=== jgrimm is now known as jgrimm-out
=== seg_ is now known as seg
=== berglh_ is now known as berglh
=== sikun is now known as sikun_
=== sikun_ is now known as work_sikun
[06:17] <dreki> While setting up sieve on my email server I mistakenly setup a symlink  (~/sieve/myfilter.svbin > ~/.dovecot.sieve) but I think that that should have been  (~/sieve/myfilter.sieve > ~/.dovecot.sieve). I have fixed the mistake but dovecot/sieve still throws an error saying that the symlink still links to the wrong file. I have made sure the symlink is fixed and I tried restarting dovecot and the server but nothing works. I still get the error and my sieve f
[06:17] <dreki> If anyone has any ideas about what I could be doing wrong I would greatly appreciate any help.
=== amoralej|off is now known as amoralej
=== _degorenko|afk is now known as degorenko
[10:47] <jamespage> coreycb, zul: https://bugs.launchpad.net/ubuntu/+source/nova/+bug/1644187
[10:47] <ubottu> Launchpad bug 1644187 in nova (Ubuntu) "ValueError on creating new nova instance" [Undecided,Confirmed]
[10:49] <tobasco> jamespage: ping, uca update breaking spawning new instances for me on 14.04 uca liberty https://bugs.launchpad.net/nova/+bug/1644187
[10:49] <ubottu> Launchpad bug 1644187 in nova (Ubuntu) "ValueError on creating new nova instance" [Undecided,Confirmed]
[10:50] <jamespage> tobasco, I just raised that with zul and coreycb prior to you joining channel
[10:51] <jamespage> tobasco, q do your compute and controller units have the same versions of the package?
[10:51] <tobasco> everything except nova-compute actually, doing them one at a time
[10:52] <tobasco> that's the issue? if so i can start pushing it out to all of them directly
[10:54] <jamespage> tobasco, I suspect some sort of break on the RPC compat looking at the stacktrace but I might be wrong
[10:54] <jamespage> zul, hey can we make sure that for ocata package updates we're using 0ubuntu1 's please
[10:55] <jamespage> I've seen  a few 1ubuntu1's
[10:56] <jamespage> for ubuntu first uploads that is
[11:01] <tobasco> jamespage: confirmed only happening on the compute not running the latest nova packages
[11:01] <jamespage> tobasco, ok please comment on the bug to that effect
[11:02] <jamespage> tobasco, the versioned-objects layer should make some guarantees about mixed version running esp within a liberty sru
[11:02] <jamespage> but
[11:02] <jamespage> it has been know to break before..
[11:02] <jamespage> lets get an upstream nova bug task raised as well
[11:02] <jamespage> tobasco, can you detail from->to versions as well please
[11:02] <tobasco> jamespage: i tagged upstream project in bug
[11:02] <jamespage> tobasco, \o/
[11:02] <jamespage> tobasco, so you did - my bad
[11:04] <tobasco> jamespage: Upgrade: nova-compute-libvirt:amd64 (12.0.4-0ubuntu1~cloud1, 12.0.5-0ubuntu1~cloud0)
[11:04] <tobasco> commenting on bug
[11:04] <jamespage> tobasco, gosh and just a minor point upgrade as well
[11:05] <tobasco> jamespage: yea, we try to keep inline with the releases of new packages, usually we are some more people but i had to handle the stuff myself so some compute nodes were left for this morning
[11:06] <jamespage> tobasco, well at least there is a workaround to get you moving again
[11:06] <tobasco> good it got resolved though, thanks a lot, need to verify our workflow on upgrades now, we didn't test the ability to run older compute vs newer controller packages
[11:10] <jamespage> tobasco, ftr we always do a complete upgrade prior to re-commissioing the cloud during SRU verification
[11:10] <jamespage> so we'd not normally pickup a mixed mode issue like this
[11:15] <tobasco> jamespage: this is interesting thought, our product team is still seeing issues, this might actually be a faulty API request to nova causing the issue
[11:16] <tobasco> and perhaps nothing to do with nova, must look further into this...
[11:16] <jamespage> tobasco, ok here for another 7 hrs
[11:16] <jamespage> light on US folks around today due to thanksgiving this week
[11:43] <hhee> guys, ubuntu server 16 lts contains php 7.0.8 in default repo, currtnct version 7.0.13. which way i need to use to use latest version?
