[14:02] <powersj> https://bugs.launchpad.net/cloud-init/+bug/1736174
[14:02] <powersj> someone should respond to that one
[16:04] <blackboxsw> alrighty, got my ducks in a row....
[16:05] <blackboxsw> #startmeeting Cloud-init bi-weekly status meeting
[16:05] <meetingology> Meeting started Mon May 14 16:05:28 2018 UTC.  The chair is blackboxsw. Information about MeetBot at http://wiki.ubuntu.com/meetingology.
[16:05] <meetingology> Available commands: action commands idea info link nick
[16:08] <blackboxsw> welcome folks to another cloud-init status meeting. This forum is used to communicate the recent changes,  current development efforts and host open office hours to help get quick discussion on bugs,  branches or features of interest to anyone developing (or consuming) cloud-init.
[16:09] <blackboxsw> We'll go through a couple of topics as usual (Recent changes, In-progress Development, Office Hours), if there are any additional topics needed just let me know.
[16:09] <blackboxsw> #topic Recent Changes
[16:10] <blackboxsw> We track our upstream work publicly on trello. Feel free to participate or ask questions about any feature work that is seen up there if there are concerns.
[16:11] <blackboxsw> #link https://trello.com/b/hFtWKUn3/daily-cloud-init-curtin
[16:11] <blackboxsw> We have landed the following work items: beware the paste :)
[16:11] <blackboxsw> * Completed release of 18.2 to Ubuntu Bionic, Artful, and Xenial
[16:11] <blackboxsw> * Implement bash completion script for cloud-init command line
[16:11] <blackboxsw> * Improved Softlayer datasource documentation
[16:11] <blackboxsw> * net: Depend on iproute2's ip instead of net-tools ifconfig or route
[16:11] <blackboxsw> * Accept-ra unset by default (LP: #1732002)
[16:11] <blackboxsw> * Command collect-logs will only include most recent boot's journal (LP: #1766335)
[16:11] <blackboxsw> * FreeBSD: Invoke growfs on ufs filesystems such that it does not prompt. (LP: #1404745)
[16:11] <blackboxsw> * Azure: poll_imds fixes to only report 'ready' state once during pre-provisioning from Joshua Chan (LP: #1765214)
[16:11] <blackboxsw> * DataSourceSmartOS: fix hang when metadata service is down from Mike Gerdts (LP: #1667735)
[16:12] <blackboxsw> * DataSourceSmartOS: change default fs on ephemeral disk from ext3 to ext4 from Mike Gerdts (LP: #1763511)
[16:12] <blackboxsw> * pycodestyle: Fix invalid escape sequences in string literals
[16:14] <blackboxsw> A big thank you to community involvement again. Thanks jocha(Microsoft) and mgerdts(Joyent) for the recent branch work supporting Azure and SmartOS clouds respectively
[16:16] <blackboxsw> We also went through another round of StableReleaseUpdates for Ubuntu on Xenial and Artful to pull in IBMCloud platform fixes. putting Xenial and artful at 18.2-4-g05926e48-0ubuntu~16.04.2 | ~17.10.2
[16:18] <blackboxsw> I think that's about it for completed development...
[16:19] <blackboxsw> #topic In-progress Development
[16:21] <blackboxsw> There are a couple of items being worked actively at the moment:
[16:22] <blackboxsw> * SRU of cloud-init tip into bionic (should land today or tomorrow) 	18.2-27-g6ef92c98-0ubuntu1~18.04.1
[16:23] <blackboxsw> * SmartOs datasource detection improvements
[16:23] <blackboxsw> * Moving OpenStack datasource to get detected earlier at 'local' stange instead of 'network' stage using ephemeral dhcp client
[16:26] <blackboxsw> * read_file_or_url fixes returing text content in all cases
[16:26] <blackboxsw> * various upstream bug fixes
[16:27] <blackboxsw> * powersj: is also investigating a move to a centralized library for our cloud testing.
[16:29] <robjo> blackboxsw: w.r.t. Depend on iproute2's ip instead of net-tools ifconfig or route was this a merge of iproute2tools branch? I don't recall seeing a merge notification but am way behind in e-mail
[16:29] <blackboxsw> think that about captures what upstream is working on. I think we can transition to office hours for ~30 mins for anyone to bring up ideas of interest
[16:29] <blackboxsw> #topic Office Hours (next ~30 mins)
[16:29] <blackboxsw> hi robjo, checking status there
[16:30] <blackboxsw> I know we landed one branch on that topic
[16:31] <blackboxsw> robjo: so we had a couple branches to packaging dependencies in ubuntu to call out iproute2 specifically as a hard package dependency.
[16:32] <blackboxsw> robjo: and the code changes (which took in some of your branch content  and review comments) landed in rev 6d48d265a0548a2dc23e587f2a335d4e38e8db90
[16:32] <robjo> OK, so I can delete my branch
[16:32] <blackboxsw> https://pastebin.ubuntu.com/p/266CyDt9gD/
[16:34] <robjo> thanks, so we'll get that in 18.3?
