=== svetlana is now known as Sveta
=== lotuspsychje_ is now known as lotuspsychje
[07:17] <lordievader> Good morning
=== cpaelzer_ is now known as cpaelzer
[09:22] <jamespage> coreycb: as openstack and ubuntu are now re-aligned on release dates we prob need to get the standing freeze exception from the release team for os packages
[09:26] <jamespage> coreycb: done - cc'ed you and sahid
[09:27] <jamespage> sahid - for reference - we're past the feature freeze date for Ubuntu 19.04 development; for the past few cycles OpenStack has released in advance of this date due to the staggered release schedules
[09:28] <jamespage> but now the release weeks are back to back, so I've requested a general freeze exception for the OpenStack packages so we can continue to upload snapshots, RC's and release versions
[09:28] <jamespage> sahid: https://wiki.ubuntu.com/FreezeExceptionProcess
[09:29] <jamespage> as a refrerence
[09:51] <sahid> jamespage: thanks for the heads-up and details
[12:36] <coreycb> jamespage: thanks
[12:51] <coreycb> jamespage: sahid: i'm going to promote stein-staging to stein-proposed and get some focus on testing next week
[12:51] <coreycb> jamespage: oh yay os-ken is tehre
[13:08] <jamespage> coreycb: yep - infinity acked it into the archive during the week
[13:08] <jamespage> I pushed it through
[13:09] <jamespage> +1 on the promotion - enable and tidy stein is a goal for next week for me ;)
[13:10] <coreycb> jamespage: great, +1 on that
[16:48] <UForgotten> It would appear that since an update on the 21st, a few of us can't build xenial machines anymore.  The problem I am observing is that I can preseed an install, and it downloads packages and gets to libc6-udeb and seg fault 18's error 4 on libc-2.23.so and then the box can't spawn anything further.
[16:49] <UForgotten> Need to get a bug opened, and looking for the fastest path to resolution.
[16:59] <tomreyn> The other person reporting this is |easy| (in #ubuntu). both UForgotten and |easy| report that their automated deployments of 16.04, using netboot or mini.iso, fail. |easy| reports they are using http://de.archive.ubuntu.com/ubuntu/ubuntu/ubuntu/dists/xenial-updates/main/installer-amd64/current/images/netboot/mini.iso and that deployments broke between 2019-02-20 18:04 and 2019-02-21 14:49
[17:01] <tomreyn> which package should (probably release related) bugs about mini.iso be reported against?
[17:02] <|easy|> UForgotten: what to to to fix it ?
[17:12] <UForgotten> |easy| how to fix it? we have to convince someone in here who can fix it that it's really a problem and get them to fix it, or figure out a workaround ;)
[17:13] <|easy|> UForgotten: a work around is not to use netboot
[17:14] <|easy|> UForgotten: the server iso is working, i know this is not nice but it work for now
[17:14] <|easy|> UForgotten: i think after some googling it ist the same bug from past come up again
[17:15] <|easy|> UForgotten: look https://bugs.launchpad.net/ubuntu/+source/debian-installer/+bug/1753449
[17:15] <ubottu> Launchpad bug 1753449 in debian-installer (Ubuntu) "Ubuntu 18.04 netboot installation failed during loading of installer component libc6-udeb" [Critical,Fix released]
[17:15] <UForgotten> oh, fun. I wasn't able to find any examples of the segfault. but it makes sense if someone did a bad code merge into xenial
[17:16] <UForgotten> ah. maybe they fixed in bionic but somehow it snuck into xenial during a backport of something
[17:16] <|easy|> UForgotten: i think so
[17:29] <tomreyn> UForgotten + |easy|: i suggest that one of you files a bug against the "debian-installer" package, the other checks whether whatis discussed there also applies to them and if so, confirms it.
[17:30] <tomreyn> dates, locations (URLs) and most of all file checksums / hashes will help identify any issues and get them solved faster.
[17:30] <UForgotten> that 1753449 bug is definitely my problem, but in xenial instead of bionic.  bionic is working fine atm.
[17:30] <UForgotten> if an older ISO will work, I'm going to try downloading and using that instead of the netboot
[17:31] <tomreyn> that's a fixed bug, you can refer to it in the new bug report.