[11:47] <tobasco> jamespage: false alarm, was apparently something wrong on our side
[11:47] <tobasco> jamespage: atleast i got our compute nodes upgrades, sorry for the hassle :(
[12:17] <SipriusPT> hello guys
[12:18] <SipriusPT> I am not able to connect to ports with TLS/SSL like 587 and 993 who are open (i have test with telnet)
[12:19] <SipriusPT> I am only able to use ports 25 and 143
[12:19] <SipriusPT> when i try to send mail with 587, i am always receiving this error
[12:19] <SipriusPT> 5.7.0 Must issue a STARTTLS command first
[12:20] <SipriusPT> and i have already mess with smtpd_tls_security to encrypt and may from master.cf
[12:20] <SipriusPT> also i didnt have submission commented
[12:20] <SipriusPT> those was the solutions that i have found
[12:20] <SipriusPT> but didnt work
[12:20] <SipriusPT> anyone knows what can be?
=== Piper-Off is now known as Monthrect
[12:36] <zul> jamespage: yeah that was me not use to the tooling that we use, its been corrected on newer versions
[13:17] <binia> SipriusPT, not sure, what php version, i remember having problems with php5.6
=== amoralej is now known as amoralej|lunch
[13:27] <linuxperia> Hi all. I have a huge problem with MariaDB MySQL Server. Today i tryed to get multi mysql instances running on my ubuntu server with one galera instance that listen to 0.0.0.0 and a the existing one mysql instance listening like always to localhost inspired by this howto here => https://www.sebastien-han.fr/blog/2012/08/06/multi-mysql-instances-with-galera/
[13:27] <linuxperia> After the starting of mysql using the changed configuration all my databases and user acess information has gone lost in the existing /var/lib/mysql directory where all the existing data was stored!
[13:27] <linuxperia> How can i recover my existing mysql server data back! I see there some innodb data files and innodb log files that hopefully can be used. please help! thanks for any helpfull tips and tricks!
[13:55] <zul> coreycb: btw keystone needs to be python-distutils in order to get the migrate.cfg stuff to work properly now
[14:01] <coreycb> zul, export PYBUILD_SYSTEM=distutils  ?
[14:02] <zul> coreycb: dh $@ --buildsystem=python_distutils --with python2,sphinxdoc,systemd
[14:02] <coreycb> zul, ok
[14:02] <coreycb> zul, did that fix it up?
[14:02] <zul> coreycb: going to run it locally and rebuild the ci version and re-deploy
[14:02] <coreycb> zul, ok
[14:02] <zul> coreycb: no keystone...no worky ;)
[14:02] <coreycb> zul, ya
[14:18] <zul> coreycb: shit that didnt do it either
[14:19] <coreycb> zul, did you talk to anyone upstream about it?
[14:20] <zul> coreycb: yeah they thought it was fine
[14:21] <coreycb> zul, are the missing files in the release tarball?
[14:21] <zul> coreycb: nope they are there
[14:21] <coreycb> zul, no they're not?
[14:21] <zul> coreycb: the migrate.cfg are in the tarball
=== amoralej|lunch is now known as amoralej
[14:27] <coreycb> zul, maybe jamespage knows a better way.  we could manually  install the files in d/rules but there should be a better way.
[14:28] <zul> coreycb: yeah im trying to figure out the better way
[14:28] <zul> coreycb: im not giving up yet
[14:28] <hhee> guys, how can i install latest php in ubuntu server? by default there is 7.0.8
[14:29] <hhee> ubuntu server 16.04 lts
[14:33] <SipriusPT> binia: i am using PHP Version 5.6.25
[14:34] <binia> yeah, i gave up :D
[14:34] <SipriusPT> did you solve it?
[14:34] <binia> nope, whatever i did, it fucking didnt work, went to 5.5 worked no probs
[14:35] <SipriusPT> i see
[14:35] <SipriusPT> thanks binia
[14:35] <binia> no probs
[15:05] <relativemedia> morning folks
[15:05] <cpaelzer> where are those morning folks?
[15:05] <cpaelzer> hi relativemedia
[15:05] <relativemedia> im on 16.04.01 x64 and the following packages need to be upgraded ... linux-headers-generic linux-headers-virtual linux-image-virtual linux-virtual open-iscsi
[15:06] <relativemedia> last time i did dist-upgrade from 14.04 to 16.04.01 i had to nuke the vm and rebuild
[15:06] <relativemedia> shouldn't have any issues this time right ?
