[00:19] <{bosco}> ok
[00:57] <d_kam> Hey guys.  Package installation question.  If I have ubuntu 16's nginx package installed, can I install nginx from another PPA ( EG, PhusionPassenger ) without issue? Or do I need to uninstall the standard nginx package first?
[00:58] <sarnold> you almost certainly need to uninstall the distro-provided package first unless the other packager has gone to lengths to make side-by-side install an option
[00:59] <sarnold> you could use apt-get download ... to grab the binary package, ar x the thing to get data.tar.* file, and tar tf that data.tar.* file to see what filenames it uses. if it uses 'standard' names, then you have to uninstall the distro one first.
[00:59] <sarnold> if everything goes into /opt/phusionphun/nginx/ or something like that then you might be fine
[01:03] <d_kam> Thanks sarnold
[01:04] <d_kam> Any idea when the standard nginx package is built against openssl 1.0.1, when the OS provides 1.0.2g?
[01:05] <d_kam> I want HTTP/2 which now requires ALPN, which requires nginx to be built with OpenSSL 1.0.2d(?)
[01:20] <sarnold> that's probably a "minimum version". 16.04 LTS ships with openssl 1.0.2g with a load of patches for security fixes, so it's probably new enough for the ALPN..
[01:25] <sdeziel> d_kam: if you get back, nginx in Xenial already supports HTTP/2
=== dkam is now known as d_kam
[08:27] <oskaress> Hey guys, when I want to add a new user to an existing group, shouldn't the command just be adduser $user $group --gecos <...> and then chpasswd on a new line to set the password?
[09:11] <jamespage> o/
[09:13] <lordievader> oskaress: The way I add users to  a group is `sudo usermod -a -G <groups> <user>`.
[09:14] <oskaress> lordievader But I want to create the user and in the same command add it to an existing group, that is possible right?
[09:15] <lordievader> oskaress: Yes, read the useradd man page: https://linux.die.net/man/8/useradd
[09:15] <lordievader> -G flag.
[09:16] <oskaress> Thank you
[09:52] <RoyK> oskaress: try man useradd ;)
[13:01] <Oskars_> Is there any way to use adduser and create a user without creaing a group with the same name?
[13:06] <andol> Oskars_: adduser --ingroup ExistingGroup NewUser
[13:06] <Oskars_> andol, Thank you, will try!
=== giraffe is now known as Guest431
[13:15] <rbasak> cpaelzer: re bug 1658469, I think the goal is still to enable HTTP/2 in apache2 in Xenial. It just needs figuring out the details.
[13:15] <ubottu> bug 1658469 in apache2 (Ubuntu) "mod_http2 is not available under Apache 2.4.23 / Ubuntu 17.04 xenial" [Low,Triaged] https://launchpad.net/bugs/1658469
[13:15] <rbasak> nacc: ^ that's right I think?
[13:18] <coreycb> jamespage: i'm going to get started on newton point releases today: https://bugs.launchpad.net/cloud-archive/+bug/1718033
[13:18] <ubottu> Launchpad bug 1718033 in Ubuntu Cloud Archive newton "[SRU] newton stable releases" [Medium,In progress]
[13:27] <Oskars_> andol, I don't get it, when I run it, it seems to work, says Adding new user <user> (1001) with group <group>, but when I run "grep <group> /etc/group" the new created user it not in the group which it earlier said he was added to...
[13:28] <Oskars_> is not* in the group
[13:29] <andol> Oskars_: It wouldn't, as only secondary groups show up that way
[13:29] <andol> Oskars_: Try `groups username`
[13:30] <Oskars_> andol, Ah I see.. 'groups <username>' worked, now I see it. Just got to figure out why I can't FTP in with the user now haha
[13:31] <andol> Oskars_: If you look in /etc/passwd you will see that the GID of the primary group is listed as the fourth value.
