=== JanC_ is now known as JanC
[07:18] <cpaelzer> rbasak: could you look into tagging and merging, as well as uploading https://code.launchpad.net/~paelzer/ubuntu/+source/multipath-tools/+git/multipath-tools/+merge/311921 today ?
[07:19] <cpaelzer> rbasak: I updated the merge request accordingly, but had to realize today this is not ubuntu-server-dev upload permission yet
[09:43] <cpaelzer> rbasak: have you seen my request to tag, merge and upload multipath?
[09:43] <cpaelzer> rbasak: or is the change to the upload permissions we discussed in the other chan close enough that I should upload then?
[09:46] <rbasak> cpaelzer: yes. I'm running an update of the server packageset now. It involves running germinate which takes a while. I just started it. Separately I have a note to gather all upload tags, which is a bit lower priority. I've added multipath-tools to that list.
[09:46] <cpaelzer> rbasak: ok thank you
[09:47] <cpaelzer> rbasak: can you ping me once the packageset update is complete so I can copy this one from bileto?
[09:47] <rbasak> cpaelzer: will do!
[09:51] <vamadir> hello
[09:55] <vamadir> i have a question, about ubuntu server`s. How to correct create users and groups for services? Like httpd, ftp, postfix...etc. Or its beter start them from root?
[09:56] <hateball> vamadir: usually service-users are created for you when you install the packages
[09:56] <hateball> vamadir: for instance installing apache will create www-data
[09:58] <vamadir> hateball: have some rules or tips on ubuntu about creating service-users?
[09:58] <ogra_> vamadir, normally you do not create them yopurself, the packages do this from the install scripts
[10:00] <vamadir> ogra_: i understand, but if i create my own service, i need to create user for this?
[10:00] <ogra_> if you dont want to run it as root, yes ...
[10:00] <ogra_> i'd look at the source of a package with a similar service in such a case
[10:01] <ogra_> specificaloly at the postinst script in the debian dir
[10:03] <vamadir> ogra_: thanks
=== disposable3 is now known as disposable2
[10:31] <Stry> rbasak: ubuntu-mysql apparmor logs: http://paste.ubuntu.com/23898693/
[10:37] <rbasak> Hmm, he's gone.
[10:37] <rbasak> For the logs: /var/log/mysql/error.log being empty isn't good. A conffile modification predating Precise (maybe Trusty?) can cause that IIRC. I wonder if he also has a tweaked AppArmor configuration?
[12:00] <panicstr> Hello All.
[12:00] <panicstr> I'm doing a release upgrade on ubuntu server. It currently shows "Restarting services possibly affected by the upgrade:" for over half an hour... has something gone wrong?
[12:09] <fredrin> I have a server with two NICs, one for uplink (eno1) and one for internal network (eno2). eno2 has been used to create a bridge interface, br0, that I uses for virtual-guests. When I enable ufw, the virtual guest are unable to send arp requests to the dhcp-server that runs on the host, but other clients that is connected to the bridge by eno2 physical and a switch are able to get IPs. When I
[12:09] <fredrin> disable ufw things work.
[12:10] <fredrin> I've added the lines that is under the NOTE section of ufw-framwork manual
[12:11] <fredrin> this is on 16.04
[12:12] <fredrin> any pointers how to debug further?
[13:30] <fredrin> ok, ufw allow bootps helped
[13:30] <fredrin> ...
[13:30]  * fredrin pats himself on the back 
[13:38] <zioproto> coreycb, jamespage when we run the Openstack network node on ubuntu, it would be nice to have the Kernel compiled with CONFIG_NET_DROP_MONITOR=y
[13:38] <zioproto> how does it work to give feedback about the Ubuntu Kernel ?
[13:43] <coreycb> zioproto, hi, that might make sense. the folks in #ubuntu-kernel would know better.
[13:43] <zioproto> thanks
[13:44] <coreycb> zioproto, thank you :)
[13:54] <zioproto> done, looks like this change will go in zesty
[13:54] <zioproto> openstack folks, might want to read this : http://www.9bitwizard.eu/packets-drop-by-drop
[13:54] <zioproto> looks like a noisy network can cause packet loss on your network node
[13:54] <zioproto> hard to debug at the moment on a ubuntu server
[13:55] <zioproto> ;)
[14:11] <zul> coreycb: where do you fiddle with the cloud archive patches again?
[14:12] <coreycb> zul, https://git.launchpad.net/~ubuntu-cloud-archive/ubuntu/+source/ca-patches/
[14:12] <zul> coreycb: thanks
[14:18] <panicstr> do-release-upgrade failed resulting in broken packages, apt not working etc... can i reboot?
[14:19] <zioproto> panicstr, why reboot ? It does not sound like a good plan
[14:20] <zioproto> panicstr, what you mean by apt not working ?
