[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#1020792: tech-ctte: Halt merged-/usr transition until dpkg filesystem damage bugs are fixed



Hi Zack,

Thanks for bringing this to the committee; even if Sean is correct that we won't act on this report, you've described the issues clearly and I think it was worth bringing to our attention.

On 26/09/2022 20:28, Zack Weinberg wrote:

It has been known for some time that dpkg has bugs which prevent it
from correctly handling merged-/usr systems.  #858331/#848622 is the
only such bug (that I can find) that has actually been recorded in the
BTS, and it *appears* to be a relatively harmless problem, affecting
only dpkg-query output.

This much is uncontroversial.

However, Simon Richter <sjr@debian.org> showed in
https://lists.debian.org/debian-devel/2021/08/msg00326.html that the > bad dpkg-query output is only the most obvious symptom of a much more
serious problem, and that, under conditions that will plausibly occur
in the archive after the release of bookworm (assuming all continues
as presently planned) upgrading packages on a merged-/usr system can
cause packaged files to disappear from the filesystem.  The files most
likely to be affected are the files that are currently packaged in
/bin, /sbin, and /lib, including, just to mention a few, /bin/bash,
/bin/systemd, and /lib/$ARCH/libc.so.6.  Thus, the dpkg bugs can
render systems unbootable on upgrade, and should be considered
critical severity.

This is a very useful message, and (at least to my mind) makes it clearer how more serious problems might well occur.

As Sean says, though, questions of what are and aren't RC bugs are typically the domain of the release team, not the TC.

I don't think you're asking us to revisit our decision on the approach taken to merged-/usr; we don't generally return to a decision once made (and a GR would normally be the approach to overturning a TC decision). Personally, I think there are circumstances where we might (e.g. a convincing argument that we missed something critical in our decision-making, or that circumstances have changed sufficiently to warrant another look), but I don't think we are in that situation here at the moment.

I think the best way to proceed would be to open a bug describing the problem that Simon outlines with RC severity; the relevant maintainers and release team can then discuss how to resolve the issues and if they warrant delaying the release or adjusting when we complete the transition. Obviously those people might want to ask the TC for advice, but I think that would be up to them at least in the first instance.

Thanks,

Matthew


Reply to: