=== markthomas is now known as markthomas|away === zz_DenBeiren is now known as DenBeiren [03:02] hi guys anyone cna help how to add additonal ssh-keys on an existing ubuntu server -i have na existing one and its pretty working ok.. [04:42] hi guys when i ssh-keygen -t rsa on my ubuntu server it generates my public key and copy this to my authorized_keys but how do i ssh putty on the server after..? [05:04] Newbie with a prob. Installing ubuntu server 12.4 onto ausu eeebox 1030 (reused machine) installs, but does not seem to boot screen goes off, no keyboard or power switch function. ISO created using ubuntu startup creator. Anyone have suggestions? [05:16] seems a common problem on forums, but no one seems to have a solution. === freeflyi_ is now known as freeflying_ [08:58] Good morning. [09:45] any tools to identify browsers of clients for ubuntu? [09:58] Alina-malina: Doesn't Apache already note that? [09:59] it mess the user_agent [09:59] and cant determine [10:00] i need something like fingerprints or something like that === Lcawte|Away is now known as Lcawte [10:00] Alina-malina: That is what the browser itself sends out... [10:00] yes, well i thought maybe some additonal things that can inject into browser [10:00] anyways [10:00] thanks for help === Lcawte is now known as Lcawte|Away === Lcawte|Away is now known as Lcawte [11:21] hi to everyone. I have a acer one with ubuntu server, no gui/desktop just terminal/console. does anyone knows something like teamviewer to share terminal? This computer is behind NAT so I cannot use a ssh. [12:15] hi all. small question how can i increase the memory allocation for iptables ? at the moment i am getting this error message whenever i try to add a iptables rule to a specific chain => iptables: Memory allocation problem === Lcawte is now known as Lcawte|Away [12:46] hello [12:47] What I need to set up in domain to configure mail server on CDN? CDN has a different IP than productive server. [12:48] xperia: do you have a multi-cpu system? [12:49] xperia: because afaik iptables makes a copy of the ruleset for each cpu, so it's possible you simply run out of system memory whith really large rulesets on a multicpu machine [12:50] adsc: yes i have a quad core cpu. the people at netfilter said i should switch to a 64 Bit Kernel because => "i.e. 1GB limit for kernel memory. moreover, iptables uses vmalloc area which is limited to 128mbyte on 32bit kernel" === Lcawte|Away is now known as Lcawte [12:52] i ahve also hughe problems with setting up ipset. it looks like ipset is not availble anymore or does not work with ubuntu. at least i am not able to rebuild it anymore on ubuntu. === kickinz1 is now known as kickinz1|afk [12:54] xperia: and you can't switch to a 64bit kernel? === kickinz1|afk is now known as kickinz1 [12:56] adsc: well this will need a total revision of the server as everything need to be switched to 64 Bit i think or i am wrong ? if it is only the kernel it would be easy. guess need a full new install of a new server. [12:56] yeah, it would be best if you could do a new install [12:58] adsc: more new additional work ohhhhh ! thank you lot for your replys and answers. appreciate it. [13:18] jamespage, glance systemd updates - https://code.launchpad.net/~corey.bryant/glance/systemd === Lcawte is now known as Lcawte|Away === bilde2910|away is now known as bilde2910 [14:16] coreycb, no templating? [14:17] jamespage, I just went with what Debian had [14:17] jamespage, I can take a stab at templates if you prefer [14:17] coreycb, +1 [14:17] it should be a rename of the files and stripping out most of the content [14:18] jamespage, ok I'll work up some templates [14:20] Hi all. Quick question. I know this isn't a security channel but in the context of SSH'ing to a headless server with PuTTY, is there a way to revoke a private/public key pair generated in puttygen (I don't see the option) or is shredding the private and public key files sufficient? [14:20] (Redoing keys due to accidentally not saving the passphrase) [14:25] Stuxnet: it is sufficient to destroy the copies of the private key [14:25] Stuxnet: Remove it from the authorized keys file? [14:25] Stuxnet: you might want to take it out from authorized_keys, that's roughly equivalent to revoking it [14:25] (if you remember all of them) [14:28] Okay, I did that first. I'm just going to shred the key file that was saved from puttygen. Thanks guys! [14:54] is there a parameter for the dhclient command that allows me to get an IP but ignore all other options OR simply display the IP received by dhclient but not actually set it? [15:02] bananapie: you might be able to replace dhclient-script with -sf with a trivial wrapper that just prints arguments and environment. I'm not sure if that'll work but hopefully that's a start if you can't find anything else. [15:11] coreycb, hmm horizon - do you have a kilo-1 mp for that? [15:13] jamespage, I didn't create an mp yet but I can. It's waiting on python-pint getting to main. [15:13] coreycb, create it now - I'll review and upload [15:13] it can sit in the queue and then as soon as the MIR lands, it goes in [15:13] jamespage, https://code.launchpad.net/~corey.bryant/horizon/2015.1-b1/+merge/246314 === Lcawte|Away is now known as Lcawte === Lcawte is now known as Lcawte|Away [15:39] hello, i keep getting errors like this datatec named[495]: error (unexpected RCODE REFUSED) resolving 'datatec.es/A/IN': 213.251.188.153#53 [15:40] i was looking in named.conf.