[02:48] anyone about to successfully get phpmyadmin to work with nginx on 14.04 server? [02:49] for some reason the auto config program that runs during the package install doesn't have nginx as an option, and when you go with neither it doesn't seem to generate a config file for it [02:50] so without the config i'm unable to give it my ip address to gain access to it [07:45] I have two nics, both with static IPs. if I set an IP to the first nic, the settings get auto-applied on the second one. why?! using GNOME. [07:49] Razva: Do you mean that both NICs get the same IP? [07:49] or what are "the settings" in this case [07:51] hateball exactly, both nics get the same IP and I have no idea why. [07:51] if I set the LAN IP on the second one, the WAN will be changed. same with WAN -> LAN. [07:52] Razva: Weird. I don't run GNOME so I dunno why that might be [07:52] Razva: if you manipulate network-manager from a terminal, does it behave the same? [07:52] hateball yyyyyyyyup... [07:53] Razva: using nmcli, that is? [07:55] hateball I've just manually set the networking from cli and rebooted, let's see what's happening... [08:16] bah, didn't worked. reinstalling with Lubuntu. I need a browser on this specific server so... [08:19] Razva: network-manager is still the same, regardless what DE you use [08:19] Razva: and there'd be no need to reinstall, you could just apt-get install lubuntu-desktop [09:18] hateball I've done a netinstall and choose Lubuntu, it works like a charm [11:01] Razva: weird. must be a bug in gnomes network manager frontend then [11:01] Razva: I use KDE Plasma and I have no problems using multiple NICs at least [11:22] Hi guy [11:22] It might be stupid question but how change default users group? [11:23] Or make sure all user will be in group X as I use domain login with lib-nss [11:27] Add the default user group to group X. [11:33] LostSoul: "default users group" is a bit ambiguous. [11:34] LostSoul: on *login* a user has a primary group defined in '/etc/passwd' and a list of secondary groups, all those listing the user name in '/etc/group'. They are all "default" groups. [11:34] I see [11:34] I want all AD users to be in certain group [11:34] LostSoul: as to "make sure all user will be in group X" that depends on your use account management scripts. That's an administrative decision. [11:35] Hmm [11:35] I want all users that are able to login via AD to be in certain group [11:35] LostSoul: that may depends on whether 'libnss' allows you to do that. [11:36] Second question is if there is bug in lib-nss as when I add more than X groups in ldap.conf then it starts connecting and getting users [11:38] LostSoul: https://help.ubuntu.com/community/LDAPClientAuthentication might help you [11:38] LostSoul: "Assign local groups to users" section [11:39] Thanks Walex2 [11:40] LostSoul: also https://bugs.launchpad.net/ubuntu/+source/pam/+bug/297408 [11:40] Launchpad bug 297408 in pam (Ubuntu) "pam_group does not support NSS groups" [Undecided,Fix released] [11:44] Hmm [11:44] But still, is it normal than after adding more than 6-8 groups in ldap.conf it stops working? [11:45] Sounds like a lot of groups. [11:48] Hmm [11:53] I've added those lines to /etc/security/group.conf and also added other stuff to pam but seems user is still not in group I wanted him to be ;/ [11:59] coreycb, ddellav: hey - I pushed a load of changes to the neutron package - generaly tidy up and bug fixes [11:59] coreycb, also reverted the agent/server mixup [11:59] coreycb, not sure whether that bug is really relevant any longer... [12:13] LostSoul: traditionally UNIX/Linux systems had a limit of 16 groups, and now most implementations support 256. But with AD could be lower limits. [12:17] hello world [12:23] Walex2: It's not user [12:24] I want to add AD matching (so that certain users can login via ssh on server) [12:24] so when I add more than 6-8 groups in ldap.conf it stops showing users (getent passwd) [12:28] LostSoul: do you get a specific error? [12:36] Nope [12:36] No error at all [12:36] Then how do you determine "it is not working"? [12:36] When I add more than X group mapping getend doesn't return AD users [12:36] Also I can't login :) [12:36] So you get a login error? [12:37] Yes [12:38] So this is wrong?: "< LostSoul> No error at all" [12:39] I mean it doesn't get users from AD when I add more than X mappping in ldap.conf [12:39] So it's not error as user is not searched in AD [12:40] when I add X mapping in ldap.conf and typing getend passwd - I'm getting user list (including AD users) [12:40] When I add X +1, then getent passwd returns only local users [12:40] So I'm wondering if there is a limit [12:43] And did you configure logging for ldap already and investigate