[13:36] <jamespage> coreycb: ack I was doing likewise on ceph 10.2.9
[13:36] <jamespage> and and ovs updates
[13:36] <Oskars_> andol, Yeah they have the same GID so that looks correct
[13:36] <coreycb> jamespage: awesome
[13:37] <cpaelzer> rbasak: well I was following naccs last comment in there, but he will know and can correct my misunderstanding if it was one
[13:55] <coreycb> beisner: neutron 2:8.4.0-0ubuntu5~cloud0 is ready to promote to mitaka-updates for bug 1668410
[13:55] <ubottu> bug 1668410 in neutron "[SRU] Infinite loop trying to delete deleted HA router" [Medium,In progress] https://launchpad.net/bugs/1668410
[13:57] <coreycb> beisner: jamespage: there are a few tempest smoke test regressions with the nova and libvirt in mitaka-proposed that i want to dig into before promoting them
[14:03] <beisner> coreycb dosaboy ivoks - neutron is now in mitaka-updates re: https://bugs.launchpad.net/neutron/+bug/1668410
[14:03] <ubottu> Launchpad bug 1668410 in neutron "[SRU] Infinite loop trying to delete deleted HA router" [Medium,In progress]
[14:04] <beisner> thx for validation on that coreycb
[14:04] <coreycb> beisner: np, ty
[14:04] <ivoks> beisner: coreycb thanks!
[14:14] <Pinkamena_D> on AWS machine, 16.04 server, I have uninstalled resolvconf. After updating contents of /etc/resolv.conf, it still gets overwritten on reboot, but there is no header about resolvconf doing it now, just the other lines. What else is modifying this file?
[14:17] <iliv> I'm installing Ubuntu Xenial 16.04.03 using server amd64 ISO installer and I ran into a problem creating RAID-10 out of 4 physical disks I have in this server. I created my "physical raid" devices, all identical to each other, utilizing all the 4 physical drives. I specify that there will be 4 active devices and 0 spares. When the installer attempts to actually create the RAID device, it says "Not enough RAID partitions available" and strangely "You
[14:17] <iliv>  have 1 RAID partitions available but your configuraiton requires 4 partitions". I checked several guides online, including official Ubuntu Wiki and I seem to be doing everything correctly but, unfortunately, this error.
[14:18] <sdeziel> Pinkamena_D: it's a wild guess but maybe it could be cloud-init? I don't know if it normally interacts through resolvconf or not
[14:18] <iliv> I tried creating RAID1 using just 2 disks but it failed too displaying similar error message with the only difference in number of available vs required partitions.
[14:18] <jamie_1> anyone have any ideas why i cant get into my root when connected with ssh? I changed the setting in /etc/ssh/sshd_config to PermitRootLogin yes    and its still not letting me. I changed and saved and then did systemctl restart ssh.service and a full reboot
[14:19] <iliv> I don't really have any ideas what could be wrong
[14:19] <iliv> here's a screenshot of the error message: https://i.imgur.com/SCbK6GX.png
[14:20] <sdeziel> jamie_1: check in /var/log/auth.log, it should tell you why root is being denied
[14:24] <jamie_1> sdeziel: its just telling me ubuntu-server sudo: pam_unix(sudo:auth): auth could not identify password for [jamie]
[14:24] <jamie_1> i know my password is correct
[14:25] <sdeziel> jamie_1: that is a log entry from sudo, not sshd. Try "grep sshd /var/log/auth.log"
[14:27] <jamie_1> sdeziel:  its just saying that it accepted my public key and opened a session for me
[14:27] <sdeziel> jamie_1: can you share the line (you can obfuscate the IPs if you want) ?
[14:29] <jamie_1> sdeziel: their all local ip's right now so it doesnt matter https://pastebin.mozilla.org/9032730
[14:30] <jamie_1> thats all it says even after my attempts to use sudo
[14:30] <sdeziel> jamie_1: OK, " Accepted publickey for jamie" seems to indicate that you didn't  tell SSH to log as root (try "ssh root@192.168.1.15" ?)