* but i dont see anything [15:48] coreycb, I can see the pint dependency [15:49] jamespage, can't see it? [15:49] coreycb, in the upstream requirements files? [15:50] coreycb, oh wait I'm being dumb [15:51] jamespage, ok [16:25] coreycb, horizon uploaded [16:25] coreycb, it will build just fine and pull in a runtime dep on pint [16:26] jamespage, great, thanks [16:26] coreycb, I rebased the download juju env file patch as well [16:26] fairly trivial [16:28] jamespage, thanks, I must have missed that it was disabled [16:28] coreycb, np [16:45] smb, talking of kernel problems, we seem to be seeing issues with partition table freshness on vivid [16:45] our ceph and swift testing scrubs existing partitions and reformats but we are seeing: http://paste.ubuntu.com/9731880/ [16:46] 100% of the time [16:46] raising a bug now [16:46] jamespage, I was about to ask about lp bugs :) [16:46] jamespage, Also that is with 3.16 or 3.18 already [16:46] smb, we only switched on vivid testing today as kilo-1 is nearly done [16:46] 3.18 I think [16:46] just re-deploying to make sure [16:47] could be if you set up recently [16:47] jamespage, glance updated - jamespage, https://code.launchpad.net/~corey.bryant/glance/systemd/+merge/246227 [16:48] jamespage, the kernel team ppa has also 3.19 for early testing. If that is of help === bilde2910 is now known as bilde2910|away [16:54] smb, 1410363 [16:54] bug 1410363 [16:54] Launchpad bug 1410363 in linux "partition table updates require a reboot" [Undecided,New] https://launchpad.net/bugs/1410363 [16:54] rather [16:54] jamespage, oh and btw... was there not a similar thing at some point err last year... [16:55] smb, apparently so - beisner remembers [16:56] Hmmm something with umount removing the mount but not releasing the partition... [16:56] coreycb, uploaded - thanks! [16:56] which went away at some point... [17:06] good day all, been googling for an answer wihtout luck, hoping to find some knowledgable assistance on an issue I'm having with my local repository [17:07] looking for a way to include the packages in a PPA in my local repo. We use our local repo to host custom versions of packages, for example to have the most up to date zsh available on our servers [17:07] I'd like to include the packages in a PPA, as opposed to building the packages on my repo. [17:08] specifically, I want to get the latest build of duplicity on a 12.04 server from my local repo. [17:08] I can build a duplicty package and make it avialable in my repo, but I'd like to be able to just use packages in the official PPA without having to build the package myself and include it in my repo === kickinz1 is now known as kickinz1|afk [17:10] reason being that I don't wan't to add a PPA to all my servers. We manage our boxes via chef, so I COULD add the PPA and ensure that the latest duplicty is installed, but that seems havy handed when I've already got a repo on my lan specifically to avoid that kind of PPA proliferation === markthomas|away is now known as markthomas === kickinz1|afk is now known as kickinz1 [17:19] jamespage, that error "2015-01-13 16:50:09 INFO mon-relation-changed mkfs.xfs: cannot open /dev/vdb1: Device or resource busy" doesn't that imply the partition is open excl already ? which would also block the reread ioctl? [17:19] apw, the block device gets unmounted, re-paritioned, re-formatted and then re-mounted [17:19] it's failing on the re-formatting bit [17:19] so its likely the unmount was foo-bar I guess === kickinz1 is now known as kickinz1|afk [17:20] jamespage, yes, it is failing in a way that implies it is not unmounted [17:20] so the first things to do is to check that the umount is working [17:21] jamespage, as saking to change partition sizes which are in use will fail, with that reboot required thing, and the subsequent format fails too [17:21] jamespage, and the error handling is non-existant for it to get that far one presumes [17:22] apw, the block device does not appear in 'mount' output or the content of /proc/mounts [17:23] could udev be playing with it perhaps [17:24] jamespage, does lsof say anything at that point, also if you do the proceedure by hand, does it work [17:24] as you are mounting it with a vfat or something in it, unmounting, changing the partition tables, (and by here it has gone wrong) [17:24] jamespage, I cannot remember the bug number but a state of where umount would seem to work btu not release the partition... thats was what I remember from last time [17:24] apw, smb: http://paste.ubuntu.com/9732538/ [17:25] that feels familiar [17:25] apw, ext4 [17:25] then switching to xfs [17:25] also mbr->gpt [17:25] though I think oh right the jbd process was involved there [17:26] smb, I seem to remember this from before [17:26] jamespage, ok so that