Re: debootstrap/1.0.102 appears to break debuerreotype autopkgtest

[Hideki Yamane]

Hi,

On Thu, 14 Jun 2018 10:15:51 -0700
Tianon Gravi  wrote:
> > Instead of stretch simply defaulting to non-merged-usr, it's now
> > _blacklisted_ from merged-usr, even if I explicitly specify
> > "--merged-usr", right?  Is that the intended implementation here?

 Yes, since releases until stretch was already shipped without merged-usr,
 so it should be. But loose restriction for test purpose is okay, IMO.

 Question with 'EXTRACT_DEB_TAR_OPTIONS="$EXTRACT_DEB_TAR_OPTIONS -k"'
 It was introduced https://bugs.debian.org/838388 , so it should not be
 applied to all releases. However, I'm not sure which "older" release
 for it, especially whether it equals to merged-usr.

-- 
Regards,

 Hideki Yamane henrich @ debian.org/iijmio-mail.jp

Re: d-i repo at dillon

[Cyril Brulebois]

Hi,

Holger Wansing  (2018-06-14):
> > I also changed some lines from svn to git (./daily-build-aggregator
> > and ./daily-build-manager and ./daily-build-overview)
> 
> They fail, because on dillon, there is still a checkout of the svn
> repo.  So dillon needs an git checkout there.
> 
> Ok for me to do that?

Sure thing, and thanks!

Don't bother regarding the daily-build-logs.git repository (or whatever
it's called), it has gotten corrupted at some point, which I've never
found time to investigate, fix, or nuke. It might make sense to re-init
it to an empty repository at some point, so that further logs can be
injected in it. Just mentioning that so that you don't wonder too much,
not trying to put it in your plate.


Cheers,
-- 
Cyril Brulebois (k...@debian.org)
D-I release manager -- Release team member -- Freelance Consultant


signature.asc
Description: PGP signature

d-i repo at dillon

[Holger Wansing]

Hi,

Holger Wansing  wrote:
> > > 2.
> > > Above you mention "re-enable cron entries". However, I cannot find any 
> > > such
> > > disabled entries.
> > > I assume the l10n-sync process is currently enabled, or am I missing 
> > > something?
> > > (I'm planning to test the modified script via some weblate commits, that's
> > > why.)
> 
> I also changed some lines from svn to git
> (./daily-build-aggregator and ./daily-build-manager and 
> ./daily-build-overview)

They fail, because on dillon, there is still a checkout of the svn repo.
So dillon needs an git checkout there.

Ok for me to do that?


Holger


-- 

Created with Sylpheed 3.5.1 under 
D E B I A N   L I N U X   9   " S T R E T C H " .

Registered Linux User #311290 - https://linuxcounter.net/

Re: [l10n-sync] Updating sync script on git migration

[Holger Wansing]

Hi,

Cyril Brulebois  wrote:
> Hi,
> 
> Holger Wansing  (2018-06-12):
> > Cyril Brulebois  wrote:
> > > I think it would make sense to have you in the d-i gid, so that you can
> > > access dillon, and re-enable crontab entries, maybe subscribing to cron
> > > mails? Of course I could re-enable those crontab entries myself but I
> > > won't be able to fix any possible fallouts in the upcoming days.
> > 
> > 1.
> > I just subscribed to cron mails.
> 
> I've seen your commit but this isn't live yet. ;)

Now that I have access to dillon, they _are_live_ :-)

I have also changed the number of sublevels for l10n-sync to 5, which is the
current amount of sublevels.

> > 2.
> > Above you mention "re-enable cron entries". However, I cannot find any such
> > disabled entries.
> > I assume the l10n-sync process is currently enabled, or am I missing 
> > something?
> > (I'm planning to test the modified script via some weblate commits, that's
> > why.)

I also changed some lines from svn to git
(./daily-build-aggregator and ./daily-build-manager and ./daily-build-overview)


Holger

> Modifying the git repository isn't sufficient (as we might want to tweak
> the crontab entries without a commit; at least we've been doing that
> from time to time). We would need to update the copy on dillon in d-i's
> home directory; switching it from svn to git in the process. I can't
> really commit to doing so right away though. Note there's “make check”
> and “make install” available thanks to scripts/Makefile, to list the
> diff between what's in dillon's crontab and what's in $VCS.
> 
> > 3.
> > While logged into dillon, I found an old example-preseed.txt file under
> > /srv/d-i.debian.org/www/
> > which is an outdated file from squeeze. I tried to remove it, but lacking
> > permission.
> > Adding myself to d-i gid is missing here, probably?
> 
> Right. You should get at least this:
> | User kibi may run the following commands on dillon:
> | (d-i) ALL
> 
> which will let you do things like “sudo -u d-i -s”, which means working
> as the d-i user on dillon.
> 
> For this you'll likely need to open a ticket on RT:
>   https://wiki.debian.org/Teams/DSA/RTUsage
> 
> Feel free to cc me when requesting getting added to the d-i group, I'll
> confirm with a reply (which I think was kind of needed/appreciated last
> time I've been in that process).
> 
> Thanks for joining!
> 
> 
> Cheers,
> -- 
> Cyril Brulebois (k...@debian.org)
> D-I release manager -- Release team member -- Freelance Consultant


-- 

Created with Sylpheed 3.5.1 under 
D E B I A N   L I N U X   9   " S T R E T C H " .

