[Piuparts-devel] Bug#918379: please decide: severity of "fails to purge" issues

[Holger Levsen]

package: release.debian.org
x-debbugs-cc: debian-policy at lists.debian.org, piuparts-devel at lists.alioth.debian.org, Adrian Bunk <bunk at debian.org>

Hi,

filing this as a bug now. Please reassign to src:piuparts once you have
decided...

On Sat, Jan 05, 2019 at 03:34:23PM +0000, Holger Levsen wrote:
> in https://github.com/anbe42/piuparts/commit/283dac3ae7e31fee51efb836468cd8ca5b61584f
> (not yet merged into the main piuparts repo) Andreas proposes to file bugs
> regarding failing to purge with severity 'serious" because "old bugs are
> filed/fixed and any failure due to a regression in sid will block migration
> anyway", while we used to treat 'failing to purge' bugs as severity
> 'important' as in practice those bugs are merely annoying (while being a
> clear policy violation).
> 
> There's also at least #918312 filed by Adrian Bunk.
> 
> The reasoning that these bugs will block migrations anyway sounds sound
> - except for new packages though!
> 
> So I would like to have the opinion of the release team if you also
> think that those bugs should be filed with severity 'serious' nowadays.
> (As it was their opinion that this shouldn't be done previously.)
> 
> What do you think?

this led to this discussion on #-release:

<ivodd>  h01ger: what does 'fails to purge' mean? the purge fails (gives
         errors) or the purge 'succeeds', but files are left?
<h01ger> both
<ivodd>  and with 'both' you mean: 'either of these will produce that error'?
<h01ger> yes. https://piuparts.debian.org/templates/mail/ lists 5 cases with purge failures
<ivodd>  personally, if the purge command fails, I wouldn't even hesitate to\
         call the serious (I'm surprised it wasn't filed as serious before)
<ivodd>  if the command succeeds and files are left over, I might hesitate a
         bit, but on the other hand, if you can't remove all the files, you
	 have no right to call that a 'successful' purge
<ivodd>  so I wouldn't mind if that was filed as serious as well
<h01ger> usually purges are done manually so failing to purge with exit 1 doesnt
         really have an effect. thats why it was important only
<ivodd>  well, it means something is wrong, and the user gets to clean it up manually
<ivodd>  but I don't think we disagree here :)
<h01ger> :)
<ivodd>  I don't really like adding new cases for RC bugs just before the freeze,
         but if this type of error has been tested for quite some time, then it
	 isn't really a 'new' case
<h01ger> ivodd: i think i will reply to the bug via submit at bugs.d.o (assigned to release.d.o) and include our replies there
<h01ger> yeah
<ivodd>  ok
<h01ger> it has been tested since lenny :)
<ivodd>  piuparts++
<h01ger> :) thanks!


-- 
cheers,
	Holger

-------------------------------------------------------------------------------
               holger@(debian|reproducible-builds|layer-acht).org
       PGP fingerprint: B8BF 5413 7B09 D35C F026 FE9D 091A B856 069A AA1C
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: not available
URL: <http://alioth-lists.debian.net/pipermail/piuparts-devel/attachments/20190105/dec8d3c6/attachment.sig>