/srv/irclogs.ubuntu.com/2014/05/05/#ubuntu-server.txt

KawaiolaAnyone set up a static Ip in a virtual instance of ubuntu server?00:52
patdk-lapdefine, virtual instance01:05
=== justizin_ is now known as justizin
Kawaiolapatdk-lap, I'm running ubuntu server in vmware the web interface version01:15
patdk-lapthe web interface version?01:15
patdk-lapnever heard of any such thing01:15
Kawaiolapatdk-lap, Uhh hold on let me look at what it is actually called01:16
Kawaiolapatdk-lap, I guess I'm just using the regular vmware01:17
patdk-lapwell, set a static ip just like you would, on any other ubuntu install01:18
Kawaiolapaterfacesdk-lap, I did that but for some reason it keeps falling back to dchp which doesn't make sense because I totally took that out of my int01:18
Kawaiolapatdk-lap, I went and made it static in nano /etc/network/interfaces01:19
Kawaiolapatdk-lap, and everytime I restart the interface it fails and rolls back to dhcp01:20
=== Guegs_ is now known as Guegs
=== tyhicks` is now known as tyhicks
=== FreezingAlt is now known as FreezingCold
=== luminous_ is now known as luminous
=== Infierno|zZz is now known as InFierno|Coding
=== chmurifree is now known as chmuri
=== chmuri is now known as ChMuRi
=== ChMuRi is now known as chmurifree
=== chmurifree is now known as chmuri
bopper1_I set up a file server using samba, ubuntu 14.04 LTS, command interface only. I can't access my server from win 8.  Please Help :(06:07
bopper1_I set up a file server using samba, ubuntu 14.04 LTS, command interface only. I can't access my server from win 8.  Please Help :(06:28
bopper1_wow wow wow06:49
bopper1_can anyone help me?06:49
=== Lcawte|Away is now known as Lcawte
* [gnubie] waves08:20
[gnubie]# uname -r && lsb_release -d && sysctl -a | grep -i swappiness08:20
[gnubie]3.13.0-24-generic08:20
[gnubie]Description:Ubuntu 14.04 LTS08:20
[gnubie]vm.swappiness = 1008:20
[gnubie]given the above information, i'm wondering why my system is already swapping when i still have around 50% of physical memory that is still free..08:21
[gnubie]anyone?08:21
[gnubie]is there am i missing with this new version?08:21
[gnubie]hello..?08:40
ghostlineshi, i have a bridged nic called br0. I use it for my vm's that I run with kvm. However as soon as I enable a 6in4 tunnel that also uses the bridge(br0) I lose ipv4 connection on the vm's but the physical host is fine(ipv4 and ipv6 work)09:12
ghostlineshow can I troubleshoot this? It seems as if I can't use a 6in4 tunnel and bridge at the same time09:13
owhghostlines: I do not know the answer, but are there any hints in any logs or dmesg?09:16
ghostlinesowh double checking those now09:18
ghostlinesok looks like the vm's are getting an ipv6 address that I configured with radvd. Maybe the ipv6 config in radvd is incorrect09:28
ghostlinesbut I would still expect ipv4 to still work, it seems as if a misconfigured ipv6 address is being preferred / used by default if it's set and ignored the ipv4 address09:29
mardraumyes, ipv6 is preferred in dual-stack.09:32
ghostlinesahh09:33
linociscohi all09:49
linociscohow to ensure secure connection between windows clients and ubuntu server in client-server architecture?09:50
owhlinocisco: To do what?09:51
linociscoowh, to deploy asterisk like deployment in the network09:52
linociscoowh, sorry09:52
linociscoowh, to avoid hacking09:52
owhHuh?09:52
linociscoowh, make sure secure connection between clients and servers in office network09:53
owhWhat are the Windows Clients getting from the Ubuntu server?09:53
owhlinocisco: Are they getting files, print queues, database connection, terminal, x-windows, what?09:53
linociscoowh, basically, clients will access file folders/ authentication like LDAP,09:55
linociscoowh, database also in future09:55
owhlinocisco: So, we're talking file and print sharing?09:55
linociscoowh, terminal also for remote access to server to create/troubleshoot users accounts09:56
owhlinocisco: So, ssh to provide terminal access, user accounts to provide samba access, disallow guest access, remove services that you're not using.09:57
