[04:58] <OssumPawesome> hello I'm trying to remotely forward a port using ssh on my ubuntu web server. So far I can get 192.168 loaded up through elinks/links2, but when I fill out the port forwarding form nothing changes and my new forwarded port is not added to the list. could anyone help me with this?
[05:27] <sarnold> OssumPawesome: are you using ssh's actual port forwarding mechanisms (-L and -R) or are you just happening to use elinks over an ssh connection to try to configure port forwarding on some -other- device like a home router?
[05:37] <OssumPawesome> the second one sarnold
[05:38] <OssumPawesome> my router
[05:39] <sarnold> OssumPawesome: many router interfaces require javascript in order for 'submit' buttons to work -- or, they even lack the submit buttons entirely on the assumption that the javascript will Just Work
[06:14] <OssumPawesome> sarnold, if youre still there - is there a way i could get javascript to work on command line or somehow open a port on my router a different way?
[06:17] <sarnold> OssumPawesome: try ssh -L 8888:192.168.1.1:80 servername --- then aim your firefox or chrome to localhost:8888
[06:17] <sarnold> OssumPawesome: of course if you have to connect to to it with ssl, it'll take a bit more work, probably you'll have to run 'ssh' as root, and also add -L 443:192.168.1.1:443 so that https://localhost/  will load the remote router page..
[06:19] <OssumPawesome> yeah im doing this remotely so everythings gotta go through ssh
[06:19] <OssumPawesome> what is ssl?
[06:22] <sarnold> OssumPawesome: SSL is secure-sockets-layer, the original name for the new TLS, transport layer security -- it's the little lock icon in the web browsers, the "s" in "https"
[06:35] <OssumPawesome> sarnold, it works! thank you so much you are a beautiful human being. I honestly did not expect to be able to do this remotely. Thanks again
[06:36] <sarnold> OssumPawesome: hehehe :) glad it worked for you :)
[06:36] <sarnold> OssumPawesome: ssh -L and -R are awesome powerful things. they can get you out of trouble or create all kinds of trouble :) have fun!
[06:37] <maxb> ssh -D is even more awesome :-)
[06:39] <sarnold> maxb: wow, I haven't seen that one before. looks awesome. ssh -D blort, then go set your socks proxy in firefox, and OMG EVERYTHING WORKS ? :)
[06:40] <maxb> Pretty much
[06:40] <sarnold> maxb: cool :) thanks
[06:41] <maxb> And then you combine it with tsocks, and you can run pretty much anything over it
[06:41] <sarnold> oooh.
[06:41] <sarnold> very cool. :)
[06:42] <maxb> Very handy for "I want to run this as if I was within my datacentre"
[06:43] <maxb> or reconfiguring the ADSL router in an office on the other side of the globe
[07:59] <pds_> hi ladies and gentlemen i'm trying to kickstart a ubuntu server 14.04 LTS with a ubuntu desktop 12.04LTS, following this tutorial http://digitalsanctum.com/2013/03/22/how-to-setup-a-pxe-server-on-ubuntu/ using this kickstart file i create with the gui (http://paste.ubuntu.com/7773935/). It boots up just fine but it doesn't seem to want to automate.
[08:07] <stemid> a kickstart file is not enough pds_, you also need to modify the boot parameters in the pxe files.
[08:07] <stemid> to point to the ks file
[08:07] <stemid> also any incorrect command in the ks file will halt the automated install and ask for input.
[08:08] <stemid> and I've only pxe booted debian, never ubuntu, but debian does not have full KS support. I still have to use preseed.
[08:09] <pds_> stemid what precisely do i need to modify?
[08:10] <stemid> well you should find a complete tutorial that also covers the PXE files. I can only speak from experience with Debian and RHEL. I tend to modify pxelinux.cfg/debian-installer/amd64/pxelinux.cfg
[08:11] <root-user> Hey guys, question here maybe someone can help. Yesterday i did "sudo chmod 600 //"  and now my system wont boot and i cant login trough console. Anybody knows a solution maybe?
[08:13] <stemid> you could try booting a live CD, mounting your root filesystem and doing chmod 0755 on it
[08:14] <root-user> Thanks, read about it but not tried it yet. btw, doest the chmod 600 // means that i have changed file permissions on the entire system :o?