Registered Linux User #311290 - https://linuxcounter.net/

Re: Re: debootstrap/1.0.102 appears to break debuerreotype autopkgtest

[Tianon Gravi]

On Thu, 14 Jun 2018 at 10:00, Tianon Gravi  wrote:
> Instead of stretch simply defaulting to non-merged-usr, it's now
> _blacklisted_ from merged-usr, even if I explicitly specify
> "--merged-usr", right?  Is that the intended implementation here?

If the attached patch is an OK direction, I'd be happy to submit it as
a merge request for easier merge/review. O:)

(sorry for missing that the previous discussion was around this issue
in debootstrap, not just that this is what debuerreotype needed to do
-- at a conference and email is a bit of a blur right now /o\)


♥,
- Tianon
  4096R / B42F 6819 007F 00F8 8E36  4FD4 036A 9C25 BF35 7DD4
diff --git a/debootstrap b/debootstrap
index 88781fc..8eca9ec 100755
--- a/debootstrap
+++ b/debootstrap
@@ -27,7 +27,7 @@ KEYRING=""
 DISABLE_KEYRING=""
 FORCE_KEYRING=""
 VARIANT=""
-MERGED_USR="yes"
+MERGED_USR=""
 ARCH=""
 HOST_ARCH=""
 HOST_OS=""
diff --git a/scripts/debian-common b/scripts/debian-common
index c5a5a39..9517af8 100644
--- a/scripts/debian-common
+++ b/scripts/debian-common
@@ -37,12 +37,11 @@ work_out_debs () {
 first_stage_install () {
 	case "$CODENAME" in
 		etch*|lenny|squeeze|wheezy|jessie*|stretch)
-			;;
-		*)
-			EXTRACT_DEB_TAR_OPTIONS="$EXTRACT_DEB_TAR_OPTIONS -k"
-			setup_merged_usr
+			[ -z "$MERGED_USR" ] && MERGED_USR="no"
 			;;
 	esac
+	EXTRACT_DEB_TAR_OPTIONS="$EXTRACT_DEB_TAR_OPTIONS -k"
+	setup_merged_usr
 
 	extract $required
 
diff --git a/scripts/gutsy b/scripts/gutsy
index cea5d7b..9fd3ea1 100644
--- a/scripts/gutsy
+++ b/scripts/gutsy
@@ -68,12 +68,11 @@ work_out_debs () {
 first_stage_install () {
 	case "$CODENAME" in
 		gutsy|hardy|intrepid|jaunty|karmic|lucid|maverick|natty|oneiric|precise|quantal|raring|saucy|utopic|vivid|wily|yakkety|zesty|artful|bionic|cosmic)
-			;;
-		*)
-			EXTRACT_DEB_TAR_OPTIONS="$EXTRACT_DEB_TAR_OPTIONS -k"
-			setup_merged_usr
+			[ -z "$MERGED_USR" ] && MERGED_USR="no"
 			;;
 	esac
+	EXTRACT_DEB_TAR_OPTIONS="$EXTRACT_DEB_TAR_OPTIONS -k"
+	setup_merged_usr
 
 	extract $required
 

Re: Re: debootstrap/1.0.102 appears to break debuerreotype autopkgtest

[Tianon Gravi]

On Thu, 14 Jun 2018 at 07:43, Tianon Gravi  wrote:
> Update is dput now -- should show up (and get re-tested) soonish. :)

It just dawned on me that this upload possibly won't actually fix the failure.

Instead of stretch simply defaulting to non-merged-usr, it's now
_blacklisted_ from merged-usr, even if I explicitly specify
"--merged-usr", right?  Is that the intended implementation here?

(after all, the "usrmerge" package still exists, so IMO if I specify
"--merged-usr" it should still work, but barring that, it seems like
it should at least be a fatal error or even a warning if something
I've explicitly asked debootstrap to do for me is blacklisted from
working)

♥,
- Tianon
  4096R / B42F 6819 007F 00F8 8E36  4FD4 036A 9C25 BF35 7DD4

Bug#901551: unblock: libgcrypt20/1.8.3-1

[Andreas Metzler]

Package: release.debian.org
Severity: normal
User: release.debian@packages.debian.org
Usertags: unblock