[14:20] <zul> jamespage/coreycb: nova b3 is still in dep-wait because of python-os-xenapi is still in source new
[14:21] <panicstr> zioproto Errors were encountered while processing: -list some 30 packages-
[14:22] <zioproto> panicstr, did you try to 'apt-get -f install'
[14:22] <zioproto> to finish the installation if those packages ?
[14:22] <zioproto> s/if/of
[14:22] <panicstr> E: Error, pkgProblemResolver::Resolve generated breaks, this may be caused by held packages.
[14:22] <panicstr> E: Unable to correct dependencies
[14:23] <zioproto> did you pin some packages ?
[14:23] <panicstr> pin packages?
[14:23] <zioproto> did you force some package to be in a specific version ?
[14:23] <panicstr> not that i recall
[14:23] <zioproto> what is the name of the package that fails to upgrade ?
[14:24] <panicstr> there's some 30 of them
[14:25] <zioproto> what upgrade are you doing ? trusty to xenial ?
[14:26] <panicstr> yes
[14:26] <zioproto> panicstr, read this: https://blog.bramp.net/post/2012/04/29/failed-ubuntu-update/ is this helpful ?
[14:27] <panicstr> this log is empty in my system
[14:27] <zioproto> /var/log/dist-upgrade/apt.log or /var/log/apt/term.log ?
[14:29] <panicstr> zioproto don't bother, got to go, thanks for your help
[14:29] <zioproto> no problem
[14:42] <coreycb> zioproto, cool thanks for pushing on that and thanks for sharing that post
[15:18] <jarad_> problem booting new install of daily build on lpar - where to find help ?
[15:20] <jarad_> "/vdevice/v-scsi@30000009/disk@8100000000000000:-1,/etc/yaboot.conf: Unknown or corrupt filesystem" - which is not true
[15:26] <coreycb> zul, i'm backporting python-wsgi-intercept python-xmltodict
[15:26] <zul> coreycb: k
[15:27] <zul> coreycb: im fixing libvirt in the CA
[15:27] <coreycb> zul, thanks
[15:27] <coreycb> zul, i think the other backport issues may be fixed by updating d/control
[15:42] <jarad__> i have a yaboot problem - tying to install powerpc64 on lpar - when booting i get
[15:43] <jarad__> "/vdevice/v-scsi@30000009/disk@8100000000000000:-1,/etc/yaboot.conf: Unknown or corrupt filesystem" - which is not true
[15:43] <jarad__> daily build that is ...
[15:44] <jarad__> any suggestions
[15:49] <zul> coreycb: the libvirt patch we carry for the ocata UCA is no longer needed. fyi
[15:50] <powersj> jarad__: take a look at LP: #1606089
[15:50] <ubottu> Launchpad bug 1606089 in yaboot (Ubuntu) "unable to boot after 'entire disk' install (16.10, ppc)" [Undecided,Confirmed] https://launchpad.net/bugs/1606089
[15:59] <jarad__> sorry what does LP: #160689 mean ?
[15:59] <ubottu> Launchpad bug 160689 in Moblin Multimedia "Suggest to add a column to display media path" [Low,Won't fix] https://launchpad.net/bugs/160689
[16:00] <coreycb> zul, ok good
[16:00] <powersj> jarad__: that issue and workaround are described in the launchpad bug here: https://launchpad.net/bugs/1606089
[16:00] <ubottu> Launchpad bug 1606089 in yaboot (Ubuntu) "unable to boot after 'entire disk' install (16.10, ppc)" [Undecided,Confirmed]
[16:07] <jarad__> thx :)
[16:10] <jamespage> zioproto, coreycb: that sounds like a useful kernel feature - thanks for getting that enabled for the zesty kernel!
[16:11] <zioproto> jamespage, it will come also in Xenial
[16:11] <jamespage> zioproto, great
[16:11] <zioproto> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1660634
[16:11] <ubottu> Launchpad bug 1660634 in linux (Ubuntu Yakkety) "Enable CONFIG_NET_DROP_MONITOR=m in Ubuntu Kernel" [Undecided,In progress]
[16:12] <dasjoe> Ugh, "Root privileges are required for running gparted." - I'm a member of disk, so I should be able to run gparted as my user
[16:25] <jarad__> powersj, ubottu: confirmed was able to boot after choosing ext2 as /boot and / - thx agin
[16:25] <jarad__> powersj, ubottu: confirmed was able to boot after choosing ext2 as /boot and / - thx again
[16:26] <powersj> jarad__: good to hear!
[16:49] <nacc> dasjoe: is that perhaps hardcoded in gparted?