[08:15] <stemid> no, you did not use -R
[08:15] <stemid> thankfully
[08:15] <stemid> so hopefully you only need to reset / to 0755 and then boot up.
[08:15] <root-user> Ah, that's a relief! Let me try it :)
[08:47] <pds_> stemid i'm been messing around with kickstart files for to damn long
[08:47] <pds_> trying preseeding now
[08:47] <pds_> http://sfxpt.wordpress.com/2013/06/09/get-the-debianubuntu-ready-and-customized-the-way-you-like-in-10-minutes/
[09:41] <root-user> stemid, u still here? I booted from live disc, and mounted my file system. Got a terminal but when i try to chmod 0755 it says "unable to open /etc/sudoers: permission denied"  Could you tell from this what I need to do?
[09:55] <pds_> is there any documentation any where on the interwebs on how to automate the installation of ubuntu server
[10:07] <root-user> @Stemid,  thanks! your guided me in the right direction, fixed now with the liveCD :)!
[12:23] <pds_> kickstarting ubuntu server getting the installer failed to download a file from the mirror.
[12:25] <hron85> Hi! Can anyone help me with joining a 14.04 server into the AD? I successfully joined the server, wbinfo -u and wbinfo -g works correctly. getent passwd $USERNAME works correctly, however the getent passwd does not lists any windows user.
[12:25] <hron85> I do not know it is related or not, but sshd says initgroups: invalid argument when i try to log in into the server with my windows account
[12:26] <peetaur2> pds_: I found kickstart (an alien system) to suck terribly in Ubuntu, so I use preseed only (which is native to Debian). I doubt it's related though.
[12:27] <peetaur2> kickstart comes from redhat
[12:28] <peetaur2> pds_: do you use a caching server? I recommend apt-cacher-ng
[12:29] <pds_> nope just a simple nginx server
[12:32] <peetaur2> is it a mirror? or unrelated to packages? I'm talking about the packaging proxy
[12:33] <hron85> anyone with samba knowledge?
[12:33] <pds_> but i'm open for suggestions reading the help.ubuntu about preseeding
[12:34] <pds_> extracted the ubuntu iso the my nginx www/ubuntu
[12:34] <pds_> join #ubuntu-nl
[12:35] <peetaur2> I'm not so sure extracting the iso is the best way to do it
[12:35] <pds_> just following http://digitalsanctum.com/2013/03/22/how-to-setup-a-pxe-server-on-ubuntu/
[12:35] <pds_> brb
[12:36] <peetaur2> pds_: what I used (maybe) was this http://www.linuxquestions.org/questions/linux-newbie-8/automated-remote-installs-of-ubuntu-using-kickstart-802660/ and then modified it heavily... so in partcular look at the "Step 3. Copy the netboot files into the tftp server."
[12:36] <peetaur2> I am not sure really where all my notes came from ;) but that link was in my notes
[12:36] <pds_> yup did that
[12:37] <peetaur2> so if you downloaded the netboot stuff, then you don't need any iso ...
[12:37] <peetaur2> and you shouldn't need a web server except for kickstart
[12:38] <peetaur2> and I couldn't get various methods to work for supplying the preseed.cfg... so I put it in the initrd cpio.gz
[12:39] <peetaur2> but you might find a way to put it on the web (but it is kinda wrong since that means you can't configure network in pressed then)
[12:39] <AtuM> pds_, http://linux.opm.si/programska-oprema/ubuntu-14-04-network-install
[12:41] <AtuM> pds_, these are my notes written after I got this to work.. kickstart is almost unusable in 14.04.. use preseed and use my notes to get ahead
[12:41] <pds_> hmmmz
[12:42] <peetaur2> AtuM: heh oh nice. Good thing I never even tried kickstart in 14.04 because I already found it bad in 12.04 or whatever I started with. (where it worked for basics, but didn't do network, disk, keyboard, language, :D [anyuthing but proxy I think] properly)
[12:43] <peetaur2> but it had a purdy GUI ;)
[12:43] <AtuM> peetaur2, well.. GUIs are sometimes evil ;-)
[12:43] <pds_> kk
[12:43] <pds_> so back to the drawing board
[12:44] <pds_> read the fricking preseed files
[12:45] <peetaur2> and FYI I found partitioning horrid in preseed... they reinvented the wheel. All they would need to give you is some parted, mdadm, lvcreate, etc. commands, but they made a thing that is limited in some ways (have to use whole disk, can't have a mix of raid and not, can't keep old partitions, etc.)
