Bug#462835: Debian bug #462835 grub-pc: alloc magic is broken
On Tue, Aug 19, 2008 at 2:34 AM, Felix Zielcke [EMAIL PROTECTED] wrote: Am Montag, den 18.08.2008, 23:06 -0400 schrieb Bharath Ramesh: Sorry for a late reply. No problem. I do not get the alloc magic is broken error any more with grub2 and lvm.mod. Grub2 never works if I have lvm.mod included. What do you mean with that? What is GRUB telling you? Does it go to rescue mode without showing any error message or do does it hang? Before updating to 1.96+20080724-8 version it would just hang with it reading Welcome to Grub Please try the freshly new uploaded 1.96+20080724-8 version. There was a bug in the LVM module that the error variable was given back to the upper layer so it goes to rescue mode even if it shouldn't in that case. The freshly uploaded 1.96+20080724-8 version I end up with the rescue mode. I have to manually load my kernel using the command line. I hope this helps. Will do my best to provide you all the information I can. I ran grub-emu with my grub.cfg which I have attached with this email. I get the following error when I use the grub.cfg in grub-emu error: invalid arch independent ELF magic I have attached grub.cfg along with this email. grub.cfg_broken gives the above error while grub.cfg_working is the only working config file to make my laptop boot. This is normal grub-emu doestn't support modules at all but lvm.mod is compiled in, so you don't need to load it to use your lvm devices with e.g. `ls' command. Once again I am sorry for this late reply. Once again no problem :) A `late' reply like yours is better then getting none at all for months. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#495717: heimdal-kdc: Heimdal KDC with LDAP backend not pointing properly to slapd socket
Juan Miguel Corral wrote: I am installing a Heimdal KDC with OpenLDAP backend, and I have found that when i try to use kadmin, I get following error: kadmin: hdb_open: ldap_sasl_bind_s: Can't contact LDAP server Permissions are right. The problem is that heimdal is trying to connect to LDAP socket in /var/run/slapd/ldapi, but slapd creates it in /var/run/ldapi. The problem gets fixed just with a symlink: ln -s /var/run/ldapi /var/run/slapd/ldapi And then I can use kadmin normally. But I think the problem should be fixed before release. This isn't a Heimdal bug, it is an LDAP issue. Heimdal uses the ldap libraries which have the path hard coded. In libldab-2.4 they changed the path from /var/run/ldapi to /var/run/slapd/ldapi I can only assume you are using an old version of slapd, as the version I have on Etch and Lenny will create the symlink automatically. I suspect they will use the new path by default too. Just to confirm: what version of slapd do you have? Brian May -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#355883: Same problem here
Hi Adam Adam Cécile (Le_Vert) wrote: Bastian Blank a écrit : On Tue, Aug 19, 2008 at 06:50:11PM +0200, Adam Cécile (Le_Vert) wrote: I can confirm this bug 2.6.26-1-amd64 and a slightly modified rebuild of the i386 flavour. My usb stick is unusable. I doubt that you see the same bug. Try #494800. Check reply by Kumar Appaiah (20080812). He re-opened this bug against kernel 2.6.26. I noticed severity has been dropped from important to normal. Any justification ? Someone which is not submitter or maintainer changed it. I'd rather say it's critical. Everybody needs usb sticks No. Many of my machines don't even have usb. No comment. Bastian I found the attached patch somewhere on lkml. It's targeted for kernel 2.6.27. I can confirm this patch fix the issue with my stick. This patch hasn`t taken to Linus`s tree yet but has been queued by Greg K-H, guess we`ll see it merged until 2.6.27. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#427492: grub-efi: method to 'bless' ?
