=== rangerpbzzzz is now known as rangerpb
=== tekniq_ is now known as tekniq
[16:00] <blackboxsw> Looks like somebody else has an openbuild cloud-init up with some stale package references. https://build.opensuse.org/package/show/Cloud:Tools/cloud-init
[16:34] <rharper> interesting
[19:04] <powersj> bah launchpad failed to create a diff...
[19:29] <blackboxsw> smoser, this bug is linked as sru https://bugs.launchpad.net/cloud-init/+bug/1644064 but the related branch is "Needs Fixing" what should it be
[19:29] <ubot5`> Ubuntu bug 1644064 in cloud-init (Ubuntu) "sshd_config file permission changed to 644 if ssh_pwauth value is true or false" [Undecided,New]
[19:31] <blackboxsw> looks like all review comments were addressed
[19:32] <blackboxsw> just not merged
[19:34] <powersj> rharper: thanks for the 2nd review
[19:36] <blackboxsw> powersj, sure, anyday (including today works well for me)
[19:36] <blackboxsw> powersj, sure. Anyday .including today, works well for me.
[19:37] <blackboxsw> geez I give up on English
[19:37] <powersj> blackboxsw: Today works for me as well
[19:37] <powersj> lol
[19:37] <powersj> just tell me where
[19:37] <blackboxsw> ok I'm just going through sru bugs to create a template.
[19:37] <blackboxsw> and trying to reproduce failing tests
[19:39] <blackboxsw> powersj, how's 3:30 sound, someone might drop a delivery by the house sometime between now and 3:30.
[19:40] <blackboxsw> I can go anywhere (your choosing as last time we didn't have internet0
[19:40] <powersj> blackboxsw: 330pm dazbog harmony/timberline
[19:40] <blackboxsw> done
[20:00] <smoser> blackboxsw, i suspect its fix-released in artful
[20:01] <smoser> looking. i just marekd it fix-committed in cloud-init
[20:27] <blackboxsw> smoser, thanks, does this bug need to be SRU'd? https://bugs.launchpad.net/cloud-init/+bug/1636531  it's just unit tests?
[20:27] <ubot5`> Ubuntu bug 1636531 in cloud-init (Ubuntu) "unittests blkid command fails on slave s390x" [Undecided,New]
[20:27] <blackboxsw> which aren't included in the deb anyway
[20:30] <nacc> blackboxsw: probably we want the unit tests to pass on older releases?
[20:30] <nacc> blackboxsw: unclear, i guess
[20:30] <nacc> blackboxsw: although the 'fix' is a jenkins change?
[20:30] <blackboxsw> well it's a unit test not leaking blkid calls change which could fail on a variety of systems
[20:30] <blackboxsw> though I just found other leaks in testing the fix
[20:31] <blackboxsw> but not really part of this bug.
[20:31] <nacc> blackboxsw: ok, yeah, i guess it's not clear to me, i was just looking at powersj's response
[20:32] <blackboxsw> yeah I'm with you nacc, I'm not quite certain why that makes an SRU cut. unless we have ungoing jenkins runs against proposed/released cloud-init too
[20:32] <blackboxsw> which I guess is a goal
[20:32] <powersj> if I recall smoser has SRU'ed test fixes before
[20:32] <nacc> powersj: yep, we have
[20:32] <powersj> as I remember doing SRU versification on them
[20:32] <nacc> powersj: i'm just not sure here if the fix is in curtin but really a jenkins change?
[20:33] <blackboxsw> powersj, how'd you verify unit tests as they aren't delivered in the deb produced. did you have to git clone the repo and attempt to run the tests again?
[20:33] <powersj> There were two fixes, 1) the jenkins path change, so no action here, and 2) smoser made a new branch with some other changes related to this issue to prevent it in the future
[20:33] <nacc> powersj: ah i see
[20:34] <blackboxsw> yeah the part 2) is what I was wondering about. as it is just unit test mocks
[20:34] <powersj> blackboxsw: pull-lp-source cloud-init <release>; cd cloud-init-*; run tests
[20:34] <blackboxsw> heh nice
[20:34] <blackboxsw> TIL ^
[20:35] <nacc> powersj: didn't you have a test that did this specifically in jenkins?
[20:35] <nacc> powersj: that is, ran the unit tests (or vmtests?) for a given release from the src for that release?
[20:36] <powersj> blackboxsw: of course if you do it that way make sure to specify the version so they know it lines up and make sure to use release-proposed in the command
[20:36] <powersj> nacc: yeah we have the proposed tests that run when cloud-init is there.
[20:36] <powersj> it does the same thing
[20:36] <nacc> powersj: yep, ok
[20:37] <powersj> nacc: to be clear though, the jenkins job runs the integration tests
[20:37] <nacc> powersj: we have so many new curtin/cloud-init jenkins jobs, i often don't know what does which :)
[20:37] <nacc> powersj: ah ok
[20:37] <powersj> so not unit tests, but maybe I should add those anyway
[20:37] <powersj> nacc: I know :( I'm not sure how to clean them up either
[20:37] <nacc> powersj: unit tests seem like a self-consistency check that's worthwhile
[20:37] <powersj> I'm glad we have expanded coverage of things, but it isn't always clear
[21:53] <smoser> blackjid, regarding "how'd you verify unit tests..."
