/srv/irclogs.ubuntu.com/2012/08/10/#ubuntu-server.txt

MagicFab_Hi - is it possible to access the server filesystem from an LTSP client? I can't seem to find any docs about this.00:05
=== cpg is now known as cpg|away
=== mdeslaur_ is now known as mdeslaur
zuladam_g: pong02:00
=== cpg|away is now known as cpg
=== n0ts_off is now known as n0ts
=== cpg is now known as cpg|away
=== n0ts is now known as n0ts_off
=== n0ts_off is now known as n0ts
=== n0ts is now known as n0ts_off
=== smw_ is now known as smw
JonEdneyI'm having an issue with a server I just installed on a VPS - had all the services working without problem, and came back a few hours later to work on it, and I cant get services to start up.  Had to connect via serial web console.03:49
Gr3mlinhay all, would i be shoot if i said i was using ubuntu server for a mincraft server?03:49
JonEdneyI checked all the logs I know of, nothing indicates a problem, yet I cant get anything to start.03:50
mardraumJonEdney: can you even ping it?03:56
mardraumGr3mlin: no, it works fine03:56
JonEdneymardraum, nope03:56
mardraumthen I suspect you have a networking issue?03:56
JonEdneyOnly serial console access, even after reboot03:56
mardraumcheck your network config and talk to your VPS provider I guess03:56
JonEdneyThat was my next step, I'm browsing over logs again03:57
JonEdneyThanks mardraum03:57
Gr3mlinfine and great! :) only it wont allow me to use both thread of the CPU for the MC server. any pointers?03:58
JonEdneywth, you're right.  There is no eth0 refernce in my interface config file03:59
mardraumGr3mlin: ubuntu isn't stopping you, that sounds like a minecraft question, this probably isn't the best forum.04:01
Gr3mlin_sorry bout that guys. router issues now.. :S04:05
Gr3mlin_so  think i said, i have an issue with mincraft server not using all the CPU cores. any one know a trick?04:05
JonEdneyHmm, would failed networking config, cause apache, mysql, ssh, dovecot, etc to just not start?04:06
JonEdneyTried starting apache, and i got disconnecte04:07
mardraumJonEdney: most would start but depends how you have them configured04:10
mardraumeg if you specifically told apache to bind to an IP that wasn't there it would be unhappy04:10
JonEdneyThat would make sense.  This is something I've been working on as part of a learning process, I could have mistyped something.04:11
mardraumhow do you know they are not starting?04:11
mardraumtry eg service ssh status04:12
JonEdneyWell, I'm logged into the serial console04:12
JonEdneyThey all say not running, I did find some info when I tried to toggle networking though.04:12
JonEdneypostconf: warning: /etc/postfix/master.cf: unused parameter: smtpd_bind_address=127.0.0.104:13
JonEdneyNever seen that before, I must investigate.04:13
mardraumI think your /etc/network/interfaces is boned04:13
mardraumif you have lost localhost...04:13
JonEdneyI did see something weird in there, referencing something I've yet to hear of..04:13
=== n0ts_off is now known as n0ts
JonEdney"iface venet0 inet manual"04:14
JonEdneyhttp://pastebin.com/x5gyJ7iB04:14
JonEdneyThats my interfaces04:15
JonEdneyWhich is NOT what I set up, I wonder if another services changed it.04:15
mardraumread the top of the file?04:15
mardraumvenet0 - openvz04:15
mardraumbtw I can ping both those public IPs04:16
JonEdneyHmm.04:16
JonEdneyI've had this set up working before, but couldn't get the mail server working, so I reinstalled and tried again, got the mail working, and this happens, strange.04:17
JonEdneyI'll tinker some more.  Thanks for your input mardraum04:17
mardraumthat postfix warning won't stop postfix from running04:18
JonEdneyHmm, and everything is coming back up, I wonder if there was a problem at my host..04:18
=== n0ts is now known as n0ts_off
JonEdneyIll have to keep investigating, now it cant be pinged.04:23
JonEdneyThanks again for the input mardraum; take care.04:33