[12:46] <pds_> so you guys would go for preseeding instead of kickstart
[12:46] <peetaur2> on a debian based system, yes for sure
[12:46] <AtuM> peetaur2, I do manual partitioning.. I only use network install so I don't need to add a dvd unit to servers
[12:47] <peetaur2> you can also just leave the partitioning blank and it'll prompt during install
[12:47] <AtuM> pds_, you don't have a choice.. preseed is debian based.. kickstart does not work for debian
[12:47] <pds_> kickstart does kinda work :)
[12:47] <pds_> but i guess i'm trying to hard :)
[12:48] <peetaur2> on the last mass install I did just that... had a pre-script that downloaded a chroot with the tools I needed, unpacked it, and then I hit ctrl+f2 and did my partitioning in there :D  (for 73 servers... would have been nice if preseed partitioning didn't suck)
[12:48] <peetaur2> my chroot had a script of course
[12:48] <peetaur2> but still manually connecting to each kvm over IP and hitting a few buttons is stupid
[12:49] <AtuM> pds_, when you get kickstart to work properly, please give me a shout and a link to your manual :)
[12:50] <pds_> let me note some stuff down
[12:50] <pds_> ask peetaur2 and AtuM stuff about preeseeding :)
[12:53] <pds_> btw i actually got it allmost working
[12:53] <pds_> http://www.tiikoni.com/tis/view/?id=bb55c79
[12:54] <AtuM> oh.. I see it's almost installed :D
[12:55] <peetaur2> I don't know what it's doing there, but what (I imagine) mine does is:    install base system from CD (like your download), and then since it's a proper netboot, installs non-base stuff (ssh server for example) from the network, using the caching server. So if you just have one repo set that is your server, but it has no extra stuff not on the cd, then it'll fail there.
[12:56] <peetaur2> so I think you need a proper netboot with a simple caching server. (apt-cacher-ng is super easy to set up... basically just install, and it works already, no repos to set up like the stupid aptproxy which was horribly designed)
[12:58] <AtuM> peetaur2, you use kickstart for this?
[12:59] <AtuM> peetaur2, I don't think pds_ got as far as installing the os. I don't see how this can be done with kickstart. I should check how fedora does it, as it basically also writes an image and then expands it to the partition.
[13:00] <pds_> well i allmost succeeded no ?
[13:01] <AtuM> pds_ depends on what file it's trying to download..
[13:02] <pds_> ubuntu local mirror
[13:02] <peetaur2> AtuM: no, I tried kickstart, and got it to work with a mix of preseed for things not supported, but later found it had side effects, so moved the kickstart parts to pure preseed, and now it works fine.
[13:02] <AtuM> pds_, is this post-install or is it searching for the installer image?
[13:03] <AtuM> pds_, installer no longer uses repo to install base os.. it copies an image and expands/extracts it..
[13:04] <AtuM> it can later use repo to install extra packages if needed
[13:05] <pds_> probably searching for the installer
[13:06] <AtuM> peetaur2, I'va also moved away from kickstart completely.. too bad preseed is not as good for partitioning. lucky for me I don't need that yet
[13:07] <peetaur2> kickstart is worse for partitioning btw ;) (at least on Ubuntu)
[13:07] <peetaur2> just for someone like me that knows the CLI well, it's far easier to use plain commands like parted, mdadm, pvcreate, lvcreate, etc. to partition a multi-disk system in loops and stuff
[13:07] <AtuM> pds_, ok, so you're still far off.. I'd recommend you to stop wasting time with kickstart and forget the nice gui it has
[13:07] <pds_> well that was pretty clear
[13:08] <pds_> well about a month of research down the drain
[13:08] <pds_> woop
[13:08] <peetaur2> haha
[13:08] <peetaur2> a month wasted
[13:08] <peetaur2> it would take you like 3 hours to get a simple preseed file going
[13:08] <peetaur2> sorry to tell you only now :(
[13:08] <AtuM> pds_, you might want to use the knowledge on redhat systems.. not such a waste in my eyes
[13:08] <peetaur2> hehe yes you can look on it positively that way ;)
[13:09] <pds_> well since kickstart worked on ubuntu (well the gui that was)  and most of my knowledge is on centos system ...