the logs? Or even used strace? [12:46] LostSoul: there is then probably a limit, but it would be surprising it is on the Linux side. [12:47] LostSoul: you can use 'ldapsearch' or 'shelldap' to query the MS-AS LDAP server to see what it returns. [12:58] Walex2: assign group works - thanks! [12:58] But still this search .. [12:59] jamespage, thanks for fixing up neutron [12:59] coreycb, np [12:59] just testing the upgrade path is ok and I'll upload [13:37] <^King> hey [13:38] <^King> How to check if a package is installed? [13:39] ^King: dpk -l [13:39] dpkg* [13:39] apt-cache policy [13:39] apt-cache policy will do that job too [14:15] why in the world does autopilot installs Landscape on a LAN IP?! aaargh [14:15] is there any way to "move" it to an external IP? [14:21] <^King> Hey, is it possible to change a account name on a ubuntu server? [14:21] <^King> Similar to changing passwords with passwd [14:23] Walex2: Still around? [14:23] I was wondering if there is way to set up same home dir to all ldap users [14:23] ^King: usermod --login foo bar # renames bar to foo [14:23] ^King: man 8 usermod for details [14:25] my juju is putting landscape to private IP [14:25] how can I access it remotely [14:26] <^King> sdeziel: Will that also change the dir name? [14:27] <^King> /home/newusername/ [14:27] ^King: no but if you pass it --move-home it will [14:27] ^King: you might also want to rename the primary group as well (groupmod) [14:28] <^King> Then I'll just recreate a new username then. [14:28] <^King> Create a new user* [14:29] hi [14:29] can anyone help me on landscape and juju [14:45] @dweaver, do you have exprience on JUJU and MAAS [14:49] Remind me which rc is loaded after ssh login to server? [14:50] jamespage can uou please help my colleague saket? we'r struggling with this here :) [15:01] coreycb, neutron uploaded [15:04] Is there way to force pam other module to set home for all user to certain folder? [15:05] jamespage, awesome, thanks again for the cleanup! apologies for the cruft. [15:06] coreycb, np [15:06] coreycb, apologies for sending your on a red-herring task with switching out plugins for neutron-server [15:06] as there is only ml2 in tree now... [15:06] jamespage, ok gotcha [15:06] coreycb, I have of course now broken neutron-openvswitch and neutron-gateway charms... [15:07] I/we [15:07] jamespage, heh, well that happens [15:07] got to play catch up [15:10] ? :) [15:49] saket_, Yes I do [15:50] saket_, do you need some help with something? [15:53] <^King> I can't install tk8.6, only 5. Why is that so? [15:53] <^King> Only tk8.5 === Guest59764 is now known as karstensrage [17:27] back [17:27] ok soooo can please somebody tell me if/how can we access the Landscape LAN IP from...our public range? [17:28] because Juju/AutoPilot setup a private IP, which obviously cannot be accessed from outside [17:35] ddellav, coreycb: lots of oslo.* releases today :-) [17:35] jamespage, ack thanks. we're working through the clients today. [17:36] Razva, its assumed that you have some access to the network that landscape and the cloud are being deployed to [17:43] jamespage yup, obviously :D [17:43] but tunneling through a proxy via Putty is not a good long-term solution :D [17:43] agreed [17:46] soooo any hints? [17:55] ? [18:21] anyone have a >=80cpu system they could use to run the testcase in bug 1358835 to verify bug 1535494 ? [18:21] bug 1358835 in numactl (Ubuntu Vivid) "numa_node_of_cpu() returns warning when cpu_index > 79" [High,Fix released] https://launchpad.net/bugs/1358835 [18:21] bug 1535494 in numactl (Ubuntu Wily) "Fix numa_node_to_cpus patch" [Undecided,Fix committed] https://launchpad.net/bugs/1535494 [18:22] else i'll just verify that it acts sanely in a smaller vm [18:26] jamespage good thing im doing those today then heh [18:27] ddellav, +100! [18:35] jamespage sorry for bugging you, but can you please point us into the right direction? [18:43] anyone...? [18:46] Razva: port forwarding on your firwall perhaps? [18:47] from where to where...? [18:50] from whatever your public range is to whatever your private range is? :) [18:50] note: I'm using leased servers [18:50] aaaand I don't currently own a switch/router/firewall [18:55] Razva: ah. then your hosting provider can probably lease you IP addresses too? I dunno how that bit works out.. [18:55] Razva: you need to talk to the hosting provider then to get guidance on how to forward ports on the infrastructure [18:55] as well as IPs, etc. [18:56] Razva, sorry - you caught me just as I had to head out for a bit [18:57] Razva, can you explain how your servers are networked a bit (this may be a bit async as have to duck out again in a few minutes) [18:58] jamespage: he may not know... [18:58] [2016-02-22 