[16:34] <blackboxsw> robjo: yes I think i marked you co-author on that branch and pulled in all your changes to cloudinit/config/cc_disable_ec2_metadata.py
[16:34] <blackboxsw> 2	
[16:34] <blackboxsw> thanks again for that, sorry for the back and forth as I hadn't seen your original branch.
[16:35] <blackboxsw> robjo: definitely in 18.3
[16:35] <blackboxsw> it landed a week or two after the 18.2 cut.
[16:35] <stanguturi> @blackboxsw, Can someone please provide inputs for the bug https://bugs.launchpad.net/ubuntu/+source/cloud-init/+bug/1766538 One of our team members has uploaded the necessary log files to the bug.
[16:38] <robjo> OK, so lets also revisit some of the other stuff I have floating about as I just did the 18.2 package for openSUSE and SLES and noticed that I am once again scarring a lot of patches :(
[16:39] <robjo> blackboxsw: https://code.launchpad.net/~rjschwei/cloud-init/+git/cloud-init/+ref/emptyStageOK should be back in your court, did you get notification?
[16:40] <robjo> blackboxsw: https://code.launchpad.net/~rjschwei/cloud-init/+git/cloud-init/+ref/noLnxDistro pending since January this may have some merge conflicts now as I had to fiddle quite a bit with the patch in my package
[16:41] <blackboxsw> stanguturi: looking
[16:41] <robjo> I think rharper is working on a different approach to https://code.launchpad.net/~rjschwei/cloud-init/+git/cloud-init/+ref/chrony when can we expect that?
[16:42] <robjo> and does rharper account for the fact that ntp has a different service name on different distributions?
[16:43] <blackboxsw> #link https://bugs.launchpad.net/ubuntu/+source/cloud-init/+bug/1766538
[16:43] <blackboxsw> #link https://code.launchpad.net/~rjschwei/cloud-init/+git/cloud-init/+ref/emptyStageOK
[16:43] <blackboxsw> ahh robjo hadn't, was on vacation Friday, will grab that/close out today
[16:43] <blackboxsw> #link https://code.launchpad.net/~rjschwei/cloud-init/+git/cloud-init/+ref/noLnxDistro
[16:44] <rharper> robjo: re: ntp/chrony, that's landed, including service names for different distros; I pulled unittests and scenarios from your branch; please look over master and see if we're missing anything from your branch w.r.t use-case/scenario
[16:45] <robjo> rharper: OK, if it's landed I can at least throw my branch away, and yes, will take a look at master, I take it this will be another in 18.3 item?
[16:46] <blackboxsw> stanguturi: ok thanks for the logs on that bug, looks like network config parsing is falling over and OVF datasource isn't being detected https://pastebin.ubuntu.com/p/qVJxDJWZRV/
[16:47] <stanguturi> @blackboxsw, Oh . Thanks. Will check with him about the test setup and update the bug. Thanks.
[16:48] <blackboxsw> updated the bug with a comment there
[16:48] <blackboxsw> thanks stanguturi
[16:50] <blackboxsw> robjo: correct as well for ntp/chrony, we held off landing it in 18.2 because of risk
[16:50] <blackboxsw> it was one of the first branches landed after the cut
[16:51] <robjo> ok, leaves the https://code.launchpad.net/~rjschwei/cloud-init/+git/cloud-init/+ref/noLnxDistro implementation of distro detection as things I'd like to get resolved, others to be addressed maybe in 2 weeks ;)
[16:51] <blackboxsw> ok noLnxDistro....  I'm updating the commit comment robjo to the trailing LP: #<bug_id>
[16:51] <blackboxsw> claiming a review slot on that now
[16:53] <blackboxsw> good unit test coverage, thanks for that
[16:54] <robjo> np
[16:55] <blackboxsw> ok this can be reviewed today, not sure if why we don't already have a get_linux_distro utility somewhere, but I'll poke around today for context
[16:55] <blackboxsw> ahh ahh, thanks for the bug robjo ok
[16:56] <robjo> the context is that the Python implementation is going away and has been deprecated
[16:56] <blackboxsw> #link https://bugs.launchpad.net/cloud-init/+bug/1745235
[16:56] <blackboxsw> thanks again
[16:56] <blackboxsw> changing that status and will help you get that landed
[16:56] <blackboxsw> good one
[16:58] <robjo> OK, blackboxsw is on the hook for two things this week ;) distro detection and the empty modules list
[16:58] <blackboxsw> #action blackboxsw review distro dection and empty modules list
[16:58] <meetingology> ACTION: blackboxsw review distro dection and empty modules list
[16:58]  * robjo on the hook to look at chrony support in master and report back to rharper
[16:58] <blackboxsw> #action robjo review existing chrony support in master per rharper's work
[16:58] <meetingology> ACTION: robjo review existing chrony support in master per rharper's work
[16:59] <blackboxsw> official now :)
[16:59] <blackboxsw> now if I only reviewed previous meeting's action items.... checking now
[17:00] <blackboxsw> 16:51 <blackboxsw> #action blackboxsw to have discussions w/ team on datasource maintaining network on each reboot per  https://code.launchpad.net/~mgerdts/cloud-init/+git/cloud-init/+merge/343712
[17:01] <blackboxsw> ok per last meeting, we have held a couple of discussions on hotplug vs. maintaining network only on reboot. on first blush it looks like we'd need to have these mutually exclusive, but we are just started to iron our what we want to do for our initial hotplug support in cloud-init and have to have a followup discussion about how to support both approaches
[17:02] <blackboxsw> #action blackboxsw carryover network hotplug vs network maintenance on reboot-only
[17:02] <meetingology> ACTION: blackboxsw carryover network hotplug vs network maintenance on reboot-only
[17:02] <blackboxsw> well that was the only action item from last meeting looks like
[17:02] <blackboxsw> ok I think that wraps up today's meeting.
