[10:15] <Skuggen> How is apparmor set up for dep8 testing?
[11:08] <cking> i'm seeing gawk segfaults on xenial gawk ppc64el, 1:4.1.3+dfsg-0.1, https://bugs.launchpad.net/ubuntu/+source/gawk/+bug/1838014 - this is a regression for sure
[11:08] <ubottu> Launchpad bug 1838014 in gawk (Ubuntu) "random segfaults in gawk on ppc6el xenial, gawk 1:4.1.3+dfsg-0.1" [Critical,New]
[13:17] <rbasak> coreycb: heads up on a nova dep8 failure
[13:17] <rbasak> ERROR 1064 (42000) at line 2: You have an error in your SQL syntax; check the manual that corresponds to your MySQL server version for the right syntax to use near 'IDENTIFIED BY 'changeme'' at line 1
[13:18] <rbasak> That's because MySQL 8.0 has deprecated and removed the use of IDENTIFIED BY in a GRANT statement
[13:18] <coreycb> rbasak: thanks, is that eoan?
[13:18] <rbasak> The user's password has to be set during CREATE USER or with ALTER USER.
[13:18] <rbasak> coreycb: yes - the MySQL 8 transition is now in progress
[13:18] <coreycb> rbasak: right ok thanks we'll take a look
[13:19] <rbasak> We will get to it if you don't, but we have rather a long list of fixes to make right now :-/
[13:20] <rbasak> coreycb: let me know if you do look into it and hit anything else. I vaguely know of all the MySQL 8 breaking changes.
[13:21] <coreycb> rbasak: ok thank you
=== imcleod_ is now known as imcleod
[15:36] <rbasak> rafaeldtinoco: oh, one more thing
[15:36] <rbasak> rafaeldtinoco: we're staging patches at https://code.launchpad.net/~mysql-ubuntu/+git/mysql-8.0-transition
[15:36] <rbasak> Since most packages aren't imported into git-ubuntu, we can't use MPs
[15:37] <rbasak> Feel free to push directly there when ready
[15:46] <rafaeldtinoco> rbasak: perfect!
[15:46] <rafaeldtinoco> tks
[16:13] <bdmurray> How could I figure out why / how a systemd service has been disabled?
[16:16] <bdmurray> vorlon: ^?
[16:23] <vorlon> bdmurray: uh grep -r
[16:29] <bdmurray> vorlon: ?
[16:34] <vorlon> bdmurray: figuring out how/why it's been disabled, well, anything could've done it including a user, so if the user didn't do it you have to look around the system in all the places software might be running that would manipulate the state of the systemd units
[16:35] <vorlon> grep -r systemd.*$unit_name /var/lib/dpkg/info ?
[16:36] <bdmurray> vorlon: I'm the user and don't think I did it. Strangely whoopsie is active but disabled
[16:36] <vorlon> huh
[16:36] <vorlon> what's the output of systemctl status whoopsie?
[16:36] <bdmurray> and I don't see anything in journalctl strange since my most recent reboot which was this morning
[16:40] <bdmurray> Loaded: loaded (/lib/systemd/system/whoopsie.service; disabled; vendor preset: enabled)
[16:40] <bdmurray> vorlon: https://paste.ubuntu.com/p/Xpnk4rDHKk/
[16:53] <vorlon> bdmurray: ok; that state can be reproduced by running 'systemctl disable whoopsie' on a system that has the whoopsie service running
[16:53] <vorlon> but that doesn't say what did this
[16:58] <bdmurray> hrm, there are some bugs where I think people are having the same issue
[17:07] <vorlon> bdmurray: how did you come to notice it?  especially given that the service only started 1h43m prior to the pastebin
[17:10] <bdmurray> vorlon: "Here's a litte story I've got to tell..." so cinnamon-settings crashed and I clicked continue on the crash report and nothing happened and I remembered bug 1814611 (b/c I got an email about it this morning) and then I looked and didrocks MP from a while ago (https://code.launchpad.net/~didrocks/apport/whoopsie-auto-ui/+merge/348479) and then I saw the call to 'sytemctl is-enabled
[17:10] <ubottu> bug 1814611 in apport (Ubuntu) "ubuntu-bug never (any longer) opens browser window" [Undecided,Confirmed] https://launchpad.net/bugs/1814611
[17:10] <bdmurray> whoopsie.service' so I tried that.
[17:11] <bdmurray> Maybe is-active is a better check but regardless I wonder what's going on.
[17:11] <vorlon> huh
[17:12] <vorlon> so is-active will return false if the service has recently crashed and is in the process of restarting; is-enabled will return false in your present situation
[17:12] <vorlon> depends whether you're trying to check state (is-active) or intent (is-enabled)
[17:46] <juliank> Maybe someone here has an idea:
[17:46] <juliank> swift post "moo" works, but the script does not: https://paste.ubuntu.com/p/8s8smBRVwQ/
[17:46] <juliank> it's using the same environment variables?
[17:52] <juliank> ah
[17:53] <juliank> swift command-line client is using OS_ variables
[17:53] <juliank> and there's a typo somewhere
[18:05] <connor_k> Aside from putting together some kind of scraper, is there a way to view all the child wiki pages descending from a single wiki page root on the ubuntu wiki? I.e., all pages descending from https://wiki.ubuntu.com/Kernel/
[18:06] <connor_k> using the search menu for something like Kernel/ gets me close, but it's paginated and I suspect it's matching a number of pages that are not specifically descended from the root one
[18:43] <sarnold> connor_k: oh god that sounds terrible. the wiki is SO SLOW scraping a subset will take an absolute eternity
[18:43] <connor_k> sarnold, actually launchpad was surprisingly spry today
[18:44] <connor_k> i did this horrible thing to accomplish what I wanted: https://paste.ubuntu.com/p/z4cxXqw8Q5/
[18:44] <connor_k> ^ certainly not intended to be an actual solution for anyone else, I'm basically throwing that script away lol
[18:45] <sarnold> connor_k: hah, cool :) how long did that thing take?
[18:45] <connor_k> sarnold, about 10 seconds to get the ~270ish results I was interested in
[18:46] <sarnold> nice