sarthorHi, I have a domain name, and I do not have static IP, but I have dyndns paid account, Can I host a website on my local machine? If yes, So what will be my DNS 1 and 2 to put in my registrar website against my domain. I am new in this world, my main goal is to learn and understand this stuff. the website is just for learning purpose. using ubuntu server 12.04 on my machine, Guide me please.04:48
mardraumsarthor: if you wish to use dyndns, you will need to ask them (or maybe check their faq?) nothing to do with ubuntu04:58
=== cpg|away is now known as cpg
=== n0ts_off is now known as n0ts
=== n0ts is now known as n0ts_off
=== tuv0k is now known as darthanubis
=== n0ts_off is now known as n0ts
trimetaSomething I've never entirely understood: if the changelog for linux-meta says that the only effect of the new kernel version is "Bump ABI", is there any reason whatsoever to reboot into the new kernel?06:24
=== n0ts is now known as n0ts_off
SpamapSadam_g: FYI, the reason bug 1021530 fell of the SRU team radar is because 1.4.0-1ubuntu1.2's changes file was generated incorrectly, it didn't list both updates, so bug 1021530 was not being tracked properly.06:28
uvirtbotLaunchpad bug 1021530 in openvswitch "[SRU] update to include stable fixes for OVS 1.4" [Medium,Fix committed] https://launchpad.net/bugs/102153006:28
SpamapSfell off rather06:28
ScottKtrimeta: It's not the linux-meta changelog that should interest you, but the changelog of the specific kernel you're installing.06:43
trimetaScottK: "aptitude changelog linux-image-3.2.0-29-generic" tells me there is no changelog.06:45
ScottKHmmm.  Not sure for kernels.06:45
ScottKI mostly read the security notices and worry about kernel reboots when I see something scary from there.06:45
trimetaLooks like nothing for 3.2.0.29, based on USN...oh well, I'll just stick with 3.2.0.27 until something happens.06:48
ScottKThere's a mailing list for the USNs.06:51
=== n0ts_off is now known as n0ts
=== n0ts is now known as n0ts_off
koolhead17hi all07:34
Anomie211http://pastebin.com/DuF0vn2T  -- Can anyone help with this error upon upgrading 9.04 -> 10.1007:34
trimetaAnomie211: Do you have some weird disk setup? It looks like xvda may have had problems mounting the root filesystem read-write, which led to a problem with udevd...although, I've never tried upgrading across different versions, so I have no idea what's normal here.07:38
Anomie211trimeta: I'm not sure? I'm with Linode and I cloned my original Linode so that I could upgrade it 9.04 -> 10.10 -> 12.04 and switch over  the IP (with 0 downtime). The clone straight away mounted in read-write. I fixed this buy eventually figuring out that 'Automount devtmpfs' had been set to 'yes' in my new Linode config (it wasn't in the original). Turned that off and it fixed the problem; updated to 10.10 and its giv07:40
Anomie211*by07:40
trimetaI don't even know what Linode is; I don't think I can provide much useful advice.07:41
Anomie211Just a VPS provider07:42
Anomie211I'll try #linode. Maybe it is problem on their end somehow07:43
Anomie211trimeta: fixed by turning 'Automount devtmpfs' back on. it was an issue with 9, which apparently was fixed in 10+07:48
trimetaAh.07:48
AdvoWorkanyone else seen this error with 12.04? error: error accessing /var/log/squid: No such file or directory from cron.daily?07:57
=== mcclurmc_away is now known as mcclurmc
_rubensounds like a flawed logrotate thingie08:05
Anomie21Processing was halted because there were too many errors.08:22
Anomie21Upgrade complete08:22
Anomie21The upgrade has completed but there were errors during the upgrade process.08:22
Anomie21Should I be worried?08:22
LachezarHello. I am having trouble increasing the Open Files limit for a user. Please advise!08:34
LachezarI am using 'sudo -u <user> <command>', and then I look at the /proc/?????/limits.08:36
LachezarWhen I do 'sudo -u <user> -i' I can see the limit has increased.08:37
LachezarBut when I do 'sudo -u <user> <command>' from an /etc/init.d script the limits stay default.08:37
=== mrevell_ is now known as mrevell
_rubenLachezar: try with 'sudo -Hu ...'?09:00
Lachezar_ruben: why would I need to change the HOME?09:06