owhlinocisco: https://help.ubuntu.com/community/Samba/SambaServerGuide09:58
vedicWhat is the fastest but safest way (safe in terms of data security) to transfer a file stored on one Ubuntu server to send to another Ubuntu server?10:21
err-orvedic: scp?10:40
vedicerr-or: Can this be automated?  As soon as file arrives, it is tranferred to another server?10:46
ogra_use rsync via ssh ... and create an upstart job that watches the dir10:48
ogra_i.e. http://paste.ubuntu.com/7398001/ is an example job ...10:50
vedicogra_: thanks10:52
vedicerr-or: How about NFS and then local system send file to remote server when new file gets created? Upload will be handled automatically?10:55
vedicogra_:10:55
ogra_NFS is fine, but not secure indeed (no encryption)10:56
vedicNFS over vpn10:58
vedicogra_:^10:58
ogra_sure, that would work ...10:59
andologra_: *cough*Kerberos*cough*11:01
dassoukiany ideas on how i can fix the errors .. I'm trying to upgrade from 10.04 to 12.04 http://pastie.org/private/nnwgfqvpsojoweeiq50xta11:06
dassoukiany ideas?11:13
pmatulismorning11:35
_rubenbah, cant find any proper open-iscsi/lio related performance tuning guides .. for some reason w_await is ~0, while on the initiator it sometimes ramps up to 2000ms11:58
_rubenthe setup is far from ideal currently, so dont really know where to start digging :p11:58
lordievaderGood afternoon.12:06
patdk-wk_ruben use a faster network12:16
=== mc_bluebeard_ is now known as mc_bluebeard
_rubenpatdk-wk: i admit that this setup is using only a single 1gigE .. which i plan to improve "some day" .. still wondering what the actual bottleneck is though12:53
_rubendoesnt appear to be bandwidth, as it sometimes fills the gigE using writes12:54
_rubenwith low iowait12:54
patdk-wkso, taking your random infomation, you can conclude that?12:54
_rubenit's the read/write (random?) io that's suffereing from waits12:54
patdk-wkcause they have nothing at all to do with each other12:54
_rubenbandwidth and latency not being related sure, but bandwidth being the bottleneck is what i tried to rule out12:55
patdk-wkwhat does writes not having i/o wait have to do with any of this?12:55
patdk-wkofcourse writes won't have much12:56
_rubenactually, reads/writes should be irrelivant here .. as on the iscsi target i dont see any issues, disk wise .. so it's either network or the iscsi stacks12:58
_rubenio sched on target is deadline, on initiator it's noop .. guess it's getting about time to redo the network at that location, to make it at least a little bit sane12:59
onoruaGuys, I'm trying to have HDD->RAID1 -> LVM -> LUKS -> EXT413:16
onoruabut it doesn't boot, what kernel parameters should I set to decrypt root?13:16
onoruaI've tried a lot from different sources in internet but not luck13:16
onoruadoes anyone has working configuration like this?13:16
onoruain my case it stop with mdadm create group disk not found13:18
pmatulisyeah, bug #1316124 , was inevitable.  i wonder if -backports is the way to go13:32
uvirtbotLaunchpad bug 1316124 in openldap "Actual Ubuntu Slapd totaly useless for "serious" use" [Undecided,New] https://launchpad.net/bugs/131612413:32
=== InFierno|Coding is now known as InFierno|AFK
mzarellaI'd like to create the symbolic link, rm /usr/sbin/node, and then update my path. How would i update my path without logging out of the ssh session and logging back in?14:58
PrincesaHi people!14:58
Princesacan you help me with zoneminder in ubuntu 12=???14:59
Princesahelp me with ubuntu server and zoneminder=??15:03
pmatuliswhat is zoneminder?15:33
=== dannf` is now known as dannf
zottaUsing apache server: It16:43
zottais possible to allow the server to read a directory without granting access to every user on the server?16:44
zottaI added www-data to the group wich owns that folder. but I still get 403 error16:45
sarnoldzotta: did you restart your apache server to pick up the group change?16:45
zottanope16:45
zottanow it works: Thanks sarnold!16:47