Hi, I had the unfortunate incident that MacBook HDD blew up while at Debconf, and without a Mac OS X installation disk, I could not set up refit, and I had to resort to PC compatibility mode. I didn't know about efibootmgr, it might be what I was looking for, but it doesn't seem to work with default lenny kernel that I have installed right now. hmmm Hi, efibootmgr can only be used when boot from efi, it's not available in bios boot mode. BTW, it seems the latest version of refit can be burned as a bootable CD, perhaps you can try that. I'm currently on an extended trip and not easy for me to check it out, and I'm not booting off EFI. Could someone with a working system confirm this? regards, junichi -- [EMAIL PROTECTED],debian.org} -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#495729: unifont: Hangul glyphs use fonts with ambiguous licensing
Package: unifont Version: 19990430-1 Severity: important The unifont package has always used Hangul Syllables glyphs generated from Hanterm fonts. Versions of unifont up to 1:1.0-4 also used glyphs from Hanterm fonts in the U+1100..U+11FF range. According to Debian Developer Changwoo Ryu, maintainer of bf-utf-source, xfonts-baekmuk, and other Debian packages, the Hanterm fonts were incorrectly listed in FreeBSD and elsewhere as being covered by the GNU Public License. Changwoo has told me that parts of Hanterm are covered under the GPL, but not the fonts. Concern over ambiguity in the terms of Hanterm font licensing resulted in their not being included in the hanterm Debian package. Changwoo Ryu removed the existing Hangul glyphs from the file unifont.bdf in his bf-utf-source package, replacing them with Hangul Syllables glyphs in the public domain. The most expedient workaround to the Hanterm fonts license ambiguity for the unifont package is to incorporate Changwoo's Hangul Syllables block from the bf-utf-source Debian package. Paul Hardy -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#495503: lazygal: order of pictures in a folder?
Ian There should be an option (command line or config file) to select
Ian sorting by name, filesystem time or exif time. Name should
Ian probably be the default.
Here's a patch that dose this, plus another nagging issue: passing
optimize and progressive flags to the JPEG encoder. Note that
the patch is cumulative vs. 0.4; i.e. it includes your 0x0 patch.
With these things in, I find lazygal almost usable :-)
Now, to make it _really_ stand out of the crowd, parallelize.
Number of threads should probably be another option, like make's -jn.
diff -u --recur lazygal/lazygal/generators.py lazygal-0.4/lazygal/generators.py
--- lazygal/lazygal/generators.py 2008-07-03 16:14:15.0 -0700
+++ lazygal-0.4/lazygal/generators.py 2008-08-19 20:28:33.0 -0700
@@ -20,6 +20,8 @@
import locale
import Image
+# lazygal has her own ImageFile class, so avoid trouble
+import ImageFile as PILFile
import genshi
import __init__
@@ -42,9 +44,14 @@
class ImageOriginal(make.FileCopy):
-def __init__(self, dir, source_image):
+def __init__(self, dir, source_image, size_name=None):
self.dir = dir
-self.path = os.path.join(self.dir.path, source_image.filename)
+
+dest_name = source_image.filename
+if size_name:
+dest_name = self.dir.album._add_size_qualifier(dest_name, size_name)
+
+self.path = os.path.join(self.dir.path, dest_name)
make.FileCopy.__init__(self, source_image.path, self.path)
def build(self):
@@ -77,6 +84,7 @@
'info')
self.dir.album.log((%s) % self.osize_path)
+
im = Image.open(self.source_image.path)
# Use EXIF data to rotate target image if available and required
@@ -85,9 +93,19 @@
im = im.rotate(rotation)
im.thumbnail(self.size, Image.ANTIALIAS)
-
-im.save(self.osize_path, quality = self.dir.album.quality)
-
+
+calibrated = False
+while not calibrated:
+try:
+im.save(self.osize_path, quality = self.dir.album.quality, **self.dir.album.save_options)
+except IOError, e:
+if str(e).startswith('encoder error'):