[14:30] <jamie_1> sdeziel: one sec
[14:32] <jamie_1> sdeziel: https://pastebin.mozilla.org/9032731
[14:32] <jamie_1> my bad wrong section XD
[14:32] <sdeziel> "Failed password for root"
[14:33] <jamie_1> sdeziel: i know its bad security but for sanity checking reasons i changed the root password to the same as my standard user and its not accepting the password
[14:34] <sdeziel> jamie_1: you might want to try with SSH keys, those are secure and you can't typo them ;)
[14:35] <jamie_1> sdeziel: how would i add my rsa to root?
[14:36] <jamie_1> im used to just accessing as user and sudoing
[14:36] <jamie_1> i like to sudo and not touch the root directory
[14:38] <jamie_1> would i just do a cp /home/jamie/.ssh/authorized_keys /.ssh/authorized_keys
[14:38] <sdeziel> jamie_1: there are many guides explaining that but the gist of it is: as jamie, escalate to root with "sudo -i", create the  the ssh dir in root's home with "mkdir -m 0700 ~/.ssh", put your *public* key in root's authorized_keys with "vim ~/.ssh/authorized_keys"
[14:39] <jamie_1> so yeah, create the directory and then move the key into there
[14:39] <jamie_1> thanks a bunch
[14:39] <sdeziel> jamie_1: mostly, just make sure the perms/owner are right
[14:44] <hehehe> hi
[14:45] <hehehe> I am playing with xclip
[14:45] <hehehe> xclip -o -sel clip > webpage.txt to get clipboard data into file - however it overwrites existing data
[14:45] <smoser> rbasak,
[14:45] <smoser> $ uvt-kvm create sm-a1 release=xenial
[14:45] <smoser> uvt-kvm: error: libvirt: internal error: qemu unexpectedly closed the monitor: 2017-09-19T14:45:45.076265Z qemu-system-x86_64: -chardev pty,id=charserial0: char device redirected to /dev/pts/22 (label charserial0)
[14:45] <smoser> warning: host doesn't support requested feature: CPUID.80000001H:ECX.svm [bit 2]
[14:45] <smoser> 2017-09-19T14:45:45.115161Z qemu-system-x86_64: -vnc 127.0.0.1:0: Failed to start VNC server: Failed to listen on socket: Address already in use
[14:45] <smoser> did you know this ?
[14:45] <hehehe> can xclip add data to a text file?
[14:46] <hehehe> instead of delete add
[14:46] <cpaelzer> I know smoser
[14:46] <smoser> hehehe, xclip -o -sel clip >> webpage.txt
[14:46] <smoser> '>>' is append
[14:46] <smoser> or in vi, open the file and type
[14:47] <cpaelzer> smoser: looking up the bug
[14:47] <smoser> :r! xclip -o -sel clip
[14:47] <cpaelzer> smoser: https://bugs.launchpad.net/ubuntu/+source/libvirt/+bug/1717442
[14:47] <ubottu> Launchpad bug 1717442 in libvirt (Ubuntu) "VNC auto port allocation affected by daemon state" [Undecided,Incomplete]
[14:47] <cpaelzer> smoser: I ended up not being able to reproduce anymore
[14:47] <cpaelzer> smoser: maybe your case can help to change that
[14:48]  * smoser hands cpaelzer the word diglett
[14:48] <cpaelzer> smoser: I saw it on xenial, but only on one machine
[14:48] <cpaelzer> the workaround is easy thou
[14:48] <cpaelzer> ok heading over to diggi
[14:48] <cpaelzer> which is unfortunate as diglett == horsea so it might still be system dependent
[14:48] <jamie_1> sdeziel: thanks a bunch, its working now
[14:49] <jamie_1> was having a hell of an time correcting postfix going back and forth between computers XD
[14:49] <hehehe> smoser: does not work yet
[14:49] <hehehe> with >>
[14:49] <cpaelzer> rharper: can I restart your rh-x1 guest?
[14:50] <smoser> hehehe, i don tknow. i dont use xclip (i use xsel for such things) . but 'some-command >> file' will most definitely write the standard output of 'some-command' to the end of 'file'
[14:51] <nacc> rbasak: cpaelzer: was going to bring that up in standup today
[14:51] <cpaelzer> nacc: the MIR on http2
[14:51] <cpaelzer> ok right place to do so
[14:51] <nacc> cpaelzer: ack, i saw it got approved
[14:53] <hehehe> smoser: so xsel can take text from clipboard and do it?