[17:47] <tomreyn> UForgotten + |easy|: I really recommend filing a bug report if you'd like this fixed.
[17:47] <tomreyn> personally i don't use those images, so won't file it.
[17:48] <UForgotten> If I can get installs working with an older iso, then that's what I'm going to do, I can't wait for this to maybe get fixed, but I will come back and open a bug if it's not a known issue
[17:49] <tomreyn> that's a good way to delay bug triage. but totally your choice. ;-)
[17:52] <|easy|> tomreyn: one more account
[17:52] <|easy|> tomreyn: ok i open a bug report
[17:53] <|easy|> tomreyn: give me some time to create an account an verify them
[18:00] <tomreyn> you don't depend on me there, since i was not able to reproduice it.
[18:02] <|easy|> tomreyn: its ok it depend on me i can reproduice it and i want to have a fix for comfort and easy use
[18:04] <|easy|> tomreyn: at moment i have a workaround and it break my automatic but i can have a working installation with work around but for future or for more then 2 machine to install it would be nice to have automatic back ;)
[18:06] <tomreyn> |easy|: cool, i'll be happy to read over it and double-check no relevant info provided on irc tonight is missing. just be sure to provide exact and reproducible (by someone who didn't take part in our chat) instructions.
[18:07] <tomreyn> so explaining how anyone else will likely run into this issue, and accompanyigńg this by urls, timestamps, file hashes, should do good.
[18:21] <|easy|> tomreyn: ok hope its clear see https://bugs.launchpad.net/ubuntu/+source/debian-installer/+bug/1817358
[18:21] <ubottu> Launchpad bug 1817358 in debian-installer (Ubuntu) "Ubuntu 16.04 netboot installation failed during loading of installer component libc6-udeb" [Undecided,New]
[18:23] <|easy|> UForgotten: feel free to add a comment to my bug report and describe youre setup
[18:23] <|easy|> UForgotten: maybee it help to fix it
[18:23] <UForgotten> thanks so much for putting that in |easy| I will keep an eye on it
[18:43] <tomreyn> |easy|: thanks for passing along the url. i'll have a look a little later (but i don't triage bug reports). if UForgotten can confirm this report as it is, by logging in and setting the "also affects me" option, this bug report would change state to confirmed, which can increase its position in the bug triage queue.
[18:44] <tomreyn> it'd also increase its credibility, ruling out the OPs environment as a cause.
[18:47] <tomreyn> UForgotten, |easy|: can i have your permission to quote what you said in #ubuntu and here on the bug report?
[18:47] <UForgotten> I'm going to post a comment with my info
[18:55] <tomreyn> |easy| reported on IRC they they were using the mini.iso with MD5SUM ed2bbe59e5dac40b92e0ddeb6b206ef8 - this info ddidnt make it to the bug report
[18:56] <tomreyn> UForgotten: which HV did you say you're using there?
[18:57] <tomreyn> or is it bare metal for you?
[18:57] <UForgotten> its libvirt/qemu/kvm using virt-install on a xenial host
[18:58] <tomreyn> traditional bios booting, i assume, not ovmf?
[19:04] <|easy|> tomreyn: you can quote me
[19:08] <tomreyn> UForgotten: can i quote you?
[19:12] <compdoc> must be a netboot thing. I install 18.04 in kvm a lot
[19:14] <tomreyn> compdoc: this is about the latest (/updates/) 16.04 mini.iso
[20:09] <tomreyn> |easy| UForgotten: It appears that I was (more?) successful in searching for duplicate reports: https://bugs.launchpad.net/bugs/1816846
[20:09] <ubottu> Launchpad bug 1816846 in glibc (Ubuntu) "segfault in libc-2.23.so netinstall installation pxe" [Undecided,Confirmed]
[20:09] <UForgotten> tomreyn sorry I was in meetings. You can quote whatever is in that bug
[20:10] <tomreyn> filing this against glibc seems wrong to me, but we shall see
[20:10] <tomreyn> UForgotten: too late, i'm done. but you can add anything you'd like to.