[13:09] <AtuM> what I've found is that there is no "ultimate" distro/os.. so every knowledge gained is helpful sooner od later
[13:32] <hron85> anyone with samba knowledge?
[13:32] <jrwren> `ask
[13:33] <hron85> jrwren: i already asked, but nobody responded....
[13:33] <jrwren> oh, the getent passwd you expect to list?
[13:33] <hron85> yeah, i get rid on getent passwd problem
[13:34] <hron85> jrwren: but the latter is still an issue
[13:34] <hron85> sshd says initgroups: invalid argument when i try to log in into the server with my windows account
[13:34] <jrwren> my knowledge in that area is 10+ yrs old. sorry. I only vaguely recall if listing all users ever worked with winbind
[13:34] <hron85> it work
[13:34] <hron85> s
[13:35] <hron85> just have to winbind enum users = yes and same with groups
[13:38] <jrwren> is nscd running?
[13:46] <pds_> looking into preseeding found some examples , wondering if there is a list with possible key values
[13:47] <peetaur2> look in debian docs... should be there
[13:47] <peetaur2> there's one that has all options but commented out
[13:48] <pds_> this one ? http://www.debian.org/releases/wheezy/example-preseed.txt
[13:49] <pds_> https://help.ubuntu.com/12.04/installation-guide/example-preseed.txt
[13:49] <axisys> how do I look a for a process that matches a name and if its cpu usage is higher than 15% then restart? I could use ps and find the info and restart.. but it might not be safe automate that way..
[13:50] <rberg-> I -think- debconf-get-selections will show a list of possible preseed options
[13:50] <axisys> manually is how we are doing it today..
[13:51] <peetaur2> pds_: yes probably
[13:51] <peetaur2> debconf-get-selections lists options, but they don't necessarily apply to the install process
[13:51] <peetaur2> a great command to know though
[13:52] <peetaur2> (preseed is not just for installing the OS)
[13:52] <pds_> well it gives a lot of options regarding key but not the values
[13:52] <peetaur2> axisys: why do you want to do that? some commands need more CPU ... (flash player for example, needs 100% on all cores :D... er no wait, kill that)
[13:52] <rberg-> ahh. I use FAI for mass installation and debconf-{g,s}et-selections to avoid debconf questions
[13:52] <pds_> flash is just evil :p
[13:53] <axisys> peetaur2: this process when goes higher than cpu .. operation starts lagging
[13:53] <axisys> operation related to the process that is
[13:53] <axisys> peetaur2: we have been seeing this for last few months
[13:53] <axisys> peetaur2: today we are thinking to automate the kill/restart
[13:54] <peetaur2> axisys: which process it it?
[13:54] <axisys> peetaur2: tac_plus
[13:56] <axisys> I am thinking of converting it to a daemontools service .. so killing/restart will be reliable.. but I want to find a quicker solution until we migrate to new system and move it under daemontools
[13:56] <jrwren> axisys: have you tried investigating why it is using so much CPU? There might be good reason.
[13:56] <peetaur2> some Cisco thing? you'd think they had enough money to hire a better code monkey to fix their stuff...
[13:56] <peetaur2> I thought Cisco was one of those "pay 3x as much as it's worth but get it done right" sort of companies
[13:57] <axisys> peetaur2: tac_plus is from tacacs+ pkg provided/opensourced by shrubbery
[13:57] <peetaur2> k... so 3rd party
[13:57] <peetaur2> well either way, I have no experience with it
[13:58] <axisys> jrwren: we did not find anything unusual .. every session ties to a core.. probably limited by disk IO and new system will have lot more memory and SSD disk
[13:59] <axisys> but trying to find a hack to find a process with high cpu usage and automate the kill/restart
[13:59] <axisys> peetaur2: ^
[13:59] <axisys> python's psutil looks pretty nice
[13:59] <pds_> http://paste.ubuntu.com/7775297/ debconf-get-selections
[14:02] <peetaur2> what's wrong with using ps -c and -o to find pid by name, then check cpu, and kill if high?