[15:06] <relativemedia> since its a minor upgrade
[15:08] <cpaelzer> relativemedia: that is only kernel updates and some ipv6 fixups in open-iscsi
[15:08] <cpaelzer> relativemedia: seems fairly safe if that is your only upgrade list
[15:08] <relativemedia> yeah thats wthat i figured
[15:09] <cpaelzer> relativemedia: but then in general since you are on a VM depending on the risk you are taking maybe snapshots are good?
[15:09] <relativemedia> takign one now :)
[15:09] <cpaelzer> relativemedia: well then you are safe++
[15:10] <relativemedia> yes indeedy
[15:10] <cpaelzer> As always, the devil is in the details ...
[15:10] <relativemedia> yeah
[15:10] <relativemedia> do snapshots take forever though :(
=== JanC_ is now known as JanC
[15:20] <relativemedia> cpaelzer, all good successful upgrade :) sweet
[15:23] <zul> coreycb: yeah manifest.in,,,stupid pbr....grumble grumble
[15:23] <coreycb> zul, ?
[15:24] <zul> coreycb:  pbr generates its own manifest with "sensible" defaults with the AUTHOR and changelog automatically
[15:24] <coreycb> zul, gotcha.  so this is a pbr bug?
[15:24] <zul> coreycb: dh_python expects a manifest to be there,
[15:25] <zul> coreycb: yeah
[15:25] <coreycb> zul, ok so i'd say open a bug against pbr/keystone and we can install those files manually for now with a comment linked to the bug in d/rules.  what do you think?
[15:26] <zul> coreycb: i have a patch that re-adds the manfiest
[15:27] <coreycb> zul, that's probably better.  can you add a link to the bug to that patch?
[15:27] <zul> coreycb: ill open a bug and add it to the link
[15:27] <coreycb> zul, thanks
=== BioKee is now known as Guest44821
[15:36] <zul> coreycb: i *think* antoher option would be patch our version of pbr
[15:37] <zul> coreycb: lemme try something
[15:37] <zul> coreycb: im just worried that other things like nova is affected the same way
[15:38] <coreycb> zul, did any other projects drop their MANIFEST?
[15:38] <coreycb> zul, if you patch pbr get the patch upstream too
[15:39] <zul> coreycb: they all dropped the MANIFEST
[15:39] <relativemedia> anyone here play with kubernetes
[15:40] <coreycb> zul, well not all of them did, but yeah a fix to pbr would be good because they might
[15:41] <zul> coreycb: relevant bug https://bugs.launchpad.net/magnum/+bug/1608980
[15:41] <ubottu> Launchpad bug 1608980 in OpenStack Object Storage (swift) "Remove MANIFEST.in as it is not explicitly needed by PBR" [Undecided,In progress]
[15:50] <relativemedia> so i have a q, can i install kubernetes on an ubuntu 16.04 server and havew it control coreos clusters?
[16:03] <coreycb> jamespage, beisner, hi can you promote libvirt - 1.2.12-0ubuntu14.4~cloud2 from kilo-staging -> kilo-proposed please?
[16:12] <Slashman> hello, I have an issue with zfs on ubuntu 16.04 kernel 4.4.0-47 and latest zfs: http://apaste.info/Iqogj result: any zfs dataset are unusable, any read/write operation takes forever...
[16:13] <Slashman> I'm looking for bug report about this, didn't find one, any idea?
[16:14] <Slashman> other log: http://apaste.info/DCX0E
[16:23] <zul> coreycb: yaaay http://pastebin.ubuntu.com/23522795/
[16:29] <coreycb> zul, \o/  pbr patch?
[16:30] <zul> coreycb: yeah i just uploaded it to the archive..next is to rebuild in the CI
[16:30] <zul> coreycb: then deploy again
[16:30] <zul> coreycb: then rebuild packages and upload
[16:31] <stgraber> Slashman: #ubuntu-kernel may be more likely to help you
=== Monthrect is now known as Piper-Off
[16:33] <jonah> Does any body here use APC smart-ups with a generator? My generator is only 2200 watts. But my ups is 2700watt max output. Yet my load is under 2000w. I spoke to APC and they said it's best not to risk testing my generator with my UPS as generators should be 3-5 times higher rated in VA/watts than the load. So I'm not sure if my setup will work and worred to test it as they've said not to!!
[16:37] <rbasak> jonah: when your generator comes online, it will need to supply both your 2 kW load as well as your recharging UPS. Depending on how fast your UPS recharges, it sounds like it will be overloaded.
[16:37] <rbasak> Presumably that's why APC have that guideline?