_rubenLachezar: oh, right, it only sets $HOME. for me that pretty much works around all sudo related issues i've run into (like cvs and gpg), but they indeed store their settings in $HOME. so in your case, dunno09:09
_rubenmight need to use su in addition to sudo09:10
_ruben'su -' in particular09:10
Lachezar?!?09:11
_ruben'09:11
Lachezarsudo su - glassfish $GLASSFISH_HOME/bin/asadmin restart-domain?09:11
LachezarIs this what you're saying?09:12
_ruben'su -' does an actual login .. nah, but close, lemme see if i can figure the exact syntax09:12
Lachezar_ruben: I need this in an /etc/init.d/glassfish09:13
_rubenLachezar: in that case i'd just try "su - glassfish -c $GLASSFISH_HOME/bin/asadmin restart-domain"09:14
Lachezar_ruben: Do I need quoting or stuff? Because the restart-domain does not get sent...09:16
Lachezar_ruben: «su - glassfish -c ".../asadmin restart-domain"» does not fix the limits.09:18
* Lachezar is rebooting...09:18
* Lachezar can't believe his eyes: restarting helped...09:21
Anomie21Getting this error from my server10:12
Anomie21PHP Deprecated:  Comments starting with '#' are deprecated in /etc/php5/cli/conf.d/mcrypt.ini on line 1 in Unknown on line 010:12
Anomie21Cannot load the ionCube PHP Loader - it was built with configuration 2.2.0, whereas running engine is API220090626,NTS10:12
Anomie21I'm assuming its because I kept my existing mcrypt.ini when I updated - anyone know where I can find the current one?10:13
bhosmerAnomie21: You can replace the # with // I think and that warning will go away.10:16
bhosmerI'm sorry replace the # with ;10:17
Anomie21bhosmer: Just got rid of that line, was expecting it to be a big file but there was only 2 lines of code in it10:28
Anomie21and the one that was causing the issue was a comment10:28
bhosmerYeah, when you upgraded php your old ini file still had the old comment style from php 4. It really wouldn't have hurt anything if it was there, that warning was just telling you about the new comment style.10:29
=== zya-afk is now known as zyga
DoomGuyI have a weired problem.. I have configured a DNS  server for caching and doing Name Resolutions for some servers11:06
DoomGuythe problem is that I can access that servers by name from windows station without problem but can't do the same from linux stations11:07
DoomGuylike mine..11:07
DoomGuycan someone helps me ?11:07
Anomie21Getting this error too now : Any ideas? Cannot load the ionCube PHP Loader - it was built with configuration 2.2.0, whereas running engine is API220090626,NTS11:12
_rubenAnomie21: you already had that error, you'll likely need to rebuild ioncube or so11:18
chmacHow do I monitor directory usage over time? My root partition is filling up, and I'd like to know where the extra space is being used.11:18
chmacTo be more explicit, I'm looking for something automated, short of writing a bash script to store du output via cron.11:18
=== cpg is now known as cpg|away
=== mdeslaur_ is now known as mdeslaur
=== mcclurmc is now known as mcclurmc_away
imminentCucumbergood morning.  the colors in ls are really starting to drive me nuts.  I have tried to implement a bash alias for ls, but it doesn't seem to work.  I've created bash aliases before, so I'm not sure what's wrong.  Here's a screenshot:  http://imgbox.com/acrEWDQg  TIA...12:00
zulmorning12:10
koolhead11hallyn: https://bugs.launchpad.net/ubuntu/+source/libvirt/+bug/99797812:11
uvirtbotLaunchpad bug 997978 in qemu-kvm "KVM images lose connectivity with bridged network" [High,Confirmed]12:11
kodapahello, I'm investigating the same issue as koolhead11 just posted12:11
koolhead11is it closed?12:11
kodapaNo12:12
kodapaConfirmed12:12
* koolhead11 is off for weekend12:12
kodapaWe're running OpenStack Essex under Ubuntu 12.04 and having problems with VM networking dies after high net io. Using virtio network bridge12:12
kodapaSeems to only affect 11.10 guests and up12:13
kodapaifdown and ifup inside the VM makes networking work again temporarily12:13
kodapaThe bug is reported since 2012-05-11 and it's pretty critical12:15
kodapaIt seems like a upstream fix in the kernel exists: https://bugzilla.kernel.org/show_bug.cgi?id=4282912:20
