=== markthomas|away is now known as markthomas === jvwjgames_ is now known as jvwjgames === markthomas is now known as markthomas|away === jvwjgames_ is now known as jvwjgames === sjm is now known as Mc_Kraken === zz_DenBeiren is now known as DenBeiren [07:04] morning anyone running percona 5.6 ? I have dependency issues with it === Lcawte|Away is now known as Lcawte === Lcawte is now known as Lcawte|Away === Lcawte|Away is now known as Lcawte === DenBeiren is now known as zz_DenBeiren [09:19] Good morning. [10:38] lordievader: Good Morning ... [10:38] o/ [10:38] ;-) ... [10:46] kickinz1: looking at Docker now. Can I see the format-patch output or your git tree please? Is this up-to-date on docker-dev? [10:47] rbasak, I think so let me check [11:16] rbasak, docker-dev up-to-date === Lcawte is now known as Lcawte|Away === Lcawte|Away is now known as Lcawte [11:42] Hello . What do you do when a message comes in, that you send packages to a arbitrary control server? They only give you the source IP which is the firewall. That is like searching the needle in a haystack. === kickinz1 is now known as kickinz1|lunch === kickinz1|lunch is now known as kickinz1 [12:57] Putting together an NFS server. 12x 6TB disks via raid controller/raid60. XFS filessytem. Will only serve as an NFS server, do I bother with anyont more then 8GB of ram? [13:00] lucidguy: That is more depended on the amount of clients [13:01] Probably around 60 clients [13:01] Then 8G should be more then enough - however, the more ram you have - the more will get cached in ram [13:01] will speed up reads [13:02] I guess it wil be on 1Gbps? [13:02] Yes. [13:04] more ram is always more ram for cache, but I think 8G will be sufficien t [13:10] I agree [13:20] I would be more concerned about writes to the raid60 [13:20] and our stripe size [13:21] to determine how much ram you need, overtop of your working set size [13:23] kickinz1: failed to build on powerpc: https://launchpadlibrarian.net/202757945/buildlog_ubuntu-vivid-powerpc.docker.io_1.5.0~dfsg1-1ubuntu1_BUILDING.txt.gz [13:23] I don't know if we care or not. [13:25] rbasak: I don't know either. [13:36] stgraber: lxc in vivid still needs commit 89a4ec737e97fdf2856fda94b816875e98155c82 === crazybluek is now known as Blueking [14:00] Upstart question again, why is it job with script stanza only eventually goes into stop/waiting state after being started? [14:01] while with pre/post-start, it remains in start/running and block further start invocation once started [14:54] hallyn: did you talk to wgrant? he pinged me about that very issue last night :) [14:54] hallyn: anyway, I plan on tagging 1.1.2 today which will include this [14:58] HI === crazybluek is now known as Blueking [15:32] hello [15:33] stgraber: no, i was just checking to see if my running-in-container update fully fixed nested lxc [15:34] excellent, thx [15:39] rbasak: https://bugs.launchpad.net/ubuntu/trusty/+source/openldap/+bug/1103353 [15:39] rbasak: won't fix it please [15:40] rbasak: issues mention at #19 seems to happen only when you install this package over your own one built against openssl not gnutls [15:40] rbasak: which means that my point at #18 is still valid [15:43] strikov: done [15:43] rbasak: tnx [15:43] rbasak: i'll put a final note there in a minute [15:45] rbasak: ah, you put one already, thanks! [15:45] :) === markthomas|away is now known as markthomas === CiPi is now known as cipi === mrt333_ is now known as mrt333 [16:46] Raid question. 12 disk raid10 array. Technically I can lose a total of 6 disks and the array will stay online. What if I loose two disks of the same pair? Array is gone? [16:59] stupid question, and I think rbasak and sarnold may want to weigh in, but would anyone hate me if I made a catch-all bug for "Can I get somemodule added to nginx?" (where somemodule is any module the community wants) [17:00] (it'd apply to my ppas project where I do bugtracking on the staging PPAs, the nginx project on LP, and the Ubuntu package) [17:02] the only reason I asked is because I've seen this request a thousand times in my email or in Debian, and want a catchall down here that says Debian should include it first, or at least get a voice [17:02] (same idea of including a new package somewhere, in that it's easier to get it into Ubuntu via Debian) [17:10] teward: sounds like you actually want an FAQ entry or template standard response instead of a bug itself? [17:11] teward: I don't think anybody will object if you manage nginx bugs how you please though. We appreciate that you do look after them. [17:11] is there a way to get changelogs from the ubuntu cloud archive