Your message dated Sat, 18 May 2019 09:15:21 -0700
with message-id <[email protected]>
and subject line Re: Bug#923732: dgit-infrastructure:
--deliberately-include-questionable-history required for every single push
has caused the Debian Bug report #923732,
regarding dgit-infrastructure: --deliberately-include-questionable-history
required for every single push
to be marked as done.
This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.
(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact [email protected]
immediately.)
--
923732: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=923732
Debian Bug Tracking System
Contact [email protected] with problems
--- Begin Message ---
Package: dgit-infrastructure
Version: 8.4
Hello,
swayidle was not actually ever rejected from NEW but was instead dcut
from DEFERRED and uploaded again a few times.
Shouldn't --deliberately-include-questionable-history be required only
once? Shouldn't dgit-infrastructure learn that the commits it thinks
are tainted aren't actually tainted?
Total 47 (delta 8), reused 19 (delta 0)
remote: gpgv: Signature made Mon 04 Mar 2019 05:36:37 PM UTC
remote: gpgv: using RSA key
9B917007AE030E36E4FC248B695B7AE4BF066240
remote: gpgv: Good signature from "Sean Whitton <[email protected]>"
remote: gpgv: aka "Sean Whitton <[email protected]>"
remote: gpgv: aka "Sean Whitton <[email protected]>"
remote: gpgv: aka "Sean Whitton <[email protected]>"
remote: gpgv: aka "Sean Whitton <[email protected]>"
remote: To
/srv/dgit.debian.org/dispatch-dir/distro=debian/repos/_tmp/swayidle_prospective
remote: * [new branch] a719af136406276be15bd51270c2c36cf42575cc ->
refs/dgit/sid
remote: * [new tag] e4a96dfac0f651a6f46a2a255dfda3e806c6115a ->
archive/debian/1.2-2
remote: * [new tag] a1d2c059f0157a8e307b9202378d7466dd0015ab ->
debian/1.2-2
remote: To
/srv/dgit.debian.org/dispatch-dir/distro=debian/repos/_tmp/swayidle_prospective
remote: * [new branch] a719af136406276be15bd51270c2c36cf42575cc -> master
remote:
remote: History contains tainted commit 2ddf9061f6aba7ee03c7da1d5dec6a4225c2e77e
remote: Taint recorded at time 2019-02-09 21:39:02 Z for package swayidle
remote: Reason: tag archive/debian/1.2-1 referred to this object in git tree
but all previously pushed versions were found to have been removed from NEW
(ie, rejected) (or never arrived)
remote: Could be forced using --deliberately. Consult documentation.
remote:
remote: Rejecting push due to questionable history.
dgit-repos-server: policy hook failed (or rejected) (256)
remote: dgit-repos-server: policy hook failed (or rejected) (256)
To git+ssh://push.dgit.debian.org/dgit/debian/repos/swayidle.git
! [remote rejected] a719af136406276be15bd51270c2c36cf42575cc -> refs/dgit/sid
(pre-receive hook declined)
! [remote rejected] archive/debian/1.2-2 -> archive/debian/1.2-2 (pre-receive
hook declined)
! [remote rejected] debian/1.2-2 -> debian/1.2-2 (pre-receive hook declined)
error: failed to push some refs to
'git+ssh://[email protected]/dgit/debian/repos/swayidle.git'
dgit: failed command: git -c push.followTags=false push
'git+ssh://[email protected]/dgit/debian/repos/swayidle.git'
a719af136406276be15bd51270c2c36cf42575cc:refs/dgit/sid
refs/tags/archive/debian/1.2-2 refs/tags/debian/1.2-2
--
Sean Whitton
signature.asc
Description: PGP signature
--- End Message ---
--- Begin Message ---
Hello Ian,
On Fri 22 Mar 2019 at 05:05PM +00, Ian Jackson wrote:
> I have spent a while trying to reproduce this. I found that there is
> indeed code which is supposed to remove taints when they are
> overridden.
>
> It wasn't tested at all. But I wrote a sketch of a test, which I
> thought would resemble what happened to this package, and the
> mechanism seems to work. See attached.
>
> Can you tell me the exact sequence of events up to the first
> unwarranted push rejection, with times where you have them ?
I just sponsored the 1.3-1 upload of swayidle, and was not required to
pass any --deliberately option. So it seems that the code to remove
taints worked this time.
Below, I include some notes I made previously as to the sequence of
events that led to the output quoted in the first message of this bug
report. I'm sorry they are not more detailed.
I'm going to go ahead and close the bug since it is no longer
reproducible. I do think that there is something wrong with the taint
removal code, but it doesn't seem we're going to uncover it for the time
being.
1) `dgit push --delayed=N` swayidle 1.2-1
2) Messing about with DELAYED, which involved (at least) two dgit
pushes, with --delayed, of swayidle 1.2-1. Each of these required
--deliberately-include-questionable-history, but that is correct. I
don't have a useful record of this.
3) swayidle 1.2-1 finally hits NEW and is later ACCEPTed.
4) A few days later, I attempt to dgit push swayidle 1.2-2 and receive
the unwarranted rejection, quoted in my mail opening this bug report.
I didn't ask my sponsee to use new version numbers, I think because
DELAYED is the twilight zone, but perhaps that would have been wise.
I note that my sponsee uses a git workflow that dgit does not yet
support (only debian/ in git), so I had to use `dgit import-dsc` in each
case.
--
Sean Whitton
signature.asc
Description: PGP signature
--- End Message ---
