Monday Triage 2025-01-27

[John Chittum]

### Bug Triage ###

Found 35 bugs

Notables and new:

1178274 <https://bugs.launchpad.net/ubuntu/+source/ruby-json/+bug/1178274>
: pcs & ruby-json MIR
    raised and assigned this week.

1841930 <https://bugs.launchpad.net/ubuntu-manpage-repository/+bug/1841930>
: w3m
    a new comment confirming older issue. Led me to check if upstream had
touched the change, they had not. past bug work shows the maintainer
stepping back, so upstream source is stagnant[0]. this was brought in a
past triage, but we can discuss again.

2056187
<https://bugs.launchpad.net/ubuntu/+source/initramfs-tools/+bug/2056187> :
initramfs-tools
    a comment from December showed a regression in a test run by MSFT.
However, the images referenced do not utilize initramfs-tools during boot.
Added a comment with the background, and asked if it couldn't be reproduced
with *only* that package installed (and overriding what's in the
UKI...which is an entire *thing*) the report shouldn't block SRU

2095203 <https://bugs.launchpad.net/netplan/+bug/2095203> : netplan.io

udev + lxc containers + netplan. the bug is very informative, and an
upstream PR is open.PR opened upstream. slyon likely already saw it
already, and maintenance of the upstream codebase lies on another team now.
don't want to lose visibility in the short term though.

2095584 <https://bugs.launchpad.net/ubuntu/+source/open-iscsi/+bug/2095584>
: open-iscsi and 2095600
<https://bugs.launchpad.net/ubuntu/+source/squid/+bug/2095600> : squid

both are related to python3.13 transition. ginggs has done initial triage
on the work. open-iscsi looks like a test-dependency issue. squid bug links
to an apparmor bug for the dead battery.[1] Can be worked around in squid,
but better to fix in apparmor long-term. adding both for discussion.

2096644  <https://bugs.launchpad.net/ubuntu/+source/run-one/+bug/2096644>:
run-one

feature request for run-one. run-one is an Ubuntu special, currently in
`main` as a dependency of `byobu`. Won't get into history, but the request
on the surface isn't a bad one. Being said, there's no active development
on run-one right now. There's more discussion we need around `byobu` and
`run-one` support in general.

2096676 </> : procmail

Dug around a bit, and I think this may be a behavioral change between
versions. Reading `man procmailex`, it actually seems like their past
recipe shouldn't have worked. the logs in the noble version show it
treating SENDER= as a variable assignment. previously, I'm assuming it was
being parsed as an action, in which case the | is appropriate. But reading
the docs, it seems like it should have always been backticks.

There are more bugs, but these took most of my day. I'll try to look at
more things that popped up tomorrow.


[0]: https://github.com/tats/w3m/issues/304
[1]: https://wiki.debian.org/Python/Dead%20Batteries

-- 
[image: Canonical-20th-anniversary]

John Chittum

Engineering Manager, Ubuntu Engineering, Server

Email: john.chittum at canonical.com

Location: Maine, USA

canonical.com

ubuntu.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.ubuntu.com/archives/ubuntu-server/attachments/20250127/833ca317/attachment.html>