without installing it? (like packages.ubuntu.com or launchpad ppas) [17:13] rbasak: considering the number of feature reqs in Debian, yes. [17:13] rbasak: I should set up a wiki page for NGINX xD [17:13] gQuigs: I think they're managed in bzr somewhere? If so you could find that in a browser somewhere. I don't know where that might be though. Does Vcs-Browser point to anything on an installed package? [17:14] rbasak: the big idea here is we don't want to *increase* the delta between debian and Ubuntu, so... since i see the requests *way* too often, thought I'd preempt em. [17:16] rbasak: done though :) === kickinz1 is now known as kickinz1|away [18:50] Hey peeps, quick question. My google foo has failed me on this one, so hopefully someone can help... [18:50] I have a question that google is failing me on... How do I have 2 separate DHCP configured interfaces (eth0, eth1) and only set the resolv.conf entries for eth0? [18:50] eth1 keeps overwriting resolv.conf when it comes up, with info that is not valid :( [18:51] mpanetta: maybe uninstall resolvconf and manage it yourself? [18:52] I don't seem to have resolveconf installed. At least the dir does not exist in /etc [18:53] Just upgraded a VM server to v15.04 and VMware tools doesn't seem to like systemd. Install is failing. Any tricks? [18:53] mpanetta: check dpkg -l resolvconf [18:53] sarnold: says un, so not installed [18:54] mpanetta: aha [18:55] hmm, I think dhclient may be editing it for me? [18:55] I could chatrr it +i I suppose [18:55] er chattr [18:55] Seems kinda like a hack tho [18:56] in theory you could set dns-nameservers for both interfaces in /etc/network/interfaces to be the DNS servers you want... but, no guarantee. [18:56] i'd test but i have to finish updating my VM labs :p [18:57] teward: Yeah, it seems to ignore that [18:57] there may be a way in /etc/network/services to set what you like.. I've never had to fight this one though, I'm running short o nideas fast [18:57] mpanetta: which release is this [18:57] 14.04? etc. [18:58] I've read that I can disable the dhclient from requesting that info, maybe that will help... [18:58] teward: Yes 14.04 [19:00] I know that the protocol allows it, whether the config files make it easy is another matter :) [19:00] Actually let me try that... [19:00] sarnold: We are about to find out ;) [19:00] My guess would be to list the specific interfaces in their own section in /etc/dhcp/dhclient.conf [19:01] and not request domain-name-servers for the interface you don't want [19:01] Ubuntu server has switched to systemd, no? Now I'm unable to install vmware-tools. Getting the following error: "initctl: Unable to connect to Upstart: Failed to connect to socket /com/ubuntu/upstart: Connection refused" Any help? [19:01] or, alternatively, just add a global "prepend domain-name-servers 8.8.8.8;" in there [19:02] Woop ok removing dns-servers and so forth from the request line in dhclient.conf fixed it [19:02] concord: please file a bug against the vmware-tools package, add the systemd-boot tag to the bug [19:02] teward: sarnold mdeslaur: thanks for the help! [19:03] sarnold: this is a package from vmware, not an ubuntu package [19:03] I'm trying to install in server v.15.04 [19:03] concord: hunh, I thought it was in our archive these days [19:03] Yea? [19:03] Hmmmmm [19:04] concord: Y [19:04] concord: https://launchpad.net/ubuntu/+source/open-vm-tools [19:05] sarnold: I will give that a try, this is vSphere 5.0.0, shouldn't matter? [19:05] concord: sorry, no idea there [19:05] sarnold: thank you! [19:06] concord: if it doesn't work out and vmware doesn't get around to fixing it soon enough, I think upstart will still be available as a fallback https://wiki.ubuntu.com/SystemdForUpstartUsers#Switching_init_systems [19:08] sarnold: concord: vmware-tools i think won't build under the later kernels in 14.10 and later [19:08] although they updated vmware-tools lately from the vmware tools ISO, and I haven't tested since [19:08] open-vm-tools should work for most of the functions otherwise though [19:32] teward: sarnold: I have installed the package and vCenter reports that HA is up and running, although it does clearly say that vmware tools are "3rd-Party/Independent" and not genuine VMware so I'm guessing support is out the window. But hey ... that's what you guys are for! [19:32] :) [19:35] concord: interesting, they asked for the packages to be moved to main.. === markthomas is now known as markthomas|away === kickinz1|away is now known as kickinz1|afk === markthomas|away is now known as markthomas === cipi is now known as CiPi === bilde2910 is now known as bilde2910|away