[14:53] <sdeziel> jamie_1: glad to hear that
[14:53] <cpaelzer> smoser: you have an a1 and an x1
[14:54] <cpaelzer> smoser: unfortunately once the symptom was removed it was as unreproducible as the bug I linked
[14:54] <cpaelzer> hmm :-/
[14:54] <cpaelzer> smoser: the bug has my thoughts so far, but for now you are unblocked
[14:54] <hehehe> installing xsel
[14:57] <hehehe> smoser: xsel -o -sel clip >> webpage.txt paste some garbabe
[14:57] <hehehe> not a text from clipboard
[14:58] <hehehe> it writes own manual to a file
[14:58] <hehehe> :D
[14:59] <rharper> cpaelzer: yes
[14:59] <cpaelzer> thanks rharper
[15:00] <nacc> hehehe: read `man xsel`, -sel isn't an option (at least in 17.10)
[15:04] <hehehe> nacc: yes I read it
[15:05] <hehehe> ok maybe it will work now
[15:07] <hehehe> ty smoser works now
[15:09] <nacc> rbasak: hrm, is insights a staic view? https://insights.ubuntu.com/2017/07/24/developing-ubuntu-using-git/ hasn't updated the top index
[15:10] <rbasak> I suspect it's a static view :-/
[15:10] <rbasak> We might need to push updates to them manually each time.
[15:11] <nacc> rbasak: yeah, I was hoping to linnk to it from the manpage
[15:11] <nacc> rbasak: should i use your actual post instead for now?
[15:17] <rbasak> nacc: sure
[15:50] <smoser> cpaelzer, thanks
[15:56] <cpaelzer> yw
[18:27] <nacc> rbasak: just realized something, our --dl-cache handling probably needs adjusting to handle GIT_DIR (rather than assuming .git) ?
[19:02] <oraqol> hey all, is there a build of conjure-up that doesn't display the following error: lxd not found please install with sudo snap install lxd && lxd init and wait for this message to disappear?
[19:02] <oraqol> on 16.04?
[19:04] <stokachu> oraqol: yea you need to run `/snap/bin/lxd init --auto`
[19:04] <stokachu> having both deb lxd and snap lxd installed with a PATH preference to use deb lxd confuses things
[19:04] <stokachu> i have a build going through QA now that'll make this much clearer
[19:05] <oraqol> ok cool
[19:05] <oraqol> I'll try that now
[19:05] <stokachu> oraqol: best to just `apt remove lxd lxd-client` if you aren't using it
[19:05] <stokachu> stgraber: ^ i thought the /snap/bin prefix was added to the environments PATH for xenial server installs?
[19:08] <oraqol> so its? sudo snap install conjure-up --classic; sudo snap install lxd; sudo lxd init --auto
[19:09] <oraqol> then 'conjure-up' and it should work?
[19:09] <stokachu> oraqol: just /snap/bin/lxd init --auto
[19:09] <stokachu> oraqol: are you reimaging ?
[19:09] <oraqol> the baremetal?  I did last night
[19:09] <oraqol> this is a pretty clean install
[19:09] <stokachu> oraqol: do you already have a lxdbr0 interface?
[19:10] <oraqol> I do not, it's named br0
[19:10] <oraqol> I'll rename the interface
[19:11] <sdeziel> stokachu: sudo has /snap/bin last: secure_path="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/snap/bin"
[19:11] <stokachu> oraqol: that's ok you dont have to as long as br0 doesn't have ipv6 enabled
[19:11] <stokachu> sdeziel: thanks, for some reason i thought it was first
[19:11] <oraqol> will it auto-detect br0 and set it as the bridge?
[19:12] <stokachu> oraqol: if it's a bridge device and doesn't have ipv6 enabled it'll show up in the conjure-up list after you select localhost
[19:13] <oraqol> ok cool
[19:13] <oraqol> lemme try it
[19:13] <oraqol> brb
[19:13] <stokachu> oraqol: make sure to ping me if it doesn't work
[19:13] <stokachu> so we can get it fixed
[19:13] <stokachu> or if you have any other issues
[19:20] <lucidguy> Ok, using Mitaka, is it really necessary to setup ssh-keys between comuter nodes for resizing etc?