+PILFile.MAXBLOCK = 2 * PILFile.MAXBLOCK
+continue
+else:
+raise
+calibrated = True
+
class WebalbumPicture(make.FileMakeObject):
@@ -217,16 +235,18 @@
self.image = image_file
WebalbumPage.__init__(self, dir, size_name, self.image.name)
-self.add_dependency(ImageOtherSize(self.dir,
- self.image,
- self.size_name))
+if self.dir.album.browse_sizes[size_name] == (0, 0):
+self.add_dependency(ImageOriginal(self.dir, self.image))
+else:
+self.add_dependency(ImageOtherSize(self.dir,
+ self.image,
+ self.size_name))
+if self.dir.album.original:
+self.add_dependency(ImageOriginal(self.dir, self.image))
# Depends on source directory in case an image was deleted
self.add_dependency(self.dir.source_dir)
-if self.dir.album.original:
-self.add_dependency(ImageOriginal(self.dir, self.image))
-
self.page_template = self.dir.album.templates['browse.thtml']
self.add_file_dependency(self.page_template.path)
@@ -496,7 +516,14 @@
self.add_dependency(WebalbumArchive(self))
def prepare(self):
-self.images.sort(lambda x, y: x.compare_date_taken(y))
+if self.album.sort_by == 'exif':
+self.images.sort(lambda x, y: x.compare_date_taken(y))
+elif self.album.sort_by == 'mtime':
+self.images.sort(lambda x, y: x.compare_mtime(y))
+elif self.album.sort_by == 'name':
+self.images.sort(lambda x, y: x.compare_filename(y))
+else:
+pass
# chain images
previous_image = None
@@ -659,7 +686,8 @@
class Album:
def __init__(self, source_dir, thumb_size, browse_sizes,
- quality=85, thumbs_per_page=0, dirzip=False):
+ optimize=False, progressive=False,
+ quality=85, thumbs_per_page=0, dirzip=False, sort_by='name'):
self.set_logging()
self.source_dir = os.path.abspath(source_dir)
@@ -675,6 +703,12 @@
self.original = False
self.thumbs_per_page = thumbs_per_page
self.dirzip = dirzip
+self.save_options = {}
+if optimize:
+self.save_options['optimize'] = True
+if progressive:
+self.save_options['progressive'] = True
+self.sort_by = sort_by
def set_theme(self, theme='default', default_style=None):
self.theme = theme
@@ -760,6 +794,10 @@
if
Bug#495730: ldd check when building vlc
Package: vlc Version: 0.8.6.h-1 Severity: normal The check to find if a vlc-nox plugin is linked with libX11 uses ldd. This will check the dependencies using system path by default - thus using /usr/lib/libvlc* instead of the newly built libvlc. A possible patch is the following: if LD_LIBRARY_PATH=$(CURDIR)/debian/tmp/usr/lib ldd $$file | grep -q libX11; then \ please note the use of LD_LIBRARY_PATH. A side-effect is that ldd, under fakeroot, will exhibit a warning saying it will not take into account symbols from libfakeroot-sysv, which fine IMO. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#493951: noting spaces that libpkg-guide may be improved
Steve is being unhelpful by not providing any patch for this RC bug he's filing. I think he's talking about a symptom not a root cause; binary incompatibility should really be addressed. I seem to recall changing libpkg-guide to address symbol versioning some time ago, and Steve should re-read the final version to point out what's missing from it. ... and I've sent the mail half-way through. I think references to 'ongoing curl transition pain' thread in debian-release would be an interesting reference to those people who are considering changing the -dev package in Debian. regards, junichi -- [EMAIL PROTECTED],debian.org} -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#493951: noting spaces that libpkg-guide may be improved
Hi, Steve is being unhelpful by not providing any patch for this RC bug he's filing. I think he's talking about a symptom not a root cause; binary incompatibility should really be addressed. I seem to recall changing libpkg-guide to address symbol versioning some time ago, and Steve should re-read the final version to point out what's missing from it. regards, junichi -- [EMAIL PROTECTED],debian.org} -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#495731: Spurious(?) character in pod makes manpage ugly.
