Alexander Sack - Debian Bugmail wrote:
Have you verified that this feature works upstream?
No, but it appears so from this:
http://wiki.mozilla.org/Thunderbird:Supported_authentication_methods
http://mailman.mit.edu/pipermail/kerberos/2005-September/008463.html
Marcus
--
To UNSUBSCRIBE,
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Gerrit Pape wrote:
I can document this in the README.Debian, is that fine with you?
Sure, if there is no real solution.
PRNGD does not seem to be packaged for Debian, so that would be a good
first step if someone wants to work on this.
Marcus
I can now confirm that this happens also on a newly booted system, not
only after suspend. (But still only sometimes.)
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
maximilian attems wrote:
what udev version are your running?
I was running udev 0.085-1. Will upgrade and try again.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
maximilian attems wrote:
please upgrade to 0.087-2 in unstable.
Yes, this fixes it.
Marcus
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Actually it seems to me the correct fix is to bump the udev dependency
version.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Michael Biebl wrote:
Maybe the Debian KDE maintainer consider to include this patch from SUSE.
There is no patch in the message you pointed to. Perhaps he means that
it is fixed in current KDE?
Thanks,
Marcus
-BEGIN PGP SIGNATURE-
Version:
reopen 343761
thanks
Nope, the problem is still there.
Marcus
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Tags: patch
Hello,
I needed a newer unionfs so I updated the package. My packages are
available from
http://www.better.se/debian/unionfs/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Eduard Bloch wrote:
Could the program be updated to work with kernel 2.6?
Maybe... where did you get the other patch?
- From the nslu2-linux project. It is now also included in the
OpenEmbedded source tree.
There is a description here:
retitle: swsusp: Cannot find swap device, try swapon -a.
thanks
Well, it seems that rebuilding the initrd with initramfs-tools did help.
At least suspend and resume is working now. (I don't really understand
why the initrd should matter when _suspending_, but that's ok...)
--
To UNSUBSCRIBE,
reopen 352656
thanks
Well, I don't find this information in README.gz.
Sorry, I was using the version from testing. It is now in html/index.html:
This package unifies the control of power managing facilities on your
PC. It supports hardware based on ACPI, APM, IDE-disks and CPU
Michael Biebl wrote:
Did you read the instructions in README.Debian about adding the users,
who shall get access to powersaved, to group plugdev?
Oops, silly me. Will try, and close the bug if it works.
Perhaps the error message could give a hint about this, instead of
claiming that the
Maybe I will add a comment like this to this section. Do you think this
makes it clear enough?
Yes, but please mention that the Debian package already does this so the
user really doesn't need to care.
(Or you could consider just removing the reference to acpid. There is no
problem, so why
maximilian attems wrote:
please use initramfs-tools, it is the default initramfs generator
for 2.6.15.
I can try, but are you sure this is relevant? My problem is not with
resuming, the problem is that it doesn't suspend.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
I'm sorry to say that the bug still appears with 2.0.1-1, despite my
earlier messages. So it will have to be reopened.
This is both with gij and Sun JRE.
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.2 (GNU/Linux)
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
I seem to have mixed up bug numbers here. This bug is indeed gone. I
confused it with #340468.
Very sorry for the inconvenience.
Marcus
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.2 (GNU/Linux)
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
FYI, the exact error message is
The connection to the data source 'test' could not be established.
libhsqldb2: file not found
(when working on a document named test.odb.)
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.2 (GNU/Linux)
Tags: patch
I updated the package for kernel 2.6.15. It is available from
http://www.better.se/debian/kernel-patch-badram/
Marcus
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Tags: patch
I have updated the package for 2.6.14 and 2.6.15 kernels. The updated
package is available at:
http://www.better.se/debian/kernel-patch-exec-shield/
My package no longer depends on kernel-patch-acl, so it need not be
removed from etch. I for one would like to see this package in
Tags: patch
I have updated the package with patches for 2.6.14 and 2.6.15. You can
get my package from
http://www.better.se/debian/kernel-patch-skas/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
This bug appears to have magically disappeared in 2.0.1-1.
Marcus
signature.asc
Description: OpenPGP digital signature
OOo 2.0.1-1 gives the same result but a different NullPointerException:
java.lang.NullPointerException
at
com.sun.star.wizards.ui.TitlesComponent.addTextListener(TitlesComponent.java)
at
com.sun.star.wizards.report.ReportWizard.buildSteps(ReportWizard.java)
at
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Ouch. The bug is back again in 2.0.1-1. This time definitely only when
using gij, not with Sun JRE.
The same workaround still works.
Marcus
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.2 (GNU/Linux)
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
I just tried this with gij instead of Sun JRE. It doesn't give an
exception, instead nothing visible happens at all when I try to start
the report guide.
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.2 (GNU/Linux)
I have updated the package for kernel 2.6.14. Bug me if you want it.
Marcus
signature.asc
Description: OpenPGP digital signature
I have updated the package for kernel 2.6.14. Bug me if you want it.
