=== r-daneel_ is now known as r-daneel
[18:12] <jocha> Hi all, so during a regular ubuntu 17.10 azure deployment I'm seeing a 15 second delay between two lines of cloudinit log, wondering if I could tap into some networking experts here who could tell me if something looks fishy here, logs here, https://paste.ubuntu.com/p/83MQQ8qQYr/
[18:12] <jocha> just realized I used here 3 times, :p
[18:12] <blackboxsw> jocha: you've been good and patient, I'm currently thrashing a bit on your branch trying to apply a patch to rework logic a bit more
[18:13] <blackboxsw> checking your comment
[18:13] <blackboxsw> your logic in the bug fix is good, I had wanted to avoid leaking the MARKER file outside scope of poll_imds  as it's not needed above that level, unfortunately I'm going through unit test changes now as a result
[18:16] <jocha> mmm ic fosure!
[18:16]  * blackboxsw wonders about bug: #1739672
[18:16] <ubot5> bug 1739672 in systemd (Ubuntu Artful) "Regression in getaddrinfo(): calls block for much longer on Bionic (compared to Xenial), please disable LLMNR" [Undecided,New] https://launchpad.net/bugs/1739672
[18:21] <blackboxsw> I know there's either a dhcp6 or ipv6 router advertisement bug that adds about 10s + to boot times. trying to fish for that bug #
[18:25] <blackboxsw> jocha: https://bugs.launchpad.net/ubuntu/+source/systemd/+bug/1765173
[18:25] <ubot5> Ubuntu bug 1765173 in systemd (Ubuntu) "networkd waits 10 seconds for ipv6 network discovery by default" [Undecided,Fix released]
[18:28] <jocha> sweet I'll take a look! thanks!
[18:52] <blackboxsw> jocha: I'm done on your branch for sure. testing it now on azure
[18:53] <blackboxsw> http://paste.ubuntu.com/p/MsVK5PZZ98/ is a suggestion to limit leaking REPORTED_READY_MARKER_FILE knowledge outside of poll_imds since nothing at outside that scope cares about the report_ready flag/file
[18:53] <jocha> @blackboxsw: thanks! also the bug you sent me looks exactly like what we're seeing, and it looks like the fix has been released, will look to try it out!
[19:45] <blackboxsw> tested your bug branch + the minor fix patch and things look good on my end . updated it for flakes http://paste.ubuntu.com/p/DtSmYqbgMV/
[20:26] <jocha> blackboxsw: Looks good to me! thanks for going through with it for me! :)
[20:29] <blackboxsw> good deal jocha, if you can apply that patch and git push, then I'll land it
[20:30] <jocha> cool will do!
[20:40] <jocha> git applied and git pushed :)
[20:44] <blackboxsw> cheers jocha
[20:47] <blackboxsw> oops fixing your commit message so each line is  < 74 chars long
[20:48] <blackboxsw> jocha: do you approve the updated commit message I added to your branch? https://code.launchpad.net/~jocha/cloud-init/+git/cloud-init/+merge/344192
[20:48] <blackboxsw> just prefaced with azure: and reformatted line length
[20:51] <jocha> approved!
[20:53] <blackboxsw> landed jocha thanks for the contribution
[20:54] <blackboxsw> it'll show in our next SRU to bionic
[20:54] <jocha> nice, ill be on the lookout
[20:55] <blackboxsw> we'll update the bug #1765214 when it officially publishes to (bionic|xenial|artful)-updates
[20:55] <ubot5> bug 1765214 in cloud-init "Multiple success messages sent to Azure Fabric if reboot occurs during pre-provisioning" [Medium,Fix committed] https://launchpad.net/bugs/1765214
[21:46] <paulmey> rharper: I updated https://code.launchpad.net/~paul-meyer/cloud-init/+git/cloud-init/+merge/344538 and added your suggested checks, ptal