zottaBut why do I need to restart a program to change user groups?16:48
sarnoldzotta: it isn't so much 'restart a program' as much as 'log that user in again'16:50
sarnoldzotta: .. and services are funny, restarting them is the moral equivalent of logging them in again16:50
zottaok...16:51
sarnoldsorry it's hard to describe :) right now I'm even curious where in the startup procedure for apache it would know about the supplementary groups. it's funny that I've never thought of it before.16:53
Ri881Hi, I want to isolate just one process. What's the best way? Do I always have to emoulate the whole OS?17:16
Ri881...emulate17:16
sarnoldRi881: depending upon your goals in isolating that one process, you have a handful of options; since I know apparmor best, I'd probably aim for an apparmor profile first, but you could also use lxc to provide it a 'clean machine' ..17:18
plmhi all17:57
plmare there a way to do a link of "arm-linux-gnueabi-gcc -V" to "arm-linux-gnueabi-gcc -v" ?17:58
=== Lcawte is now known as Lcawte|Away
lordievaderplm: An alias?18:05
plmlordievader: yes18:06
plmlordievader: any idea?18:11
lordievaderAdd "alias arm-linux-gnueabi-gcc -V='arm-linux-gnueabi-gcc -v'" to your bash profile or somewhere else where it gets loaded.18:11
lordievaderPersonally I put aliases in ~/.alias and then source it from my zsh config.18:12
plmlordievader: $ alias arm-linux-gnueabi-gcc -V='arm-linux-gnueabi-gcc -v'18:14
plmbash: alias: arm-linux-gnueabi-gcc: not found18:14
plm$ arm-linux-gnueabi-gcc18:14
plmarm-linux-gnueabi-gcc: fatal error: no input files18:14
plmthat is strange..18:14
lordievaderHold, on it needs some more quotes.18:15
=== PeterSchwaller_ is now known as PeterSchwaller
lordievaderHmm I spoke to soon, the alias itself cannot have arguments it seems, but you could name it arm-version or something.18:18
plmlordievader: how?18:18
lordievaderalias arm-version='arm-linux-gnueabi-gcc -v'18:19
plmlordievader: ohh problem that i need that, becouse I have a bug in autotools I thing, look the error compiling python:  configure:3509: arm-linux-gnueabi-gcc -V >&518:21
plmarm-linux-gnueabi-gcc: error: unrecognized command line option '-V'18:21
plmlordievader: So I was thinking to change that -V to -v18:22
lordievaderSeems to me like that is a problem with autotools. x86_64-linux-gnu-gcc doesn't know -V either.18:25
lordievader(I might be very wrong at this ;) )18:26
plmlordievader: yes, becouse that problem in autotools I'm trying use other way to cheat it?18:27
plms ?/.18:27
plmlordievader: or how I to change that in autotools, do you know?18:30
lordievaderI don't know autotools. Is this for the RPi btw?18:30
lordievaderIf so, I have cross compiled a kernel for it with help from http://elinux.org/RPi_Kernel_Compilation#Ubuntu18:31
plmlordievader: not, a arm9 embedded linux18:31
plmlordievader: is not way with power of linux to do a kind of that alias/link? =D18:31
=== RaptorJesus_ is now known as RaptorJesus
plmlordievader: I will to a script with thaat name and inside put the -v =D18:37
plmand original I change the name18:37
plmnot works :-(18:45
=== InFierno|AFK is now known as Infierno
=== Guegs__ is now known as Guegs
jsmitsnlafter upgrading to 14.04 from 12.04 my amavisd-new is not running in normal (background) mode. only in foreground or debug mode. Any pointers?20:16
=== SPQN is now known as AndreiCurelaru
=== Lcawte|Away is now known as Lcawte
=== Lcawte is now known as Lcawte|Away
=== geowany_ is now known as Geowany
=== FreezingAlt is now known as FreezingCold
hehehey folks23:11
hehehow I can see which service is using secondary ip?23:11
sarnoldhehe: netstat -anp or ss23:12
hehety23:14
hehesarnold: localhost on box with many ips usually uses just 1?23:31
heheit seems here it uses 2 :(23:32
sarnoldhehe: all of 127.0.0.0/8 routes to localhost23:33
sarnoldhehe: normally 'localhost' resolves to 127.0.0.1 and the machine's hostname resolves to 127.0.1.1.23:33
heheok i got box with many ips I want 1 ip to be used for website and 1 for another service23:33