[14:02] <peetaur2> sounds like a short script
[14:03] <peetaur2> -C that is
[14:04] <pds_> hmmm examples of preeseeding files are nice but i need to find the correct keys :)
[14:04] <peetaur2> axisys: ps -C tac_plus -o "pid,%cpu"
[14:04] <axisys> peetaur2: looking at that switch now
[14:05] <peetaur2> the big example probably has most install-related keys, just not all values
[14:05] <peetaur2> then find the docs that go with it to explain values
[14:05] <pds_> i've been looking for explinations of keys but couldn't find a decent website for it just yet
[14:06] <peetaur2> well, link the official docs to prove you looked :P and then ask about an option and maybe we know
[14:06] <axisys> peetaur2: I am reading the man ps
[14:07] <pds_> https://help.ubuntu.com/12.04/installation-guide/i386/preseed-contents.html
[14:08] <pds_> i would like to install ubuntu server in english but with a belgian keyboard
[14:09] <peetaur2> pds_: k, well that's the ubuntu guide... the debian one would be more complete. But you also need a question.
[14:09] <peetaur2> pds_: sadly I think that's one thing I couldn't get right... at this company we use English language but German (eliminate dead keys) keyboard
[14:10] <pds_> well darn
[14:11] <peetaur2> pds_: here's what I set in my file, but I don't know if it worked, but clearly variant/model USA is wrong. http://dpaste.com/31K4JCK
[14:12] <axisys> peetaur2: hmm so ps -C tac_plus -o "pid,pcpu,cmd" --sort pcpu is a good start.. looking for a switch like .. display only if the pcpu is higher that 15.00 w/o awk
[14:12] <axisys> peetaur2: thanks for that hint
[14:13] <peetaur2> just parse it (rule is if there's a -o sort of thing for format, yes you can parse it) with awk, and then use bash... eatenup=15; if [ "$cpu" -gt "$eatenup" ]; then ...
[14:14] <peetaur2> trim away the % first
[14:14] <pds_> got something like this
[14:14] <pds_> http://dpaste.com/0T7FECW
[14:15] <peetaur2> pds_: deleting keys might mean it will ask instead
[14:15] <pds_> peetaur what do you mean?
[14:16] <peetaur2> I have many more keys in my file, don't I?
[14:16] <peetaur2> if you remove them instead of changing the values, it might ask instead during the install
[14:17] <peetaur2> and also you have  debian-installer/locale  twice
[14:17] <peetaur2> (and so do I... hehe)
[14:20] <pds_> let me check
[14:23] <pds_> continueing http://dpaste.com/0TEETD2
[14:53] <K4k> Tossing this out again in case anyone is here today that might know the answer.
[14:53] <K4k> If I switch the name of GID 27 from sudo to wheel, will this cause any odd behavior that anyone can think of?
[14:58] <peetaur2> K4k: you can probably make 2 groups with same GID
[15:01] <bitfury> Hi everyone
[15:01] <K4k> peetaur2: Oh, that's actually possible!? I didn't even think of that because I just assumed it wouldn't let me
[15:03] <bitfury> anyone know what version of tomcat gets installed when I select it during install?
[15:04] <bitfury> does it go for the latest stable tomcat8
[15:04] <cfhowlett> !info tomcat bitfury
[15:04] <cfhowlett> !info tomcat
[15:04] <bitfury> !info tomcat8
[15:04] <bitfury> :\
[15:05] <bitfury> what about this http://packages.ubuntu.com/utopic/tomcat8
[15:05] <peetaur2> K4k: I know it's possible for users, but not sure about groups (since the group file syntax says id and members on same line)
[15:06] <bitfury> oh nvm, failed to read it's universe
[15:26] <K4k> peetaur2: manpage for groupadd indicates there is a -o flag for non-uniq GIDs. Thanks!
[15:27] <peetaur2> K4k: nice :)
[15:44] <user123321> Suppose I want to install 2 Ubuntu or LUbuntu servers with identical server programs in each one, is CARP good for HA and LB?
[16:22] <K4k> ACK! Ok, I got my GIDs mixed up. We use GID 10 for wheel in ldap. Ubuntu sets GID 10 as uucp... I wonder if switching the GID for uucp to 14 (same as on Redhat) would break anything so that sudo could be moved to GID 10
[16:34] <K4k> nope, doesn't appear it does :) yippeee
[16:46] <zartoosh> HI how could I have entries in /var/log/boot.log timestamped?
[17:03] <TJ-> zartoosh: That is saved by "/etc/init/plymouth-log.conf", part of the Plymouth graphical boot manager
[17:06] <TJ-> zartoosh: I'm not aware of a way to have those messages timestamped unless you edited the Plymouth source-code directly, or intercepted Upstart's console output
[17:07] <zartoosh> TJ-, thanks for your answer, okay I do not want to touch any source code, so I guess the best is to intercept the upstart's console output.
