[06:44] is it possible that cloudinit will rerun with local configuration if network is not archieveable [06:45] i assume that it only run per instance with default configuration。 [07:20] 😊 [12:26] smoser: powersj: saw this in the log here: <_ix> 22:34:48> Any idea why I'd be getting `` in my network device table output [12:29] would we have a cloud-init.log from a centos boot in ci ? [12:52] rharper, well, its where it is rendering the table for display. :-( [14:08] smoser: right, missing a str method or something [14:59] will check it out. see if I can reproduce it [15:29] rharper: we reproduced it w/ lxd, I'm working on a fix for simpletawble [15:29] simpletable even [15:36] blackboxsw: cool [15:53] looks like that issue with simpletable also broke ssh_authkey_fingerprints [15:53] fix committed locally, adding a couple more unit tests [16:14] is there a clean way to remove a \n from a set_fact? I'm defining a value for it and not sure why that value, IE 0\n or 1\n is being set that way and not just 0 or 1 [16:14] wrong room... === Guest66098 is now known as mgagne [16:36] smoser: https://code.launchpad.net/~chad.smith/cloud-init/+git/cloud-init/+merge/332066 [16:36] intheclouddan[m]: :) [16:49] powersj, when c-i runs on a MP [16:49] it does not collect any artifcats, is that correct ? [16:49] smoser: correct only the console output of the tests [16:49] i'd have hoped to have been able to look at the artifacts from the integration test of the MP abov e from blackboxsw and see that it had hte desired content in /var/log/cloud-init-output.log [16:50] smoser: put in a card for me and I can change that [16:51] * blackboxsw installs on aws, cleaned-boot env to check [16:52] https://trello.com/c/fWq8GFKB [16:53] smoser: thx [16:57] http://paste.ubuntu.com/25714850/ an aws clean-boot run [16:58] with my branch [17:08] blackboxsw, it failed integration test for maas [17:08] i think just because no good reason [17:08] but... [17:08] <_ix_> I'd be down to test whatever you folks have got. [17:09] _ix_, the simpletable issue you reported is in process of being fixed [17:09] https://code.launchpad.net/~chad.smith/cloud-init/+git/cloud-init/+merge/332066 [17:09] <_ix_> That's really great. Thank you! [17:09] yeah that looks like the same case _ix_ [17:09] yeah we just ran into that today too [17:10] should have a daily copr build up today with that fix [17:13] powersj, https://jenkins.ubuntu.com/server/job/cloud-init-ci/398/console [17:13] do you recognize that as "transient yum noise?" [17:13] with a better placement of the '?' [17:13] * smoser triggers rebuild [17:14] https://jenkins.ubuntu.com/server/job/cloud-init-ci/399/ [17:16] * blackboxsw is trying to build in my centos lxc [17:16] smoser: yes [17:16] Cannot retrieve metalink for repository: epel/x86_64. Please verify its path and try again [17:16] is networking noise/issues [17:17] "mirror.bytemark.co.uk" because we know those super fast US -> UK pipes are always faster... not [17:19] faster than .jp pipes usually . "faster" is a relative term. [17:19] :) [17:19] true [17:34] ci looks happy this time https://code.launchpad.net/~chad.smith/cloud-init/+git/cloud-init/+merge/332066 [17:42] also just built and installed on centos7 in an lxc here to validate [17:54] smoser: are you working on the apport bug? otherwise I can peek [17:55] def raw_input_char(self, prompt): [17:55] '''raw_input, but read only one character''' [17:55] heh, kinda built into the DialogCLI class [18:04] blackboxsw, you can take a look for sure [18:04] i coudlnt figure out what was getting called [18:28] dmulford_, you there? [18:28] https://code.launchpad.net/~smoser/cloud-init/+git/cloud-init/+ref/fix/rh-subscription-add-null-check [18:28] smoser: yep, what's up? [18:29] just posted on your mp [18:30] blackboxsw, you want to pull your MP [18:30] or i can [18:30] thanks I'll pull. keeps me in practice. [18:30] smoser: ahh perfect :) [18:30] ok. [18:32] blackboxsw, ok. pull