Marcus
signature.asc
Description: OpenPGP digital signature
Package: lintian
Version: 1.23.14
Severity: normal
I have a package which uses dbconfig-common but has no debconf
templates of its own. Lintian gives two warnings: no-debconf-templates
and missing-debconf-dependency.
Both of these seem to be bogus. My package depends on dbconfig-common
(which in
Package: initscripts
Version: 2.86.ds1-7
Severity: normal
The postinst script fails, probably because I am using file-rc:
---
Setting up initscripts (2.86.ds1-7) ...
/var/lib/dpkg/info/initscripts.postinst: line 62: cd: /etc/rcS.d: No such file
With which JDK? Ah, I see. Sun 1.5...
Right, and the working system had gij or kaffe, I think. But switching
it to Sun 1.5 made the bug appear on that system too. So it seems to be
related to that.
I also have an oobase file which uses an ODBC database (and not HSQLDB).
I can work on that file
Package: openoffice.org-base
Version: 2.0.0-5
Severity: important
Start oobase. In the dialog box that shows up, choose to create a new
database file. Any attempt to manipulate the database, such as
clicking on Tables to create tables gives a pop-up error message:
The connection to the data
Package: openoffice.org-base
Version: 2.0.0-5
Severity: normal
When trying to create a report using the report guide (selecting
Reports and clicking on Use Wizard to Create Report...), instead
of the report guide, a Writer window pops up. It shows a document
named Untitled1 containing no text
Package: vrms
Version: 1.11
Severity: normal
graphviz is listed as non-free by vrms, even though it is now in main.
-- System Information:
Debian Release: testing/unstable
APT prefers testing
APT policy: (990, 'testing'), (500, 'unstable'), (500, 'stable'), (1,
'experimental')
Architecture:
Package: vrms
Version: 1.11
Severity: normal
ncompress is listed as non-free by vrms, even though it is now in main.
-- System Information:
Debian Release: testing/unstable
APT prefers testing
APT policy: (990, 'testing'), (500, 'unstable'), (500, 'stable'), (1,
'experimental')
Package: vrms
Version: 1.11
Severity: normal
zoo is listed as non-free by vrms, even though it is now in main.
-- System Information:
Debian Release: testing/unstable
APT prefers testing
APT policy: (990, 'testing'), (500, 'unstable'), (500, 'stable'), (1,
'experimental')
Architecture: i386
Hmm, it still looks broken to me. In my .llgal/captions I have lines like
DIR: december Open subgallery december
I get the following output:
-
$ llgal -R
Reading the captions file and preparing entries: 100%
Preparing entries: 100%
Found
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Package: llgal
Version: 0.12-1
Severity: normal
llgal generates HTML files with a content-type specifying
charset=ISO-8859-1, but directory names are not encoded correctly. I
have LANG=sv_SE.UTF-8 and have a directory named 'Täby' (that's T, a
with
Well, what's exactly the real problem here ? Does the link point to an
invalid URL instead of your directory ?
Actually the link works, it just looks ugly. Like this:
Täby
If I replace iso-8859-1 by utf-8 in the header then it displays
correctly. I guess it makes sense to use the local
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Tags: patch
The upstream patch for 2.6.14.1 applies to the Debian kernel
(linux-source-2.6.14 2.6.14-4) and seems to work for me (tested on one
system for a few days only).
I made a trivial patch for the Debian package (just added the new patch
from
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Is there something unusual in your configuration?
I don't think so.
I use Heimdal, perhaps that's a difference? And I have the user accounts
in LDAP (but the passwords managed by Kerberos, outside of LDAP).
Permissions on /tmp for ticket caches?
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
I have an Athlon XP 2400+ system. I have tried athcool for a few days,
and have observed no problems so far. The motherboard is Epox EP-8RDA
with nForce2 chipset.
Send me an e-mail if I can help out in any way.
Marcus
-BEGIN PGP SIGNATURE-
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Russ Allbery wrote:
Hm. That looks okay. Could you add debug to the end of the two
pam_krb5.so lines and then send me the resulting log output from syslog
Here it is:
Nov 23 10:06:37 myhost sshd[18820]: (pam_krb5): none:
pam_sm_authenticate:
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Could you send the contents of your /etc/pam.d/common-auth and
/etc/pam.d/common-session files?
/etc/pam.d/common-auth:
-
auth sufficient pam_krb5.so ignore_root
auth requiredpam_unix.so try_first_pass nullok_secure
-
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
libpam-krb5 1.2.0-1 has been uploaded. Let me know if that fixes the
problem for you when you get a chance to try it.
No, I still don't get it does not fix it. I have tried both with and
without privilege separation.
-BEGIN PGP SIGNATURE-
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hello,
I am the maintainer of input-utils. A user reported a problem (bug
#326149) on the powerpc architecture that I need some help
investigating, since I don't have a ppc machine.
Apparently on his system the event devices start at
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Yves-Alexis Perez wrote:
On my box, that's not the case (i'm using udev)
So was he...
crw-rw 1 root root 13, 64 2005-11-17 11:05 /dev/input/event0
crw-rw 1 root root 13, 67 2005-11-17 11:05 /dev/input/event3
There is a gap here.