[17:08] <TJ-> zartoosh: I'm thinking that'll be just as difficult!
[17:09] <zartoosh> TJ- in /etc/init/plymouth-log.conf : It flushes the boot log to disk, so it is not part of creation of entries ...
[17:09] <TJ-> zartoosh: Correct
[17:10] <TJ-> zartoosh: what is flushed to disk is the contents of the console buffer
[17:11] <TJ-> zartoosh: so unless you have a way to intercept and prefix timestamps as the messages are written to the console, there's no way to accurately timestamp them
[17:12] <zartoosh> TJ-,  oh I see, okay, thank you so much. I did googling yesterday, it seems there is a program called "grab_serial" that may help me on this ...
[17:55] <zartoosh> Hi I have installed ubuntu 14.04. I boot in uefi mode.  Whenever I reboot or powercycle system stops the boot process at grub menu and I have to manually type enter. Is there a parameter I can change so it automatically boot? thanks
[18:02] <genii> zartoosh: You can try: edit /etc/default/grub    and see if there is a line like: GRUB_TIMEOUT=        and put some small number there like 5 ( the number is how many seconds to wait before automatic booting). Then after, to do sudo update-grub
[18:02] <genii> If the GRUB_TIMEOUT line is not there, to add it.
[18:05] <zartoosh> genii, there is a GRUB_TIMEOUT set to 2. but still it needs enter... Could be screen resolution ?
[18:06] <genii> zartoosh: If it's not prompting you to choose a valid screen resolution each boot, that is probably not it
[18:06] <zartoosh> genii, it is not prompting, so it is not display issue. Let me paste the /etc/default/grub ...
[18:09] <zartoosh> genii  http://paste.ubuntu.com/7776445
[18:09] <genii> zartoosh: Looks fine there.
[18:10] <zartoosh> genii, could it be because of the uefi mode?
[18:11] <genii> zartoosh: That would be my guess. I don't know enough about EFI/UEFI to be of much assistance there, however
[18:12] <zartoosh> genii, thanks
[18:28] <user123321> Suppose I want to install 2 Ubuntu or LUbuntu servers with identical server programs in each one (eg: Apache and might be other server programs), is CARP good for HA and LB? Does anyone have experience with CARP?
[19:24] <bitbyte> i’m trying to generate a CA to my webserver know any guides ?
[19:25] <patdk-wk> you don't use CA's on webservers
[19:28] <sarnold> heh, the first guide I find uses des3 and 1024 bit rsa. otherwise decent-looking but perhaps a decade out of date..
[19:32] <zartoosh> hi I have installed ubuntu 14.04 on my system. Network interface name has changed from eth0 to em1. I removed the biosdevname package with the hope that interface name goes back to eth0, it didn't, any one could help me please?thx
[19:33] <sarnold> zartoosh: look for something like /etc/udev/rules.d/70-net-persisnte-rules or similar
[19:33] <rberg-> and delete it!
[19:33] <rberg-> I hate that file :)
[19:34] <lordievader> Allways in for trouble that file!
[19:35] <rberg-> it will be regenerated on boot but with the names you want this time
[19:35] <zartoosh> sarnold, rberg-  it seems it did not, during reboot it says it can  not make the link for eth0 device.
[19:38] <zartoosh> I actually preseeded my installation and during installation I had the biosdevname removed. Then in reboot the file 70-pres... had the device node as "eth" however it did not work either.
[19:39] <lordievader> zartoosh: Does it show any wired network interface?
[19:39] <zartoosh> lordievader, yes it does
[19:40] <lordievader> zartoosh: Ok, great. So what exactly does not work?
[19:41] <lordievader> Is it that the NIC is still called em1?
[19:41] <zartoosh> lordievader,  the network interface does not come up either as eth0 or em1
[19:42] <lordievader> zartoosh: Could you pastebin the output of both "lspci -k|grep Network" and "ip a s"?
[19:43] <zartoosh> lordievader, I do not have network access to that system I will do it with use of usb, will take me a few minutes, thank you for trying to help me ...
[19:43] <lordievader> zartoosh: No problem, and take your time ;)
[19:44] <zartoosh> One more question , in pervious ubuntu installation like 12.04 I could do Alt+F2 and get console access, but on 14.04 seems this is disabled ?
