https://help.ubuntu.com/12.04/serverguide/18:49
mgriffin_that was a really silly decision, probably they didn't know innodb would fallback to not using libaio1 if it was simply not found18:51
StathisAsarnold: thanks18:52
m0nk3yjoeGuess not.. =(18:53
sarnoldm0nk3yjoe: it might be better to ask the more specific question anyway..18:53
m0nk3yjoeI'm trying to figure out the package pdns-backend-mysql18:53
sarnold"well" means different things to different people, but a concrete question can either be answered or not :) hehe18:53
smw_Hi all, I have a headless server and want to run a gui program which I don't need to access the gui for. What is the best way to do this?19:22
smw_I am not trying to control a gui program remotely (like vnc), I just want it to run even though there is no display19:23
axisyshow to check total RES usage by a process with multiple threads?19:27
sarnoldsmw_: investigate xvfb19:29
smw_sarnold, thanks19:29
smw_sarnold, works perfectly, thanks :-)19:40
sarnoldsmw_: woot :)19:41
smw_sarnold, I came in here and got a great answer. In #linux I got a lecture on not running X on a server :-\19:43
smw_I was also told to rewrite the program not to need X and one person ask me "why do you even think that is possible"19:44
sarnoldsmw_: sheesh :) I find if you don't know the answer, it's fair to hazard a clearly-labeled guess, but arguing is another approach entirely :)19:46
cwillu_at_worksmw_, in fairness, it _does_ imply that the program is at least mildly silly20:11
cwillu_at_workeven on windows, a service needs special checkboxes enabled in order to interact with the desktop20:11
cwillu_at_worksmw_, what's the program if you don't mind my asking?20:12
smw_cwillu_at_work, it is a webbrowser. We already have it automated with extensions/rpc, but we need it running full time on a server20:36
smw_cwillu_at_work, and an indication that I am doing something abnormal is not a reason to argue about whether or not servers should run browsers/x11 programs20:38
cwillu_at_worksure20:40
cwillu_at_workI'm just asking20:40
smw_cwillu_at_work, yeah, we have a lot of work already put into automating some tasks through the browser, but running it on a server seems better than a desktop :-)20:40
=== cmagina is now known as cmagina-away
Ricky27Hey. I run a server with a fairly good CPU and 32GB RAM (only ~6GB used on average). Any idea if I could run anything? Like a BOINC project that requires not all of CPU but more RAM? I'm paying for this machine anyway22:28
thesheff17BOINC is almost all CPU intensive...you can specify the number of CPU it uses and you can also specify nice 19 ./bonic so it has the lowest priority on those CPU when things are running.22:30
Ricky27thesheff17: thanks! Do you think this will interfere with a standard apache2 setup if i allocate 6 of 8 cpu cores?22:32
sarnoldRicky27: it shouldn't, see also ionice to ensure your apache has higher priority IO requests...22:33
thesheff17hehe yea what sarnold said22:33
Ricky27sarnold & thesheff17 thank you!22:33
=== klaas_ is now known as klaas
=== vidplace7_ is now known as vidplace7
nOStahlhey guys, my /boot is full heh. cant run sudo apt-get autoremove it complains about it being full to run the operation, how can I get rid of some old kernels manually the right way?23:19
sarnoldnOStahl: you can rm old kernel and initrds from /boot by hand until you've freed enough space to continue with the tools23:20
nOStahlah ok23:21
sarnoldnOStahl: just be sure to leave the kernel and initrd for the kernel version you're currently running, the newest, and make sure to leave at least two, in case one doesn't work well. :)23:21
nOStahlaye23:22
nOStahlty its going now23:22
sarnoldwoot23:22
nOStahlwhy would it have filled up the boot partition on its own? heh23:22
nOStahlshouldnt apt-upgrade be managing that on its own23:23
sarnoldonce upon a time, old kernels were never automatically removed... I think that there is some kind of automatic limiting in place now, but I've never looked into it deeply enough to understand. :/23:23
nOStahlhmm some thing is broken23:29
sarnoldnOStahl: I thought that went easily, there's sometimes more to fixing the problem..23:32
nOStahlits giving broken pipe when I try to sudo apt-get -f install23:35
nOStahldependancy issues23:35
sarnoldnOStahl: can you pastebin the errors?23:35
nOStahlthere's some flag set somewhere telling it there's no space left on the device23:37
nOStahlcant pastebin from this terminal access for some reason23:42
nOStahlbut its got broken dependancies23:43
nOStahlthe following packages have unmet dependancies: linux-headers-server : depends : linux-headers-3.2.0-48-generic but it is not going to be installed23:43
nOStahllinux-image-server : depends linux-image-3.2.0-48-generic but it is not going to be installed23:44
sarnoldnOStahl: the pastebinit package can paste to a long list of pastebins, it might be handy if you're trying to fight your way through an obnoxious firewall23:44
nOStahlyet if I try sudo apt-get -f install    it runs through like its working but then broken pipes23:44
sarnoldnOStahl: you can apt-get purge linux-headers-server linux-image-server  -- then fix all the other errors -- then re-install them, to make sure you get updates23:45
nOStahlok23:47
nOStahlwill try purge23:47
nOStahlnope no luck23:52
nOStahlits stuck on those packages...23:52

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!