heheI added 1 more ip to eth0:0 but someone nginx starts using in23:34
heheit23:34
hehesince nginx using localhost I presume localhost somehow grabs ot?23:34
hehevery bizzare23:34
sarnoldhehe: what does your nginx  'listen' line look like?23:34
hehelisten 80 default_server;23:35
heheI tried to change it to ip and it rejects it23:35
sarnoldhehe: something like: "listen   192.168.1.14:80;" should work fine23:35
sarnold(I don't know what "default_server" does, I haven't seen that before)23:35
hehenginx: [emerg] bind() to 0.0.0.0:80 failed (98: Address already in use)23:39
heheit seems for some reason both nginx and service nr 2 cant yet to sit nicely23:39
heheon their ios23:39
heheips23:39
heheok just have to find where 00000 isd23:40
heheis23:40
sarnoldhehe: 0.0.0.0 is a wildcard bind; it'll fail if something else has a specific IP bound on :80 ..23:40
heheok23:40
heheI was thinking since I got multiple ips23:40
DenBeirenu-oh23:40
DenBeirenthink i did something terribly wrong23:40
hehei can run many services on 8023:40
DenBeirenpurged all of the kernels and rebooted23:40
sarnoldhehe: one per IP :)23:40
hehelol I got 523:41
hehei told dumb  nginx use separate it23:41
heheIP23:41
heheas per u howto :d23:41
hehewhat da it wants :)23:41
DenBeirennow my server won't come up,.. any ideas on where to check or what to do first?23:42
hehebackup23:42
heherestore from it23:42
hehewhy da u purged kernels?23:42
DenBeirenno backups as far as i know23:42
sarnoldDenBeiren: probably easiest is to boot a rescue environment off a CD or something, mount your old system somewhere, chroot there, use apt-get to install a kernel, sync, umount, reboot, and hope it works :)23:42
DenBeirenik had 99% of /boot filling up23:43
DenBeirenso i purged a few of the (very) old kernels23:43
hehewell then gix it23:43
hehefix :D23:43
heheif u smart enough to purge kernels u can fix it23:43
hehe:P23:43
DenBeirenaccidently i pasted a (incomplete) command and purged them all :s23:43
DenBeirenhehe: you really give me too much credit :p23:44
heheu aimed to do u done23:44
sarnoldeveryone's got to kill their server at least once :)23:44
hehe:D23:44
hehenow u fcked it23:44
hehehahaha23:44
DenBeireni've killed it multiple times :-)23:45
hehethats why I like aws amis23:45
heheDen!!!!!!!!!!!!!!23:45
hehemultiple? :D23:45
hehehehe23:45
DenBeirenbut i've been killing it since 6.04 :-)23:45
heheme 0 times23:45
hehesarnold: started nginx and socket.error: [Errno 98] Address already in use23:45
DenBeiren"learing" from each kill :-)23:45
sarnoldhehe: you'll have to configure all your services to ensure that no two want to use the same IP:port23:46
hehei got nxinz and 1 more23:47
heheso I got to tell  it to nginx somehow23:47
DenBeirensarnold: mounting, syncing, etc is not something i am feeling cozy with,..23:47
hehethen service 2 uses whats available23:47
heheDenBeiren: then install windows? :P23:47
DenBeirenmaybe faster/better/cleaner to do a reinstall?23:47
heheits like car parts23:48
hehesimple23:48
sarnoldDenBeiren: re-install is drastic overkill for this problem but if it what you feel comfortable doing, don't let me stop you23:48
heheput new engine in - start - go23:48
DenBeirenhehe: pls don't go there,.. my car is in the garage :'(23:49
heheso fix it23:49
DenBeirenand no, i didn't do anything to it :-)23:49
hehebe  a man :)23:49
sarnoldwell you didn't buy a new car, right? :)23:49
hehe LAST_ACK   whats that?23:50
hehe1 pesky last ask on ip I want23:50
hehe:D23:50
heheok means flushed soon23:51
heheand then ip is free!23:51
sarnoldhehe: oh man, that's reaching back too many years, but iirc it's when you've received a FIN but haven't yet sent your own FIN ACK combination...23:51
hehehow I kill it on my side?23:52
hehe:d23:52
sarnoldah, looks like FINs have been sent on both sides and you're just waiting for the ACK from the remote peer: http://tangentsoft.net/wskfaq/articles/bitmaps/state-diagram-small.gif23:52
hehetrash nginx took second ip again23:53
hehe:D23:53

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