[19:45] <henkjan> zartoosh: try ctrl-alt-f2?
[19:45] <zartoosh> henkjan, I am sure I have tried that too, I am rebooting I will try it soon. t
[19:46] <lordievader> zartoosh: That was probably ctrl + alt + f[1-6], indeed.
[19:46] <lordievader> However, a server should boot straight into tty1.
[19:46] <zartoosh> lordievader, I did F2 key, but will try it soon again. thank you again.
[19:49] <zartoosh> henkjan, lordievader  Yes the <ctrl> was missing now that issue is gone, thanks
[19:49] <lordievader> ;)
[21:10] <zartoosh> lordievader,  http://paste.ubuntu.com/7777127  and http://paste.ubuntu.com/7777134
[21:18] <bitfury> I'm getting constant broken pipe messages when connecting to a server via ssh
[21:19] <bitfury> ClientAliveInterval, CountMax and TCPKeepAlives have been adjusted on the server side
[21:19] <bitfury> what am I missing? :\
[21:43] <atpa8a> hello
[21:46] <atpa8a> with vlans, can i assign an IP to both the vlan interface and raw interface?..
[21:47] <atpa8a> basically... i want to have an untagged traffic and tagged traffic
[22:11] <bekks> atpa8a: So not tag at all, not explicitely untag/tag.
[22:16] <atpa8a> bekks: i mean... is it... auto eth0\n iface eth0 inet dhcp\n auto eth0.100\n iface eth0.100 inet dhcp\n raw_device eth0\n?
[22:17] <atpa8a> dhcp here only for example
[22:20] <bekks> atpa8a: No, that are two assignments.
[22:21] <atpa8a> how can i do something like that?
[22:34] <MACscr> ok, im doing pxe installs of ubuntu for my servers and it appears that plymouth is getting installed, which i dont see a reason to have on a server. Whats the best way to disable it from being installed or removing it alltogether?
[22:35] <ikonia> just disable the splash option in grub
[22:35] <MACscr> that doesnt do it
[22:35] <ikonia> it does
[22:36] <MACscr> no, plymouth still tries to load
[22:36] <MACscr> plus i think its better to just not install it at all on a server
[22:37] <ikonia> what are you trying to disable, the splash or the whole plytmouth process
[22:37] <MACscr> all of plymouth
[22:38] <ikonia> you're going to have problems doing that
[22:39] <MACscr> why?
[22:44] <justizin> anyone have any idea wtf is happening here? : http://pastebin.com/LRenJktv
[22:44] <justizin> trying to set up amazon ses per http://docs.aws.amazon.com/ses/latest/DeveloperGuide/postfix.html
[22:44] <MACscr> justizin: ask in #postfix
[22:46] <sarnold> justizin: apparmor or selinnux or similar? check dmesg | grep DEN to see if apparmor is involved..
[22:48] <justizin> nothing from dmesg | grep DEN
[22:48] <justizin> apparmor doesn’t have a process running, though i forget if it keeps one
[22:48] <sarnold> it doesn't
[22:48] <justizin> what’s wierd is that strace says it’s trying to open the target file O_RDONLY, when its’ job is to create it
[22:49] <sarnold> is the open(.. O_RDONLY) an existence-check or something? that's strange..
[22:52] <justizin> i guess it could be, but it seems odd that it would exit on no such file or directory, since it should be creating the file
[22:54] <sarnold> justizin: try ls -ld / /etc/ /etc/postfix /etc/postfix/sasl_passwd.db -- see if the permissions look sane
[22:54] <TJ-> justizin: can the postfix user read/write that file?
[22:54] <justizin> yah i did, i mean, again, i’m root, and the target file doesn’t exist
[22:55] <justizin> it doesn’t exist :)
[22:55] <justizin> the error is not lying, but i’m confused as to why the program that should create the .db file fails when it doesn’t exist
[22:55] <justizin> https://gist.github.com/bitmonk/c1aac7825f28d57835c2 <- more detail
[22:57] <TJ-> justizin: your 'strace postmap' is using the .db file as the input
[22:57] <justizin> oh yeah, that was the wrong command
[22:58] <justizin> han gon
[22:58] <TJ-> justizin: while your at it, can you add the result of "ls -al /etc/postfix/" ?