implies that the journaling block device is holding on to it? [17:26] apw: probably [17:27] found a work around, though I'm still interested in my original question re: mirroring packages from a PPA in my personal apt repo [17:27] apw, Yeah that was the weird state I remember. [17:27] jamespage, so, is it possible to perform these steps on an instance by hand [17:27] to see what happens? [17:27] apw, Which IIRC I could not reproduce locally [17:31] apw, so if I just unmount /mnt (where cloud-init put it) I see the vdb parition in lsof [17:31] apw, for reference its not partitioned - just /dev/vdb [17:32] jamespage, are you mounting it r/o or r/w ? [17:33] e/w [17:33] r/w [17:34] smb, apw: https://bugs.launchpad.net/ubuntu/+source/ceph/+bug/1371526 [17:34] Launchpad bug 1371526 in linux "ceph-disk-prepare command always fails; new partition table not avaliable until reboot" [Undecided,Expired] [17:34] jamespage, might be interesting to try mounting it r/o, to see how that fares [17:35] jamespage, Oh right that was the bug [17:35] yeah [17:36] apw, if I unmount it and then remount it ro, it fails [17:36] rw is OK [17:36] fails how ? [17:38] apw, could be trying a second jbd which cannot get exclusive lock [17:38] jamespage, this partition is being mkfs'd "outside" the instance i assume, what version is that space running [17:38] apw, trusty [17:39] apw, I wonder whether this might be a race as in when cloud-init does the mount relative to startup... [17:42] jamespage, where can i find the code which prepares this disk outside [17:42] jamespage, or ... can we get a dd of the partition without mounting it [17:43] apw, I would propose I try to try to repro tomorrow with a cloud-image and data on ext4 [17:43] smb, sounds reasonable [17:43] smb, i would think just preparing the fs outside, and then trying to mount it in the instance, bear in mind the instance boot has done a resize too [17:43] apw, http://cloud-images.ubuntu.com/ [17:44] just done a resize on vda === jpds is now known as Guest41488 [17:46] apw, Yeah, I think you can dd the cloud-image on any bigger vda and it resizes itself on first boot [17:47] apw, Oh and cloud-localeds might be the tool for the data [17:47] apw, Just from my memory that was restricted to isofs [17:48] smb, yeah we need to know exactly how that things is made, as i suspect it is relevant, note in jamespage's original bug a second vdc disk was not affected when mounted and unmounted from inside, so there likely is a correlation to the outside creation [17:49] apw, Either outside creation or when all that stuff is done relative to the boot sequence [17:49] apw, smb: https://github.com/openstack/nova/blob/master/nova/utils.py#L794 [18:20] hello [18:20] I installed ispconfig [18:20] And I tried to configure mail server [18:20] I clicked somewhere in ispconfig panel [18:20] and whole site is broken [18:20] server even shows php code.... wtf [18:21] There is no helpful information in log. [19:25] Hi all. What is the current kernel for Ubuntu serve 14.04 (fully updated install)? [19:25] !info linux-image-generic trusty [19:25] linux-image-generic (source: linux-meta): Generic Linux kernel image. In component main, is optional. Version 3.13.0.44.51 (trusty), package size 2 kB, installed size 28 kB [19:26] superboot: ^ [19:28] lordievader: Thanks a lot. :) [19:34] I feel like I'm out of date. Is update-rc.d still the way to handle service automatic start/stop? [20:02] hi all [20:02] i install apache2 and php5 [20:04] the problem if script cannot parse by php, the webserver will reply respond 500 ? [20:04] thanks [20:07] fandi: Yes, if php crashes Apache will most likely report a 500 error. [20:09] lordievader: ok .. how to make error not 500 :) [20:10] lordievader: because i only change simple script and all vhost and all my cluster server respond 500 :) [20:10] also my varnish :) [20:11] fandi: Write non crashing code? === martinst is now known as martins-afk [20:12] lordievader: ok .. :) [20:12] lordievader: do you familliar with varnish :) [20:12] lordievader: how to increase hit rate on varnish ? [20:12] No, I don't know varnish. [20:13] lordievader: ok thanks [21:13] I want to compile a library and an app that uses it and have few questions relating to that. How do I update? Do I just compile the new version normally and everything is replaced and leaves no unneeded trash? Also if there're any possible problems when compiling one part (library) and installing other things from repos? === markthomas is now known as markthomas|away === esde is now known as Guest54912 === Guest54912 is now known as esde === martins-afk is now known as martinst === martinst is now known as martins-afk === mfisch is now known as Guest85585 === popey_ is now known as popey === jrgifford_ is now known as jrgifford__ === markthomas|away is now known as markthomas [22:52] after installation of ubuntu server, I end up in initramfs console [22:52] during installation, the drive was sdb1 [22:52] now it is sda1 [22:52] can I change that somehow in initramfs console [22:52] cat /proc/cmdline shows the wrong command line === hamnstar_ is now known as hamnstar