[17:02] <blackboxsw> any other parting shots folks?
[17:03] <blackboxsw> Thanks again for your time. It's always a pleasure.
[17:03] <blackboxsw> Next meeting two weeks, same bat time...
[17:03] <blackboxsw> #endmeeting
[17:03] <meetingology> Meeting ended Mon May 14 17:03:58 2018 UTC.
[17:03] <meetingology> Minutes:        http://ubottu.com/meetingology/logs/cloud-init/2018/cloud-init.2018-05-14-16.05.moin.txt
[17:04]  * blackboxsw heads to publish those notes now to cloud-init.github.io
[17:29] <larsks> smoser: ci asplode, and I'm pretty sure it's not related to my patch: https://jenkins.ubuntu.com/server/job/cloud-init-ci/1115/console
[17:42] <powersj> larsks: I'm getting those failures when running your branch locally as well, but master is fine
[17:42] <larsks> That's very strange.
[17:42] <larsks> The only change was a move to the add_patch method that smoser suggested.
[17:42] <larsks> I guess I will take a closer look this evening. Thanks for checking.
[18:06] <larsks> Meh, fixed it. Made bad assumptions about add_patch.
[19:00] <robjo> rharper: looks as everything for the chrony support/ ntp naming is there, the comment I have is that this implementation is inconsistent with other concepts used
[19:01] <robjo> so far we had distro specific settings in the distro class implementation, now with ntp with part of the implementation in the cc_ntp module and other parts in the distro
[19:02] <robjo> I think in the long run this will be confusing and I think it is worth it to have a task to eventually move the distro specific bits into the distro implementation
[19:09] <smoser> powersj: did you run something here manually ?
[19:09] <smoser> https://code.launchpad.net/~rjschwei/cloud-init/+git/cloud-init/+merge/334992
[19:10] <powersj> smoser: nope it would say started by me
[19:10] <smoser> hm..
[19:11] <smoser> so why did that run ?
[19:11] <smoser> and https://code.launchpad.net/~rjschwei/cloud-init/+git/cloud-init/+merge/335406 also
[19:11] <powersj> those could have been someone pressing the "rerun" link, which isn't something I have a lot of experience with
[19:14] <blackboxsw> powersj: even the rerun attributes to a person I though
[19:14] <blackboxsw> t
[19:14] <blackboxsw> smoser: can I land https://code.launchpad.net/~smoser/cloud-init/+git/cloud-init/+merge/343738
[19:14] <powersj> hmm I think so since you'd have to be logged in
[19:22] <smoser> blackboxsw: i think so...
[19:22] <smoser> convince yourself that we're effectively doing the same check on METADATA_SOCKFILE in DataSourceSmartOS
[19:23] <smoser> i am pretty sure that as it is right now for a lxd container inside a LX_BRAND we would have the 'self.md_client.exists()' call return False and take the "No metadata device '%r' ..."
[19:24] <smoser> (line 230)
[19:24] <smoser> right?
[19:49] <blackboxsw> smoser: ok see your followup message on the branch. for some reason I'm not highlighting in IRC today. sorry reply posted to your branch. I think minimally I'd like to see the socket file check in DataSourceSmartOS if we could.
[20:25] <smoser> blackboxsw: i'm pretty sure we are checking that
[20:33] <rharper> robjo: smoser and I went back and forth on this many times;  it's a struggle between having all of the ntp knowledge within the module itself, asking only distro for paths/defaults vs pushing a portion, but not all of the ntp handler into the distro class;
[21:02] <paulmey> @rharper /@smoser , can I get another review of https://code.launchpad.net/~paul-meyer/cloud-init/+git/cloud-init/+merge/344538 ? Thanks!
[21:50] <thomasking> Is there an issue with growpart and resizefs on Ubuntu 18.04? Both show disabled even though configs for each are enabled/auto.
[21:58] <thomasking> added bug, https://bugs.launchpad.net/cloud-initramfs-tools/+bug/1771221
[22:38] <blackboxsw> responded on bug, looking for a bit more data.