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Yves-Alexis Perez wrote:
Events are dynamically created/deleted
by udev, and they start at 0. But maybe this is a kernel bug, or udev bug..
No, I think it is perfectly OK. input-utils will have to be fixed to
deal with it.
Thanks for the help!
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hello,
I am the maintainer of input-utils. A user reported a problem on the
powerpc architecture that I need some help solving:
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=326149
Apparently on his system the event devices start at
Daniel Kahn Gillmor wrote:
This looks very similar to bug 296464 to me. the extra prompt at boot
time is likely just a prompt for the crypted swap partition.
My earlier mail to this bug indicates that this is not the so.
Apparently mkinitrd incorrectly determines that the root device needs
Are you using udev?
signature.asc
Description: OpenPGP digital signature
What is your exact kernel version?
Marcus
smime.p7s
Description: S/MIME Cryptographic Signature
Package: wnpp
Severity: wishlist
* Package name: tikiwiki
Version : 1.9.1
Upstream Author : Luis Argerich, Garland Foster, Eduardo Polidor, et. al.
* URL or Web page : http://tikiwiki.org
* License : LGPL
Description : the Tiki content management system
Tiki
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
nexus:~# ls /dev/input/
event1 event2 event4 event6 event8 micemouse1 mouse3
event10 event3 event5 event7 event9 mouse0 mouse2
Do you have any idea why your system does not have /dev/input/event0?
Marcus
-BEGIN PGP
I just compiled libgcrypt with --noexecstack, which adds the correct
markings. This means that users of exec-shield enabled kernels can now
get stack protection for applications linked with this library.
So far it does not appear to have broken anything, and the discussion of
bug #321721
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
It would be useful for exec-shield users to have libcrypto correctly
marked. Otherwise exec-shield is disabled for several important network
servers where it is most useful to have it working, including the
following: sshd, named, slurpd, cyrmaster,
Package: initrd-tools
Version: 0.1.82
Followup-For: Bug #303403
I have reproduced the bug with kernel 2.6.12-1-686 with a RAID
setup. I have also found a workaround.
The system has both root and a swap partition on RAID1. The swap
partition is ecnrypted with a random key, but the root partition
It seems that this _may_ have something to do with the filetest pragma
in Perl:
http://search.cpan.org/dist/perl/lib/filetest.pm
As that page says, the filetest permission operators normally only check
the Unix mode bits, and ignore ACLs, but this can be changed with a pragma.
That
Package: llgal
Version: 0.9.10-1
Severity: normal
The program makes use of the Perl filetest operators (such as -r) to
check if files are readable. These operators by default do not
understand access control lists, and only check the standard mode
bits. Hence llgal can complain that some files
There seems to be a Debian package here:
http://kanotix.com/files/debian/acerhk/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
can you please upgrade and check status on newer image like 2.6.12
The problem disappeared with 2.6.10, and 2.6.11 also works. (I didn't
try 2.6.12 yet.)
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Hi Yann,
did you accidentally close this bug? I don't see what this has to do
with kernel-patch-nfs-ngroups...
Thanks,
Marcus
PS. And by the way, I have BadRAM patches for 2.6.11 and they seem to
work well here.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe.
Package: album
Version: 3.04-2
Severity: normal
It seems that album uses some acl-incomatible method of determining
file access rights.
Album will silently ignore image files which have the access control
list set so that they are readable by the current user, but which
would otherwise not be
Package: module-assistant
Version: 0.8.7
Severity: normal
The package should require libterm-size-perl:
# module-assistant update
Can't locate Term/Size.pm in @INC (@INC contains: /etc/perl
/usr/local/lib/perl/5.8.4 /usr/local/share/perl/5.8.4 /usr/lib/perl5
/usr/share/perl5 /usr/lib/perl/5.8
Package: zaptel-source
Version: 1:1.0.7-3
Severity: serious
Building the module with make-kpkg gives the following error:
$ make-kpkg --append-to-version=-thales --revision=11 --added-modules=zaptel
modules_image
...lots of lines deleted...
make -C /fs/src/kernel/kernel-source-2.6.11
Package: cloop-src
Version: 2.01.5-4
Severity: normal
The module does not understand the --rootcmd option to make-kpkg, so
building the module fails unless fakeroot is used explicitly for the
whole make-kpkg command:
$ make-kpkg
who's the owner for the database? have you tried changing it?
The owner was 'postgres'. I created the database according to the
instructions in README.pgsql.
I have now recreated the database with nagios as owner instead. This
seems to fix the problem!
ps now shows a different command line:
Package: nagios-pgsql
Version: 2:1.3-cvs.20050116-1
Severity: normal
When running nagios, CPU usage for the postgres process is permanently
around 5-7%.
The ps command shows the following:
8131 ?D 3:37 postgres: nagios nagios 127.0.0.1 DELETE
top gives the following:
PID USER
by any chance do you have this problem if you set LANG to C
Yes, the problem is still there.
Looking at the PgSQL logs, I see statements of the following type
repeating continuously:
---
2005-02-20 19:06:05 [32291] LOG: statement: BEGIN TRANSACTION
901 - 968 of 968 matches
Mail list logo