[17:01] <rbasak> cpaelzer: PEBKAC. I got the exception for multipath-tools working. Not quite active yet.
[17:01] <rbasak> cpaelzer: OK, you should be able to upload multipath-tools now.
[17:05] <hrw> hi guys
[17:06] <dasjoe> nacc: it seems to be, parted just sent a generic warning but let me edit the GPT
[17:06] <nacc> dasjoe: interesting, probably worth asking upstream, but it might be a safety check in their code :/
[17:07] <hrw> who I should speak with to make ubuntu-lts/arm64 image for docker.io?
[17:08] <hrw> I work on adding arm64 support to openstack-kolla and need some kind of official ubuntu/arm64 image for containers
[17:34] <blizzow> I'm having difficulty installing python-boto.  apt keeps complaining that it python-boto needs python-requests and that relies on python-urllib3 1.13. Then it complains that 1.18 is set to be installed. Anyone seen this?
[17:35] <rbasak> blizzow: can you pastebin your output?
[17:39] <blizzow> The way I got around the problem was to use aptitude and it proposed downgrading, but that doesn't really work with ansible.  Hold on a minute and I'll fire up a VM to get the output.
[17:41] <nacc> blizzow: which version of ubuntu?
[17:42] <blizzow> 16.04
[17:44] <nacc> blizzow: fresh lxd 16.04 container, python-boto 2.38.0-1ubuntu1 installed fine
[17:44] <nacc> along with python-requests 2.9.1-3, ptyhon-urllib3 1.13.1-2ubuntu0.16.04.1
[17:45] <nacc> blizzow: python-urllib3 1.18 is not published in any ubuntu
[17:45] <nacc> blizzow: currently, so perhaps a PPA in use?
[17:53] <blizzow> http://pastebin.ca/3762394
[17:54] <blizzow> Ah, elasticsearch PPA.
[17:55] <blizzow> How do I pin the python-urllib3 back to the ubuntu version?
[17:55] <nacc> blizzow: i think instead you would say only to install certain packages from the elasticsearch PPA? although perhaps that would break their software then
[18:00] <zul> coreycb: lovely....just lovely
[18:09] <blizzow> nacc, that's a fine solution, I guess how would I whitelist only certain packages from their PPA?
[18:12] <nacc> blizzow: iirc, `man apt_preferences`, with something like specific Package entries from that PPA
[18:13] <nacc> blizzow: but i'm not sure if that will let elatsicsearch work, if it in turn needs the 1.18 version
[18:14] <blizzow> yeah, urgh. This machine won't even be using elasticsearch. I guess I can't use a monolithic apt sources anymore :/
[19:42] <hallyn> soren: hey
[20:09] <blueking> is it possibly to ask firewall rules here ?
[20:09] <sarnold> sure
[20:10] <blueking> what I am unsure about are what ports one should open  lan -> wan and lan -> local.. how to know ?
[20:10] <blueking> wan -> anything usually blocked
[20:22] <blueking> sarnold   u forgot me ? ;)
[20:23] <dasjoe> nacc: apparently Gnome doesn't think of the hard requirement to be root as a bug, see https://bugzilla.gnome.org/show_bug.cgi?id=641758 and https://bugzilla.gnome.org/show_bug.cgi?id=693275
[20:23] <ubottu> Gnome bug 641758 in application "Cannot run as non-root" [Major,Resolved: duplicate]
[20:23] <ubottu> Gnome bug 693275 in application "gparted fails to start because it says it needs to be root" [Major,Resolved: notabug]
[20:23] <nacc> blueking: well, to be fair, sarnold only said 'sure', didn't say they would answer your question
[20:23] <dasjoe> They argue a user can't inform the kernel about partition table changes, so running as root is required
[20:24] <nacc> dasjoe: interesting
[20:25] <dasjoe> nacc: actually re-reading the tables as a user results in "BLKRRPART failed: Permission denied" for me, so they may have a point there :)
[20:25] <blueking> nacc  I know, just  wanted to highlight his nick
[20:26] <nacc> dasjoe: :)
[20:36] <sarnold> blueking: it depends entirely on what you want your system to -do- :)
[20:46] <blueking> sarnold sarnold  4 or 5 subnets  lan-main (free access to net)  lan-kids (time limited when kids can have access to net)  lan-fileserver (less secure subnet, fileserver has its own firewall, 4 nics on fileserver one for wan (through router), and 1-3 nics LAG lacp   vlan-trunking (vlan tagged for lan-main and lan-kids), not decided yet how to configure this.  and guest zone
[20:47] <sarnold> blueking: wow, nice
[21:08] <dasjoe> Is "sarnold sarnold" like the B1s' "roger roger"?
[21:10] <sarnold> yes yes :)
=== dzragon^ is now known as dzragon