[21:53] <smoser> there is no ubuntu deb if the unit tests didn't run
[21:53] <smoser> as they're run as part of build
[21:55] <powersj> blackboxsw: https://paste.ubuntu.com/24495122/
[21:56] <blackboxsw> thanks didn't realize disconnected from freenode
[21:58] <blackboxsw> ok thanks, yeah I see unit tests being run in the build recipe before the package build. ok https://launchpadlibrarian.net/317312745/buildlog_ubuntu-xenial-amd64.cloud-init_0.7.9-1495-g2796dab-0ubuntu1+1267~trunk~ubuntu16.04.1_BUILDING.txt.gz
[21:59] <smoser> blackboxsw, yeah..i was going to find al ink to that
[21:59] <smoser> but for some reason... launchpad.net isn't resolving for me
[21:59] <smoser> fun
[22:01] <powersj> smoser: can you add artful to https://launchpad.net/~cloud-init-dev/+archive/ubuntu/test-archive
[22:01] <powersj> and https://launchpad.net/~curtin-dev/+archive/ubuntu/test-archive
[22:01] <smoser> not until i make launchpad.net resolv :)
[22:01] <powersj> hahaha
[22:01] <powersj> sorry just read that
[22:03] <smoser> http://paste.ubuntu.com/24495160/
[22:03] <smoser> someone really doesnt want me to et an ip for launchpad.
[22:03] <smoser> (iv'e restarted network manaer, and killed the dnsmasq process that it is running)
[22:03] <smoser> but 'host launchpad.net 8.8.8.8' should be asking 8.8.8.8 directly
[22:08] <smoser> powersj, ok.. so added launchpad.net to /etc/hosts
[22:08] <smoser> i can do that, but i
[22:08] <smoser> i am pretty sure you could have too
[22:08] <smoser> "Package details"
[22:08] <smoser> "copy packages"
[22:08] <powersj> for curtin maybe, I have no rights on cloud-init
[22:09] <powersj> ah ha! ok I see you already did it for curtin
[22:09] <smoser> the key thing to remember is that when you do it..
[22:09] <rharper> smoser: if you're at a sprint, sometimes you need to push 8.8.8.8 infront of your caching nameserver; I've had to do that if DNSSEC was enabled, if youre on zesty or artful, systemd-resolvd tries to use DNSSEC
[22:09] <smoser> *copy existin binaries*
[22:09] <powersj> ah so no rebuild
[22:09] <smoser> rharper, see my paste
[22:09] <smoser> i'm on xenial here.
[22:10] <smoser> but even directly asking (host launchpad.net 8.8.8.8)
[22:10] <smoser> didn't work
[22:10] <smoser> i'mi pretty sure that is directly asking 8.8.8.8 for launchpad.net
[22:10] <smoser> so something (i think) is udp intercepting
[22:10] <rharper> yes
[22:10] <rharper> they intercept dns
[22:10] <smoser> and has a negative cache for that.
[22:10] <rharper> locally
[22:11] <rharper> for example, they deny phone updates so there's no phone storm of updates on the network
[22:11] <rharper> there are a list of other DNS hijacks via the sprints AP
[22:11] <rharper> I used to run DNSSEC enabled local cache and it *always* broke at the sprint
[22:11] <rharper> not quite sure why launchpad.net would be busted
[22:11] <smoser> oh. i'm not on canoncial
[22:12] <smoser> on mariott
[22:12] <rharper> oh well, they could do the same thing at the hotel
[22:12] <smoser> yeah. but seems still busted to cache a negative result
[22:12] <rharper> ie, hijacking dns for shit n giggles
[22:12] <rharper> if you bring up vpn, it works right ?
[22:12] <smoser> well, i think i have vpn sending only vpn traffic
[22:13] <rharper> you can flip the bit to send it all
[22:13] <smoser> yeah
[22:13] <rharper> which would direct dns through vpn (at least temp)
[22:13] <smoser> but i assume this problem will go away at some point
[22:13] <smoser> and i just put /etc/hosts entry in
[22:13] <rharper> well, it does, until you forgot you put in there and then at some point in the future you freak out again
[22:13]  * rharper did that with diglett, bastion and other hosts
[22:14] <rharper> why u no connect! ?!
[22:14] <nacc> rharper: i think DNSSEC has been deisabled again, right?
[22:14] <rharper> in systemd-resolvd, I think so
[22:14] <smoser> right.
[22:14] <rharper> or resolved
[22:14] <nacc> rharper: yeah, i got a prompt at some point last week
[22:14] <rharper> flippin spelling
[22:15]  * rharper is playing with pihole 
[22:15] <rharper> putting that rpi2 to work
=== rangerpb is now known as rangerpbzzzz
[22:31] <blackboxsw> generally speaking are these the typical things we ask for when users submit bugs against cloud-init http://pastebin.ubuntu.com/24495391/?
[22:34] <smoser> blackboxsw, all of /var/lib/cloud/ is good too
[22:34] <smoser> and /var/log/cloud-init-output.log
[22:34] <smoser> and /me goes away
[22:35] <blackboxsw> +1 smoser
[22:39] <powersj> blackboxsw: https://paste.ubuntu.com/24495434/ is what our integration tests pull after each test
[22:39] <blackboxsw> thx ppo
[22:39] <blackboxsw> thx powersj
[22:40] <powersj> smoser: thanks for the ppa updates
[22:57] <blackboxsw> ok updated bug filing request https://bugs.launchpad.net/cloud-init/+filebug
[22:58] <blackboxsw> just to request the info we care about when someone files bugs
[22:59]  * blackboxsw relocates home after picking up some food 
=== blackboxsw is now known as blackboxsw_bl
=== blackboxsw_bl is now known as blackboxsw_bbl