uvirtbotbugzilla.kernel.org bug 42829 in kvm "KVM Guest with virtio network driver loses network connectivity" [Blocking,Resolved: code_fix]12:20
jacobwhallo, is there a printable reference for byobu keys?12:26
=== mcclurmc_away is now known as mcclurmc
=== zyga is now known as zyga-afk
pmatulisdid inetd always need to be installed or was it removed from the default install at one point?12:45
hallynkodapa: (koolhead11 has left the building) I've so far not reproduced that one (bug 997978).  I guess I'll try with nc from a different host and to two separate vms...  but i have a feeling there's another piece i'm missing12:56
uvirtbotLaunchpad bug 997978 in qemu-kvm "KVM images lose connectivity with bridged network" [High,Confirmed] https://launchpad.net/bugs/99797812:56
hallynthe kernel patch mentioned in that bug is already in precise, and not having vhost_net has not helped me reproduce12:57
hallynhm, now admittedly i had a 'while [ 1 ]; do echo -n 1; sleep 5; done' loop going to show me if net was still up.  i suppose that could have kept the network up.  i'm trying now with that as well as nc from another host to two separate vms13:06
=== zyga-afk is now known as zyga
kodapahallyn: I'm trying to reproduce the bug in our lab too13:31
kodapahallyn: trying with nc, ping and iperf atm13:31
kodapahallyn: seems like we need to spawn lots of connections, not just traffic13:31
kodapahallyn: another theory is bonding, our production environment uses bonding, our lab does not13:31
kodapahallyn: when did the patch get into precise?13:35
hallynkodapa: not sure offhand13:39
hallynyeah i'm not using bonding, i don't remember if everyone in that report was13:39
hallynlemme add that to the mix here13:40
=== dendro-afk is now known as dendrobates
=== Arrrghhh is now known as arrrghhh
arrrghhhhey all.  i have a VM of ubuntu server.  i want to SHRINK the disk.  is that possible, assuming the space OK?14:40
arrrghhhor if i shrink the disk, will it freak out when i power it back up?14:40
escottarrrghhh, you need to boot the vm with an iso image shrink and shrink the filesystem inside the vm image first14:51
arrrghhhescott, sorry i didn't follow that14:51
escottarrrghhh, the system installed in the vm has a filesystem tailored to the size of the virtual disk originally presented to that os. you have to shrink that filesystem first14:52
arrrghhhso boot a livecd, shrink the guest file system, then shrink it in esxi?14:52
escottarrrghhh, yes14:52
arrrghhhyea, that makes sense.14:52
arrrghhhcool14:52
=== dendrobates is now known as dendro-afk
hallynstgraber: looking at http://www.stgraber.org/2012/01/04/networking-in-ubuntu-12-04-lts/ - claims br0 shouldn't come up unti lbond0 is ready, but my precise box comes up with br0 not up, and when i dhclient br0 it immediately comes up15:16
=== dendro-afk is now known as dendrobates
uvirtbotNew bug: #1030316 in postfix (main) "evolution don't hadle IDN-Mail adresses" [Undecided,New] https://launchpad.net/bugs/103031616:00
jparnellis this thing active?16:42
arrrghhhi was going to say define alive16:45
arrrghhhbut i guess not.16:45
=== dendrobates is now known as dendro-afk
=== dendro-afk is now known as dendrobates
fire_it's dead :D16:54
=== mcclurmc is now known as mcclurmc_away
zuljdstrand: cinder should be ok now17:10
jdstrandzul: ok17:19
jdstrandI'll take another look17:19
zuljdstrand: can you do me a favor, can you do a review of cliff-tablib and tablib please? they are needed for python-quantumclient17:19
jdstrandzul: are there MIRs for them? I don't see them on the list17:26
zuljdstrand: not yet they are stuck in binary-new17:27
jdstrandzul: oh! sure17:33
zuljdstrand: thanks17:33
Psi-JackBlah, now when I just try to activate them, soon as they do their interval, they re-mark it inactive.17:48
=== cpg|away is now known as cpg
Psi-JackYeah, there is NOTHING in the agent log at all about it the #3 or #4 log item processors besides the fact, it IS actually watching them, initially, then it just isn't anymore.18:00