that, and then you want to do an artful new-upstream-snapshot ? [18:32] and then we can do one for x and z also [18:32] and while i wait i'll just think about the sru bugs. [18:34] smoser: thank you, this looks great! I will pull and test it out as you've requested [18:55] smoser: got a lead on the apport fix http://paste.ubuntu.com/25715415/ [18:56] that'll force read two chars if choices are > 10 (10 includes one key for the builtin "C" cancel). [18:56] anyway. onto the MP for cloud-init. then I can manufacture a 'good' fix for apport I think [18:57] was just testing on artful [19:01] merged simpletable fix. working on artful now [19:03] only change in apport behavior is that any choices with greater than 10 items, will have to hit after single-character choices [19:03] i didnt see where that was [19:03] that you had to hit enter. how does/did you do that ? [19:04] i'd let the choices go to 100 at least before complaint. its a long list, but you raising RuntimeError is just going to make someone not file a bug. [19:04] that they were going to file [19:05] smoser when you say 'y' for simple choices (Y/N/C) it is just one character [19:05] hm... funny. you could instead of raising RuntimeError you could file a bug "too many choices in question 'XXXX'" [19:05] if it's a 18 item choice (like what cloud you are on) you have to hit >enter> for choice 9 or 'c' [19:06] yeah, that makes sense. but ididnt see where you did that in your patch [19:06] it reads 2 chars. [19:06] ah [19:06] .strip() [19:06] yeah 1 chars plus the strip [19:06] 2 I mean [19:07] I was wondering what you thought about that RuntimeError. it's something that should file a bug against package X. [19:08] maybe it just appends to the report that is being created . .... 'question X had too many choices so user's couldn't respond' [19:08] s/user's/user/ [19:11] blackboxsw, why cuoldnt user respond ? [19:11] you arbitrarily picked '50' as too long [19:12] oh. it gets bad when over 100. casue then the user has to hit enter twice for < 10 :) [19:14] heh, will I just thought is was an arbitrarily long set of options to the point that it was unreadable [19:14] just a line in the sand. that doesn't really need to be drawn [19:14] but, the logic itself only handles up to 99 choices [19:15] blackboxsw, i'd let it go to 100. personally. [19:15] i have a scrollbar on my terminal [19:15] heh true [19:26] blackboxsw, https://hackmd.io/BwUwrAhgRg7ADFAtMCBOJAWCDHQGwyIBmcAJqiBnhiAExRRA?both [19:26] cank you see that ? [19:27] kind of etherpad + markdown. [19:27] <_ix_> Hello again -- do the successful builds appear in copr immediately? [19:35] powersj, can you confirm... [19:35] https://jenkins.ubuntu.com/server/view/cloud-init/job/cloud-init-copr-build/ [19:36] if i rebuild that, it will do trunk -> copr -> _ix_-gets-what-he-wants [19:36] smoser: I can see that [19:37] <_ix_> smoser: Very cool. I think we can be patient and wait for the regular nightly and give it a shot tomorrow. [19:38] blackboxsw: yes that pushes to the cloud-init-dev copr repo [19:38] smoser: ^ [19:39] _ix_, i just pushed "go" [19:39] so i'm guessing 20 minutes-ish [19:41] ok smoser I'm ready for pushing to artful I think [19:41] got time for a hangout? [19:41] to check that I've gone through the proper release steps [19:41] yeah [19:42] i'll join in m inute [19:55] smoser: https://code.launchpad.net/~chad.smith/cloud-init/+git/cloud-init/+merge/332076 [20:18] https://code.launchpad.net/~chad.smith/cloud-init/+git/cloud-init/+merge/332078 [20:27] https://code.launchpad.net/~chad.smith/cloud-init/+git/cloud-init/+merge/332080 [21:31] FYI there is a new daily build on COPR https://copr.fedorainfracloud.org/coprs/g/cloud-init/cloud-init-dev/ [21:44] blackboxsw, https://bugs.launchpad.net/cloud-images/+bug/1722663 [21:44] Ubuntu bug 1722663 in cloud-images "Azure: /etc/resolv.conf points to /run/resolvconf/resolv.conf in artful" [Critical,Confirmed]