[22:59] <sarnold> justizin: so strange that it says "Permission denied" in the short-paste but "No such file or directory" in the strace output..
[22:59] <justizin> sarnold: that, too
[23:00] <justizin> no that’s from my erroneous paste, new gist coming
[23:01] <justizin> https://gist.github.com/bitmonk/3fecf9b13e02a408e82b
[23:01] <justizin> so it is actually permission denied, that is less confusing
[23:01] <justizin> except that i am root
[23:01] <justizin> i am able to create that file with ‘touch'
[23:02] <TJ-> justizin: "open("/etc/postfix/__db.sasl_passwd.db", O_RDWR|O_CREAT|O_EXCL, 0644) = -1 EACCES (Permission denied)"
[23:02] <justizin> right
[23:03] <TJ-> justizin: The file it is failing for is "__db.sasl_passwd.db", presumably a temporary
[23:03] <justizin> i can touch that as well
[23:03] <sarnold> justizin: notice the line 299, it changes it's effective uid to 1006.
[23:03] <justizin> ah
[23:03] <justizin> it’s becoming postfix
[23:04] <sarnold> hrm, shouldn't >1000 be for user accounts?
[23:04] <TJ-> Yeah, it drops privs, which is why I asked if the postfix user has access to the directory
[23:04] <sarnold> the postfix user doesn't have write access to the directory
[23:04] <parallel21> can cifs be mounted over different subnets?
[23:05] <sarnold> parallel21: afaik cifs can be mounted over the internet, no?
[23:05] <justizin> yeah i chown-ed /etc/postfix to the postfix user and i get the same
[23:05] <justizin> oh
[23:06] <justizin> it’s becoming ‘hm'
[23:06] <justizin> right because.. augh.. i hate this, i want so bad to undo it ;d
[23:06] <justizin> we run _everything_ as one user, because no reason
[23:06] <sarnold> :(
[23:06] <parallel21> thanks sarnold
[23:06] <justizin> and that does it
[23:06] <justizin> thanks (facepalm)
[23:07] <justizin> well, it’s actually wierd, hum, we run postfix as the postfix user
[23:07] <justizin> ah it runs as the source file owner
[23:08] <justizin> -o squashes that behavior
[23:10] <justizin> obviously we should not be doing that
[23:11] <justizin> but every time i try to change it in one place i get ramblings about following our existing practices ;d
[23:11] <sarnold> so, how's the resume? :) all polished up and ready to go I hope? :)
[23:11] <justizin> anyway thanks for bouncing ideas around w/ me guys
[23:12] <sarnold> good luck justizin :)
[23:12] <justizin> pff, i’m still relatively new here, i just have to smash some better practices into place
[23:12] <sarnold> oh, okay, so it's not hopeless
[23:12] <justizin> the problem is, noone is ever like ‘spend the next two weeks un-fucking our config mgmt’
[23:12] <justizin> no it’s just like years of bad habits woven into the code
[23:13] <justizin> the rationale for doing it the stupid way is not so bad, which is basically that for the most part each machine only runs one thing, and if you break into that, you have access to anything very important on that machine
[23:13] <justizin> but mail is an obvious exception
[23:13] <sarnold> I can understand the point, but working against he assumptions of the software involved is another matter :)
[23:13] <justizin> i totally agree, that’s my counter
[23:13] <justizin> i’m like okay but.. we get so much for free!
[23:14] <justizin> it just doesn’t feel that way because we are set in our ways about going against the grain, so years of changes in upstream behavior and packages have completely eclipsed us
[23:14] <justizin> the attitude is often like ‘oh yeah you have to do that thing because <software package> is dumb'
[23:14] <justizin> and it’s like no, it’s not dumb, it expects to have its’ own user
[23:14] <justizin> that’s completely reasonable 1970s practice! ;d
[23:14] <sarnold> haha
[23:15] <sarnold> well, good luck fighting the good fight ;)
[23:15] <justizin> yeh
[23:15] <justizin> there are worse fates ;)
[23:15] <sarnold> :)
[23:15] <justizin> anyway tks
[23:23] <MACscr> ok, got another question. I am provisioning my ubuntu systems with pxe and my finish template is failing and i have to hit continue for it to finish the install because of the failed exit code. Is there a log where i can see the error after it boots for the first time or should i be able to switch to another console window and see the error somewhere?