zasternIf I create an upstart script for a daemon I've built, and I want it to run at boot, is it enough to just create a symlink for it in /etc/init/?18:27
=== jhulten_ is now known as jhulten
arrrghhhzastern, why not use the built-in update script?18:56
zasternarrrghhh: I don't know about that - can you elaborate?18:56
arrrghhhupdate-rc.d18:56
zasternarrrghhh: what does that do exactly18:56
arrrghhhsorry, i was looking it up.  forgot the command myself18:56
zasterni dont want to run something that just makes any installed daemons run at aboot18:56
arrrghhhit automatically populates rc.X depending on when you want it to start.18:56
arrrghhhno no18:56
arrrghhhyou tell it what to update18:56
arrrghhhcheck the manpage ;)18:57
zasternarrrghhh: I don't want to create things in runlevels . . . isnt that init?18:57
zasternim trying to do things in an upstarty-way18:57
=== cpg is now known as cpg|away
arrrghhhthat is how you do it in an upstarty-way18:59
arrrghhhit's the "official" way to add daemons to startup19:00
arrrghhhyou can also remove daemons from startup using the same tool19:00
uvirtbotNew bug: #1035441 in krb5 (main) "package libk5crypto3 1.10+dfsg~beta1-2ubuntu0.3 failed to install/upgrade: libk5crypto3:amd64 1.10+dfsg~beta1-2ubuntu0.3 cannot be configured because libk5crypto3" [Undecided,New] https://launchpad.net/bugs/103544119:21
arrrghhhzastern, does that make sense?19:25
arrrghhhsorry i didn't explain real well, doing 5 things at once.19:26
arrrghhhwell i'll probably be on a bit later.  headed home now.19:26
=== arrrghhh is now known as Arrrghhh
pdtpatrickinteresting - i think postgresql's packaging needs some work. sysctl does not read the file it places in /etc/sysctl.d/30-postgres*19:43
=== Lcawte|Away is now known as Lcawte
=== koolhead17|zzZZ is now known as koolhead17
=== Arrrghhh is now known as arrrghhh
=== dendrobates is now known as dendro-afk
=== dendro-afk is now known as dendrobates
=== dendrobates is now known as dendro-afk
=== dendro-afk is now known as dendrobates
=== JanC_ is now known as JanC
=== dendrobates is now known as dendro-afk
kodapahallyn: any progress?21:57
=== koolhead17 is now known as koolhead17|zzZZ
hallynkodapa: i still can't reproduce22:07
hallynkodapa: given that you say you're using bonding and can only reproduce it in the production lab,22:07
hallynkodapa: can you post some info on your bond setup and the switches you're using?22:07
hallynmost of the odd network issues i've seen have been due to bad switches, badly configured switches, or use of bond modes which don't work well with the network hardware22:08
kodapahallyn: we don't have this issue with any other servers, using the same bonding setup and switches22:10
kodapaonly with openstack essex under 12.0422:10
hallynkodapa: <frown>  so maybe i need to install essex22:11
hallynkodapa: that same setup in your dev lab has not reproduced it yet though right?22:11
hallyni'm really thinking bonding is to blame, only bc that seems to usuallyb e more fragile22:11
kodapahallyn: no, but I've not tried generating many connections yet22:16
hallyni tried pinging a set of about 20 hosts in parallel, but you're probably talking something more22:17
kodapayeah22:17
=== cpg|away is now known as cpg
Vampyhey guys.. im forgetting the package that allows pasting to pastebin from cli, can soeone remind me what it is22:40
patdk-wkpastebinit22:41
Vampyapt-get install pastbinit?22:41
patdk-wkno22:41
Vampyis it default install in ubuntuserver22:41
Vampybasically i'm gettin a few error during an email server setup, and would rather pastebin then try and type it out..22:43
greppyVampy: apt-get install pastebinit22:45
Vampygreppy, thanks22:45
arrrghhhlol22:46
Vampynow to use that i would run the command ie apt-get install pastebinit | pastebinit correct?22:49
greppysomething like that, double check the man page :)22:50
Vampykk will do22:50
hallynstgraber: I'm thinking we should spit out a warning if lxc gets called with a tight umask22:58
Vampyok so first proble here is im unable to telnet into the localhost via port 25 heres the PB url http://paste.ubuntu.com/114016023:01