Package: mr Version: 0.33 Severity: wishlist It seems to me that this is a typo. It results in an ugly manpage heading, SYNOPSIS / mr [options] checkout. diff -ud /usr/bin/mr /tmp/buffer-content-5627xv6 --- /usr/bin/mr 2008-07-23 05:21:36.0 +1000 +++ /tmp/buffer-content-5627xv6 2008-08-20 15:30:25.301115281 +1000 @@ -7,7 +7,7 @@ mr - a Multiple Repository management tool =head1 SYNOPSIS -/ + Bmr [options] checkout Bmr [options] update Diff finished. Wed Aug 20 15:30:25 2008 -- System Information: Debian Release: lenny/sid APT prefers testing APT policy: (990, 'testing') Architecture: i386 (i686) Kernel: Linux 2.6.25-2-686 (SMP w/1 CPU core) Locale: LANG=en_AU.utf8, LC_CTYPE=en_AU.utf8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash mr depends on no packages. mr recommends no packages. Versions of packages mr suggests: pn bzrnone(no description available) pn cvsnone(no description available) ii darcs 2.0.2-2 an advanced revision control syste ii git-core 1:1.5.6.3-1.1 fast, scalable, distributed revisi ii mercurial 1.0.1-3 Scalable distributed version contr pn subversion none(no description available) -- no debconf information -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#495322: [pkg-boost-devel] Bug#495322: Bug#495322: strange error :-) please edit subject
On Sun, Aug 17, 2008 at 10:36:57AM +0200, Erik Thiele wrote:
Am Samstag, 16. August 2008 16:08:15 schrieb Steve M. Robbins:
On Sat, Aug 16, 2008 at 10:29:47AM +0200, Erik Thiele wrote:
Note that Boost 1.34.1 is the old version of Boost. Version 1.35 is
packaged for Debian, and Boost 1.36 was recently released (Debian
packages are in progress).
This code compiles without warnings using libboost-date-time1.35-dev.
pn libboost-date-time-dev keine (keine
Beschreibung vorhanden)
rc libboost-date-time1.33.1 1.33.1-10 set of
date-time libraries based on generic
rc libboost-date-time1.34.1 1.34.1-11 set of
date-time libraries based on generic
ii libboost-date-time1.35-dev 1.35.0-5 set of
date-time libraries based on generic
ii libboost-date-time1.35.0 1.35.0-5 set of
date-time libraries based on generic
ii g++4:4.3.1-2 The GNU C++ compiler
this is my versions. and i get the presented results. i checked it again.
The only difference apparent on my system is the compiler:
[EMAIL PROTECTED] 495322.cc
#include boost/date_time/posix_time/ptime.hpp
void foo(boost::posix_time::ptime x)
{
x.date();
}
[EMAIL PROTECTED] -Werror -c -Wall 495322.cc
[EMAIL PROTECTED] --version
gcc (Debian 4.3.1-9) 4.3.1
Copyright (C) 2008 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
[EMAIL PROTECTED] --list 'libboost-date-time*-dev'
Desired=Unknown/Install/Remove/Purge/Hold
| Status=Not/Inst/Cfg-files/Unpacked/Failed-cfg/Half-inst/trig-aWait/Trig-pend
|/ Err?=(none)/Hold/Reinst-required/X=both-problems (Status,Err: uppercase=bad)
||/ Name Version Description
+++-=-=-==
pn libboost-date-time-de none(no description available)
ii libboost-date-time1.3 1.35.0-5 set of date-time libraries
based on generic programming co
[EMAIL PROTECTED]
signature.asc
Description: Digital signature
Bug#495732: netbase: /etc/service is out of date (IANA port numbers)
Package: netbase Version: 4.32 Severity: normal Please update /etc/services to the latest IANA port numbers (last updated 2008-08-13) http://www.iana.org/assignments/port-numbers And example: entry is missing for gnunet 2086/tcp GNUnet gnunet 2086/udp GNUnet -- System Information: Debian Release: lenny/sid APT prefers testing APT policy: (990, 'testing'), (500, 'unstable') Architecture: amd64 (x86_64) Kernel: Linux 2.6.25-2-amd64 (SMP w/2 CPU cores) Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/bash Versions of packages netbase depends on: ii ifupdown 0.6.8+nmu1 high level tools to configure netw ii lsb-base 3.2-19 Linux Standard Base 3.2 init scrip netbase recommends no packages. netbase suggests no packages. -- no debconf information -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