13:50:27] note: I'm using leased servers [18:58] [2016-02-22 13:50:40] aaaand I don't currently own a switch/router/firewall [18:59] might be tricky to figure this out then... [18:59] ^ that's why I suggested they reach out to the ones they lease from for some guidance on forwarding ports, etc. that way [18:59] (it may be something that has to be done by the provider) [18:59] (though that'd be highly irregular) [19:23] jamespage I have my own IP class + my own VLAN [19:24] each server has one nic connected to the public IP class + another nic connected to the vlan [19:24] at this point it seems that Landscape is installed into a LXE container on one of the machines (not on the maas machine), which I suppose it's ok [19:25] but now the problem is that...there's no real way to access Landscape from "outside", except doing a proxy/tunnel [19:31] aaaand problem #2 [19:32] seems that Autopilot doesn't "wants" to setup a 5 node OpenStack cluster. everything is "green" but Cinder fails to deploy. saket_ is digging in the logs but he cannot find anything relevant. any hints? === alexisb is now known as alexisb-afk === AndyTechGuy is now known as _atg [20:20] [ERROR] Failed to upload leases: 'str' object has no attribute 'mac' [20:20] oh I really love this. [20:20] no documentation in this kind of issues, and minimal logs. === uaa is now known as Guest61494 === Guest61494 is now known as damascene === _atg is now known as AndyTechGuy [21:00] ddellav, I dropped python-doc8 from ironicclient as it's not needed for package builds [21:01] coreycb ack === damascene is now known as ahmed_sabir === ahmed_sabir is now known as damascene [21:12] zul, jamespage, can you subscribe a team to python-requestsexceptions? === alexisb-afk is now known as alexisb [21:22] Hello. I'm currently in the process of installing ubuntu server (15.10) on a machine that was a gentoo machine up until now. The machine has a raid6(dm_crypt(lvm)) disk layout. I'm not familiar with configuring raid/dm_crypt/lvm in the ubuntu installer and I'm afraid to play around because I can't afford to lose data on the disk. [21:23] I'm in manual disk setup, and it recognizes the raid6 array but marks it as 'auto-read-only'. I'm unsure if using the 'Configure software raid' option will allow me to access the raid array read/write and if I can do so without losing data [21:24] Could anyone provide some input on that please? [21:27] cluemann: IIRC, there should be no problem adopting your raid array and your dm-crypt setup [21:32] I was indeed able to activate my md_crypt container and the logical volumes are now available. [21:32] dm_crypt* [21:38] cluemann: it's usually auto-read-only until something tries to write to it [21:39] Oh okay thank you RoyK [22:01] ddellav, for heatclient I think you just need to swap 3.4 with PYTHON3S in d/rules override_dh_auto_test [22:01] coreycb ok i'll try that [22:15] I have vsftpd set up for logging in as a user and r/w files [22:16] Is there a way to point the anonymous login to that user for read only access? [22:16] My installation is done and all my data is intact, thanks for the input! :) [22:22] ogra_: there still? I have a question for you if you do not mind [22:27] PCI compliance is failing on a server running Ubuntu Linux 14.04.2, specifically for CVE-2015-3185 and CVE-2015-3183. How can I verify these vulnerabilities are present and resolve them? [22:28] wimpog: according to http://people.canonical.com/~ubuntu-security/cve/2015/CVE-2015-3185.html, all you need to do is pull apache2 version 2.4.7-1ubuntu4.5 or higher [22:29] sdeziel: same thing for CVE-2015-3183? [22:30] yes [22:30] Thank you! [22:30] you are welcome [22:30] sdeziel: i just ran dpkg -l | grep apache [22:30] and I have 2.4.7-1ubuntu4.9 [22:31] or am I not checking it correctly? [22:32] wimpog: maybe the PCI checking engine is not aware that Ubuntu backported the fix from 2.4.16 upstream to Trusty's 2.4.7 [22:33] sdeziel: is dpkg -l the right way to verify the version? [22:34] wimpog: yes [22:35] Thank you again! There is nothing else I could do here, isn't it? [22:36] sdeziel: [22:41] wimpog: the apache2 service is restarted on package upgrades so it shouldn't use a binary older than the one provided by 2.4.7-1ubuntu4.9. If you want to be absolutely sure you are running the patched version you can use this: for p in $(pgrep apache2); do ls -l /proc/$p/exe; done | grep deleted [22:41] wimpog: if nothing is output you are all good (running the patched binary) [22:42] sdeziel: thank you for this script. I get no output [22:43] wimpog: good. I have to go, have a good day|night [22:43] Have a good day [23:00] <{soma_coma}> Woke back up, how is everyone? === LewisCawte is now known as Lcawte|Away