Bug 316453 – Stable =sys-apps/parted-2.3

• Status: RESOLVED FIXED
• Product: Gentoo Linux

[Description Akkara 2010-04-21 04:48:05 UTC]

The existing stable version of parted, sys-apps/parted-1.9.0, seems to have trouble dealing with the newer 4K sector hard disks.  Specifically, once one gets to extended partitions, it seems to insist on aligning them on cylinder boundaries - which aren't 4K aligned.

I tried keywording sys-apps/parted-2.2 and this supports 4K aligned partitions much better -- or at least it doesn't try to fight me when I ask for something. I had already manually worked out all the needed boundaries down to the byte address while wrestling with 1.9, and I didn't check to see whether it additionally supports a 'round to 4K' option which would make things easier.


Reproducible: Always

[Comment 1 Thomas Kahle (RETIRED) 2010-04-21 05:00:21 UTC]

Seems to have no serious open bugs and is in the tree long enough.

[Comment 2 Andreas Kinzler 2010-04-22 07:57:41 UTC]

current stable version is severely broken, see http://gparted-forum.surf4.info/viewtopic.php?id=13777

[Comment 3 Christian Faulhammer (RETIRED) 2010-04-22 11:56:01 UTC]

Created attachment 228745 [details]
test log

Fails test with USE="-debug"

[Comment 4 Jeroen Roovers (RETIRED) 2010-04-22 14:11:25 UTC]

(In reply to comment #3)
> Fails test with USE="-debug"

USE=debug is recommended by upstream (bug #304231).

[Comment 5 Christian Faulhammer (RETIRED) 2010-04-22 14:21:07 UTC]

(In reply to comment #4)
> (In reply to comment #3)
> > Fails test with USE="-debug"
> 
> USE=debug is recommended by upstream (bug #304231).

 Then test should bail out or be restricted if debug is not set.

[Comment 6 Markos Chandras (RETIRED) 2010-05-16 19:26:27 UTC]

amd64 will not process this until the test restriction is fixed

[Comment 7 Jeroen Roovers (RETIRED) 2010-05-17 11:35:03 UTC]

The 2.2 ebuild now checks for USE=debug or doesn't run the test suite.

[Comment 8 Andreas Schürch 2010-05-18 11:19:35 UTC]

I get a sandbox violation on x86 with it!?

--------------------------- ACCESS VIOLATION SUMMARY ---------------------------
LOG FILE "/var/log/sandbox/sandbox-23322.log"

VERSION 1.0
FORMAT: F - Function called
FORMAT: S - Access Status
FORMAT: P - Path as passed to function
FORMAT: A - Absolute Path (not canonical)
FORMAT: R - Canonical Path
FORMAT: C - Command Line

F: linkat
S: deny
P: .libs/libfs.lax/lt1-interface.o
A: /var/tmp/portage/sys-apps/parted-2.2/work/parted-2.2/libparted/fs/.libs/libfs.lax/lt1-interface.o
R: /var/tmp/portage/sys-apps/parted-2.2/work/parted-2.2/libparted/fs/.libs/libfs.lax/lt1-interface.o
C: ln .libs/libfs.lax/libext2.a/interface.o .libs/libfs.lax/lt1-interface.o
--------------------------------------------------------------------------------

[Comment 9 Jeroen Roovers (RETIRED) 2010-05-18 13:26:04 UTC]

(In reply to comment #8)
> I get a sandbox violation on x86 with it!?

Then please file a new bug report.

[Comment 10 Paweł Hajdan, Jr. (RETIRED) 2010-05-18 15:29:19 UTC]

x86 stable, I didn't get any sandbox violation and the test suite passed both with debug and -debug.

[Comment 11 Jeroen Roovers (RETIRED) 2010-05-19 18:28:30 UTC]

More trouble. :-\ Maybe it's a good idea to revert x86 stabilisation.

[Comment 12 Alexey 2010-05-20 11:35:17 UTC]

In my case due configuration of sys-apps/parted-2.2 sandbox detect VIOLATION to write in /dev/dri/card0
So, I should add SANDBOX_WRITE="/dev/dri/card0" in /etc/sandbox.conf, and then parted-2.2 compiles OK.
But now gparted-0.4.5 can't compiles, it crash at configure phase.
Gentoo x86 (stable)

[Comment 13 Jeroen Roovers (RETIRED) 2010-05-20 12:21:09 UTC]

(In reply to comment #8)
> I get a sandbox violation on x86 with it!?

File a new bug report and stop responding to this one, please.

(In reply to comment #12)
> In my case due configuration of sys-apps/parted-2.2 sandbox detect VIOLATION to
> write in /dev/dri/card0
> So, I should add SANDBOX_WRITE="/dev/dri/card0" in /etc/sandbox.conf, and then
> parted-2.2 compiles OK.
> But now gparted-0.4.5 can't compiles, it crash at configure phase.
> Gentoo x86 (stable)

CC yourself on that bug report or file a duplicate bug report, but please stop responding to this bug report.

[Comment 14 Thomas Kahle (RETIRED) 2010-06-29 16:00:32 UTC]

Created attachment 236929 [details]
test failures on x86 with 2.3

Hi, tests fail on x86.

[Comment 15 Jeroen Roovers (RETIRED) 2010-07-01 17:39:05 UTC]

(In reply to comment #14)
> Created an attachment (id=236929) [details]
> test failures on x86 with 2.3
> 
> Hi, tests fail on x86.

Nothing to do with the architecture, I guess. Maybe you don't have the required filesystem support in your kernel.

FAIL: t4100-msdos-partition-limits.sh (exit: 1)
FAIL: t4100-dvh-partition-limits.sh (exit: 1)

Removed those from the test suite.

I've also changed src_test() to not die on `emake check' anymore. All the failing tests I have removed so far required missing kernel/userland features to work, so that's always an upstream issue and not ours.

Arch teams, please proceed.

[Comment 16 Thomas Kahle (RETIRED) 2010-07-02 07:55:06 UTC]

Ok, archtesting program complete on x86. Looks good to go together with {g,py}parted in bugs 316537, 316539.

[Comment 17 Markos Chandras (RETIRED) 2010-07-04 08:29:11 UTC]

amd64 done

[Comment 18 Jeroen Roovers (RETIRED) 2010-07-06 14:54:41 UTC]

Stable for HPPA.

[Comment 19 Christian Faulhammer (RETIRED) 2010-07-09 13:30:00 UTC]

stable x86, thanks Thomas

[Comment 20 Tobias Klausmann (RETIRED) 2010-07-12 09:25:25 UTC]

Stable on alpha.

[Comment 21 Markus Meier 2010-07-21 18:59:14 UTC]

arm stable

[Comment 22 Raúl Porcel (RETIRED) 2010-08-01 10:55:15 UTC]

ia64/s390/sh/sparc stable

[Comment 23 Joe Jezak (RETIRED) 2010-08-11 21:41:02 UTC]

Marked ppc/ppc64 stable. Closing since we're the last arch.