patdk-wkvampy, we know it doesn't work, that is what you keep saying23:05
patdk-wkthe question is, how did you break it? :)23:05
patdk-wkor what is your config look like23:05
Vampywhat config file would you like.. i'll pastte it23:05
patdk-wkI dunno23:06
patdk-wkwhat mta are you using?23:06
Vampypostfix23:06
patdk-wkit's your system, not mine :)23:06
patdk-wkpostconf -n | pastebinit23:06
arrrghhhhah23:07
Vampypaste.ubuntu.com/114016823:07
Vampyim also using Flurdy's mail server howto23:07
patdk-wkno wonder23:08
patdk-wkyou don't know how to use smtpd_*_restrictions23:08
Vampypatdk-wk, this is the howto i'm foloowig http://flurdy.com/docs/postfix/#test-common23:09
patdk-wkand you did the whole thing without testing it as you went?23:10
patdk-wkdoes postfix even start?23:11
patdk-wkanything in mail.log?23:11
Vampythought i had tested, but i've been working on it over bout 3 days23:11
Vampylemme paste mail.log23:11
Vampythat's in etc or var/log23:12
Vampymvmnd found it23:12
Vampypaste.ubuntu.com/114017723:13
patdk-wktypo's everywhere it looks like23:14
patdk-wkaccording to mail.log23:14
patdk-wkplus your database doesn't match your postfix config23:14
patdk-wkcan't locate your certificate23:14
patdk-wknote, read mail.log, fix errors, mail work :)23:15
patdk-wkgenerally23:15
Vampykk i see /etc/postfix/msyql as an area to look in.. i should invest that spot first?23:15
patdk-wkbut you need to fix those three, then after that, I'm sure there will be other errors, as it can make more progress23:15
Vampykk headin to that file to see the error23:15
patdk-wkI dunno, make it match however your database is setup23:15
patdk-wkor make your database match how it's set23:16
Vampyok when i log in to mysql thru that db, i have no issues loggin in or editing, why is it not allowing postfix?23:18
patdk-wkI didn't say it did23:19
patdk-wkI said they don't match, fix your schema23:19
patdk-wkread the error23:19
patdk-wkit's very self explanatory23:19
patdk-wkcolumn email doesn't exist23:19
patdk-wkit doesn't say anything about not being able to login23:20
Vampypatdk-wk, the refering column is from "where_field= " correct?23:24
patdk-wkdunno, likely23:24
patdk-wkthere is too many different ways you could make that file23:24
Vampyand that was exactly the area i needed to fix.23:25
Vampyand unused parameter isnt a bad error correct?23:25
patdk-wkit's generally not good23:26
Vampyok 2 pastes for ya paste.ubuntu.com/1140194 and 114019523:28
Vampythose are master.cf and main.cf23:28
patdk-wklike 138 is screwed23:29
patdk-wkneeds to be indented23:29
patdk-wkand should uncomment lines 16-2123:29
Vampywhich file23:30
patdk-wkmaster.cf23:30
Vampyline 13823:30
Vampyneeds to line up with  L139?23:31
patdk-wkwell, with the others around it23:31
Vampyok23:31
Vampy16-21 needs uncommenting?23:32
patdk-wkif you want email to work, when your not local to the server23:32
Vampyok done.. any other errors in master.cf?23:32
patdk-wknot that I notice23:32
Vampyok main.cf?23:33
patdk-wkreally should use a proper certificate23:33
patdk-wkyou really want to relay via gmail?23:34
Vampyis certificated.crt a default crt?23:34
patdk-wkno23:34
Vampyok, please explain23:34
patdk-wkthat could take me a few days23:35
patdk-wkdo you not know anything about servers? and you want to run an email server?23:35
Vampyi'm learning23:35
patdk-wkemail is the hardest thing to do, and it needs lots of maintance23:35
Vampywhich is why i picked it first23:36
patdk-wkya, wrong way23:36
patdk-wksee, your doing email23:36
patdk-wkbut in order for that email setup you used to work23:36
patdk-wkit depends on WEB, DATABASE, DNS, ...23:36
patdk-wkall things you are suppost to understand well before doing23:37
Vampyok, this is one i've choosen as i do not have the money to buy hosting, or a server, to run my business23:38
Vampyis there a better channel for new learning to server ?23:39
arrrghhhVampy, it honestly might be a good idea to pay for hosting23:47
arrrghhhit would be much cheaper in the long run when you consider all the other setup things required23:47

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