[19:55] <oraqol> still getting the following: error: Unable to talk to LXD: Get http://unix.socket/1.0: dial unix /var/snap/lxd/common/lxd/unix.socket: connect: permission denied
[19:56] <oraqol> these are my steps: sudo snap install conjure-up --classic; sudo snap install lxd; /snap/bin/lxd init --auto;
[19:56] <lordcirth_work> oraqol, and which step fails?
[19:57] <lordcirth_work> init, I'd guess?
[19:57] <oraqol> https://pastebin.com/VXcpCY4n
[19:57] <oraqol> network config ^
[19:57] <oraqol> and yes, at lxd init --auto
[19:58] <stokachu> oraqol: whats `ls -l /var/snap/lxd/common/lxd/unix.socket` show
[19:58] <oraqol> srw-rw---- 1 root lxd 0 Sep 19 15:51 /var/snap/lxd/common/lxd/unix.socket
[19:58] <stokachu> oraqol: and what's `id` show?
[19:58] <oraqol> uid=1000(oraqol) gid=1000(oraqol) groups=1000(oraqol),4(adm),24(cdrom),27(sudo),30(dip),46(plugdev),113(lpadmin),128(sambashare)
[19:58] <rbasak> nacc: ah, yes.
[19:59] <stokachu> oraqol: you aren't in the lxd group
[19:59] <oraqol> lxg group
[19:59] <oraqol> yup
[19:59] <stokachu> and your br0 doesn't seem to have any addresses
[20:00] <stokachu> oraqol: you could create lxdbr0 with `/snap/bin/lxc network create lxdbr0 ipv4.address=auto ipv4.nat=true ipv6.address=none ipv6.nat=false`
[20:01] <oraqol> I don't need the VMs to communicate with another physical network, just internal, but I want to be able to communicate with the eno1 netwok
[20:01] <stokachu> ok
[20:07] <oraqol> 'Expected 4 octets in 'None' after attempting to assign br0 as the network bridge in the conjure-up gui.  I imagine the wizard requires br0 to have an IP address in order to continue?
[20:08] <oraqol> possible workaround, attach br0 to eno1 and apply it's ip settings to the bridge?
[20:09] <stokachu> oraqol: yes to both
[20:10] <stokachu> we should have a better error for that though
[20:10] <stokachu> ill file a bug
[20:26] <oraqol> looks like it's working!  Thanks guys!
[20:41] <oraqol> hmm looks like it's stuck on 'waiting for machine' in the deploy display.  Usually doesn't take this long, and it's not utilizing any of my 12 cores
[20:42] <oraqol> 'Running step: 00_deploy-done.
[20:53] <kus_ubuntui686> hi, what ruby on rails applications are available on ubuntu?  I want to install something quick from apt to see if my setup is working properly
[20:56] <sarnold> kus_ubuntui686: using apt-rdepends --reverse ruby-activerecord   gives me the strong impression that gitlab and redmine are the only real 'ruby on rails applications' that are packaged for ubuntu. and they look _HUGE_.
[22:14] <trippeh> oh no, I can't reach the motd service from home ;)
[23:19] <sarnold> trippeh: thanks :)
[23:32] <trippeh> I thought this was no more, but just got it in artful: E: Dynamic MMap ran out of room. Please increase the size of APT::Cache-Start. Current value: 25165824. (man 5 apt.conf)
[23:33] <trippeh> oh my. I'm leaking mem
[23:36] <trippeh> ah ha my ansible fail, it was pushing a big hugepage sysctl config to this small vm
[23:38] <trippeh> <- smrt
[23:39] <sarnold> :)
[23:43] <trippeh> (unfortunately Transparent Huge Pages keeps causing us problems, so using good oldskool pre-allocated hugepages)
[23:44] <trippeh> s/good/awful/