[16:39] <rbasak> I don't have experience of running a UPS with a generator, but I do have experience of a UPS-only scenario where the main circuit is only rated slightly higher than the load.
[16:40] <rbasak> It constantly tripped after a power failure because the UPS added to the load. We had to drop the load down until it charged every power outage. Manually, before the UPS ran out.
[16:40] <jonah> rbasak: so it may work if the load is reduced enough to enable recharging?
[16:40] <rbasak> That would mitigate my particular point, but I don't have enough experience to be tell to tell you that there won't be some other problem.
[16:41] <beisner> coreycb, libvirt 1.2.12-0ubuntu14.4~cloud2 promoted to kilo-proposed re: https://bugs.launchpad.net/ubuntu/+source/libvirt/+bug/1640676
[16:41] <ubottu> Launchpad bug 1640676 in libvirt (Ubuntu Trusty) "[SRU] libvirt 1.2.12 live-migration corrupts some instances" [High,Incomplete]
[16:41] <rbasak> Power factor, for example - you stated your load in kW, not VA.
[16:41] <coreycb> beisner, thanks
[16:42] <beisner> yw coreycb
[16:52] <zul> coreycb: im killing the pending ci jobs since those are going to be invalid if the manifest is not there
[16:52] <coreycb> zul, ok
[16:53] <zul> coreycb: might as well fix CI issues as well :)
=== degorenko is now known as _degorenko|afk
=== iberezovskiy is now known as iberezovskiy|off
=== amoralej is now known as amoralej|off
[18:03] <Someone_Else> I can't browse a mounter NFS share. First, ls listst 0 files, second try: "no such file or directory". df shows the mounted volume at first, however, changes to "stale file handle". ls list the directory permissions as "d?????????", NFS shares on other hosts work fine (using the same server)
[18:03] <Someone_Else> Using Xenial
=== Piper-Off is now known as Monthrect
[18:11] <relativemedia> hey all im on 16.04.01 and i notice when i run apt-get update i get a lot of things like The repository 'https://packages.gitlab.com/gitlab/gitlab-ce/ubuntu xenial Release' does not have a Release file.
[18:12] <tarpman> relativemedia: pastebin your sources.list (and any sources.list.d fragments) and the output of apt-get update
[18:12] <relativemedia> sources.list(.d) is managed by puppet so .list is empty
[18:12] <relativemedia> ill pastbind .d
[18:14] <relativemedia> https://gist.github.com/mikedevita/7f72580b1075dccc1f66bacfc4863746
[18:14] <relativemedia> weverything is a local mirror except gitlab]
[18:16] <tarpman> https://packages.gitlab.com/gitlab/gitlab-ce/ubuntu/ is returning 404 for me.
[18:16] <relativemedia> same just noticed that
[18:16] <relativemedia> hah
[18:17] <tarpman> oh, they have some weird setup
[18:17] <tarpman> directory listings don't work. fine
[18:17] <relativemedia> yeah
[18:21] <tarpman> relativemedia: can you do 'apt-get -o Debug::Acquire::http=1 -o Debug::Acquire::https=1 update' and pastebin that?
[18:21] <tarpman> at a glance, https://packages.gitlab.com/gitlab/gitlab-ce/ubuntu/dists/xenial/Release and related files look ok
[18:22] <relativemedia> it might be because of ssl :\
[18:22] <relativemedia> iev got a sneaky suspicion
[18:25] <Someone_Else> relativemedia: Is apt-transport-https installed?
[18:25] <relativemedia> yes it is
[18:26] <relativemedia> but ihave issues clone git repos on other urls
[18:26] <relativemedia> and curl fails on their gitlab url
[18:28] <Someone_Else> The issue seems to be GitLab related
[18:29] <Someone_Else> Can you comment the lines out in the GItLab apt-file and run gitlab-ctl reconfigure?
=== Monthrect is now known as Piper-Off
=== Piper-Off is now known as Monthrect
=== Monthrect is now known as Piper-Off
[20:37] <Fieldy> hello, is anyone else having problems installing things? for over an hour now i've been having this happen, with the various us.archive.ubuntu.com servers: 0% [Connecting to us.archive.ubuntu.com (91.189.91.15)]
[20:37] <Fieldy> just never gets anywhere
[20:37] <Fieldy> each time is a different ip naturally in the IP rotation
[20:39] <sarnold> Fieldy: hrm, both IPv4 addresses I get back appear to work fine
[20:40] <sarnold> funny enough I don't get 91.189.91.15 as one ofmy choices -- I get 91.189.91.26 and 91.189.91.23
[21:03] <relativemedia> hey all something funky with my ca certs... fatal: unable to access 'https://github.com/creationix/nvm.git/': Problem with the SSL CA cert (path? access rights?)