Please unblock package libgcrypt20, this is a upstream bugfix/security
release of the stable branch, fixing CVE-2018-0495.
https://www.nccgroup.trust/us/our-research/technical-advisory-return-of-the-hidden-number-problem/

unblock libgcrypt20/1.8.3-1

thanks, cu Andreas


1.8.2-2_to_1.8.3-1.diff.xz
Description: application/xz


signature.asc
Description: PGP signature

Re: Re: debootstrap/1.0.102 appears to break debuerreotype autopkgtest

[Tianon Gravi]

On Thu, 14 Jun 2018 at 07:31, Tianon Gravi  wrote:
> Ah, which is exactly what I did in [1], but the 0.7 release isn't
> uploaded to Debian yet -- I'll do that ASAP to fix this regression. :)

Update is dput now -- should show up (and get re-tested) soonish. :)

Thanks! (and thanks for your work on debootstrap)

♥,
- Tianon
  4096R / B42F 6819 007F 00F8 8E36  4FD4 036A 9C25 BF35 7DD4

Re: Re: debootstrap/1.0.102 appears to break debuerreotype autopkgtest

[Tianon Gravi]

On Thu, 14 Jun 2018 at 02:23, Hideki Yamane  wrote:
> > debootstrap should default to non-merged-usr for stretch, but it should
> > be possible to enable merged-usr via the command-line parameter to avoid
> > the regression in debuerreotype.
>  ...
>  We can use "--no-merged-usr" and "--merged-usr" options explicitly.

Ah, which is exactly what I did in [1], but the 0.7 release isn't
uploaded to Debian yet -- I'll do that ASAP to fix this regression. :)

[1]: https://github.com/debuerreotype/debuerreotype/pull/35


♥,
- Tianon
  4096R / B42F 6819 007F 00F8 8E36  4FD4 036A 9C25 BF35 7DD4

Re: Re: debootstrap/1.0.102 appears to break debuerreotype autopkgtest

[Hideki Yamane]

Hi,

> The patch for #839046 also disabled --merged-usr for stretch as stretch
> was added to the blacklist in first_stage_install().
> 
> debootstrap should default to non-merged-usr for stretch, but it should
> be possible to enable merged-usr via the command-line parameter to avoid
> the regression in debuerreotype.

 Stretch is blacklisted (in scripts/debian-common) and I've tested on my
 box, it is non-merged-usr environment. Is it stretch?

 We can use "--no-merged-usr" and "--merged-usr" options explicitly.


-- 
Regards,

 Hideki Yamane henrich @ debian.org/iijmio-mail.jp

Re: debootstrap/1.0.102 appears to break debuerreotype autopkgtest

[Ansgar Burchardt]

Hi,

Paul Gevers writes:
> I looked at the test¹ and it compares the result of the current run of
> debuerreotype with a stored hash. Luckily debuerreotype use diffoscope
> to investigate the delta. It seems that debuerreotype is hit by this
> change in debootstrap:
>
>   * Enable merged-/usr by default (Closes: #839046)
> This is applied for buster and later.
>
> I am not sure if this should NOT have let to a change in debuerreotype,
> as I believe that is testing stretch.

>From the test log:

│ -lrwxrwxrwx   0000 2017-01-01 00:00:00.00 bin -> 
usr/bin
│ +drwxr-xr-x   0000 2017-01-01 00:00:00.00 bin/
│ +-rwxr-xr-x   000  1099016 2016-11-15 18:49:00.00 bin/bash

The patch for #839046 also disabled --merged-usr for stretch as stretch
was added to the blacklist in first_stage_install().

debootstrap should default to non-merged-usr for stretch, but it should
be possible to enable merged-usr via the command-line parameter to avoid
the regression in debuerreotype.

Ansgar

debootstrap/1.0.102 appears to break debuerreotype autopkgtest

[Paul Gevers]

Dear maintainers,

With a recent upload of debootstrap the autopkgtest of debuerreotype
version 0.6-1 started to fail. See:
https://ci.debian.net/packages/d/debuerreotype/
and
https://qa.debian.org/excuses.php?package=debootstrap

I looked at the test¹ and it compares the result of the current run of
debuerreotype with a stored hash. Luckily debuerreotype use diffoscope
to investigate the delta. It seems that debuerreotype is hit by this
change in debootstrap:

  * Enable merged-/usr by default (Closes: #839046)
This is applied for buster and later.

I am not sure if this should NOT have let to a change in debuerreotype,
as I believe that is testing stretch.

On top of that, I wonder if this test is sensitive to changes in the
security archive.

Currently this regression is delaying the migration of debootstrap to
testing by 13 days. Could we please discuss together what the best way
forward is for this regression in testing? (Just so you know, I am
empowered to have the migration software ignore the results of this
specific test if we all agree this isn't a regression that should block
debootstrap).

More information about this email and the reason of it can be found on
https://wiki.debian.org/ContinuousIntegration/RegressionEmailInformation

Paul

¹
https://github.com/debuerreotype/debian-debuerreotype/blob/master/debian/tests/stretch



signature.asc
Description: OpenPGP digital signature