[21:03] <relativemedia> i ran update-ca-certificates and it looks ok
[21:04] <tarpman> relativemedia: need more details. try openssl s_client -connect github.com:443 -showcerts
[21:04] <relativemedia> i did that, came back rc0 ok
[21:05] <relativemedia> is there a way to reconfigure ca-certificates completely, like nuke /etc/ssl or whatever and redo it
[21:08] <tarpman> relativemedia: export GIT_CURL_VERBOSE=1 and try your clone again
[21:09] <relativemedia> * error reading ca cert file /etc/ssl/certs/ca-certificates.crt (ASN1 parser: Error in TAG.)
[21:09] <relativemedia> lol wtf
[21:09] <tarpman> this sounds vaguely familiar. didn't you have a broken/invalid certificate somewhere in that directory before?
[21:10] <relativemedia> a awhile back
[21:10] <relativemedia> deleted it though
[21:11] <relativemedia> is there a way to nuke /etc/ssl/certs and regenerate it completely?
[21:11] <tarpman> anything in /usr/local/share/ca-certificates?
[21:11] <tarpman> probably not without reinstalling the entire system
[21:11] <tarpman> doubt you'll be able to purge the ca-certificates package without wiping out most of your system
[21:11] <relativemedia> yup stuff in there
[21:11] <relativemedia> ill duke it
[21:12] <relativemedia> yeah i was about to purge that but noticed it'd remove a bunch
[21:12] <tarpman> the update-ca-certificates manpage says anything in /usr/local/share/ca-certificates will be included in ca-certificates.crt
[21:12] <relativemedia> BINO
[21:12] <relativemedia> BINGO* fixed it
[21:12] <relativemedia> rm /usr/local/share/ca-certificates/*
[21:12] <relativemedia> stupid bugger
[21:12] <relativemedia> tarpman, it'd help to read the manpage right :P
[21:14] <sarnold> fwiw apt-get install --reinstall may let you get the 'overwrite this package' kind of results
[21:14] <relativemedia> thanks sarnold
[21:14] <sarnold> it wouldn't have helped in this case since it should avoid /usr/local ;) but still
[21:14] <relativemedia> i assume its a bad idea to move root's homedir to an nfs mount?
[21:16] <ihubuntu>  hello i try make auto installer ubuntu 16.10 but i can't skip select locale keymap and country http://pastebin.com/VkFEQnUW
[21:19] <sarnold> ihubuntu: do you get any warnings or errors in any log files, or other conolses?
[21:21] <ihubuntu> where get log in /var/log is empty
[21:23] <sarnold> do dmesg or journalctl work?
[21:23] <genii> If /var/log is empty, you probably have larger problems
[21:23] <sarnold> isntall environment is weird
[21:23] <sarnold> I don't know how it works :)
[21:23] <genii> Oh, installer. Look at out put from console 4 then
[21:27] <ihubuntu> a lot of error
[21:27] <ihubuntu> but i can't scroll log
[21:28] <sarnold> shift pageup perhaps?
[21:31] <ihubuntu> no shift pageup isn't working
[21:33] <tarpman> sarnold: the reason I didn't suggest that is it wouldn't (AFAIK) overwrite modified conffiles, or indeed delete locally-added ones...
[21:34] <tarpman> and yes I'm aware of the various --force-things
[21:34] <tarpman> maybe rm -rf /etc/ssl and dpkg --force-confnew would do it, but I wasn't about to recommend that wihtout testing locally first :D
[21:35] <sarnold> :)
[21:50] <ihubuntu> wah pressed options very different in documention
[21:57] <macskay_> Hey guys, I got disconnected earlier this morning but still got a question. I have a running LDAP on my system which works fine and multiple users that can successfully access the services which are connected to LDAP. I also got a "Change PW" PHP Script, which works as well. So the LDAP pw is updated successfully.
[21:58] <macskay_> The users are also allowed to use UNIX, but the PW does not change for unix, meaning they can't login to the server via SSH with the new password. Why is that?
[21:58] <sypher> macskay_: I *think* that passwords are cached for a certain period of time. At least, I've seen that in some configurations, but that may have been the function of third-party systems rather than native LDAP tools.
[22:54] <coreycb> zul, i pushed updates to horizon, it installs ok with the changes but needs work to get the static assets working