On Thu, Feb 26, 2009 at 1:58 AM, Milan Zamazal wrote:
> Did you start with newly created *.css files? crm114 has introduced a
> change that makes the old *.css completely incompatible on 64 bit
> systems.
>
> If your problem persists even with newly created *.css files, would you
> send me your c
Package: crm114
Version: 20080330-2
Severity: important
Since upgrading from etch to lenny, crm114 has failed to categorize e-mail
at all. It always returns a spam probability score of 0.0 and a status
of UNSURE. This result is the same no matter what training I engage in.
It even reports 0.0 UN
Further inspection reveals that a fresh etch install works, as well.
I've tracked the problem to the fact that my system is missing
/usr/lib/python2.4/site-packages/python-support.pth which is reported
as owned by python-support. Reinstalling python-support fixes this
problem. (That said, I would
On Wed, Apr 2, 2008 at 10:43 AM, Vincent Danjean <[EMAIL PROTECTED]> wrote:
> severity 473995 normal
> tag 473995 +moreinfo
> thanks
>
> Ari Johnson wrote:
> > With a fresh install of tailor and as far as I can tell also a fresh
> > install of python (I
Package: tailor
Version: 0.9.26-2
Severity: grave
Justification: renders package unusable
With a fresh install of tailor and as far as I can tell also a fresh
install of python (I do not use python, so none of its configuration
should have been changed other than by APT), running tailor gets this
Package: php
Version: php5
Severity: normal
strip_tags() produces incorrect output for certain input that contains
an apostrophe. The basic bug is that, if strip_tags() encounters an
apostrophe anywhere other than the first character of its input string,
then all input from the '<' that opens the
Package: flexbackup
Version: 1.2.1-5
Severity: wishlist
Tags: patch
I would like to be able to exclude certain directories from flexbackup's
backup process, without having to edit the configuration file. In
particular, I want my users to be able to specify directories that
should not be backed up
tags 412853 + security
I am requesting that this be tagged as a security issue because:
(1) I have submitted a patch and the maintainer has not responded; it
is an easy fix
(2) Because the package strictly reports false positives for its
md5sum checks of suid and sgid files, and because its purpo
m
produces the same md5 checksums as the 'md5sum' program does. A patch
is attached to this message.
I hope you find my patch helpful and timely for inclusion in etch. Thanks!
Ari Johnson
sxid-64bit-clean.patch
Description: Binary data
Note that this bug is related to #204840. In that bug, the command
was 'split -50' and was changed by the fix to 'split -l 50', which is
correct. Somewhere in between (I don't know when, exactly), this line
was changed to 'split -l -50', which doesn't seem to be valid for
*any* version of split.
Package: php5-gd
Version: 5.2.0-8
Severity: wishlist
The php5-gd package has the following dependency tree:
- php5-gd
+- libx11-6
+- libxpm4
| +- libx11-6
+- libgd2-xpm
+- libx11-6
+- libxpm4
As a result, the php5-gd package requires substantial parts of X11 to be
installed. I
Package: dist
Version: 3.70-31
Severity: important
The file /usr/share/dist/U/Chk_MANI.U on line 42 causes the generated
Configure script to include a call to "split -l -50", which causes
split(1) from coreutils 5.97-5.3 to complain as follows:
split: -50: invalid number of lines
Try `split --help
On 2/28/07, Ben Collins <[EMAIL PROTECTED]> wrote:
On Wed, 2007-02-28 at 11:09 -0500, Ari Johnson wrote:
> On 2/28/07, Ben Collins <[EMAIL PROTECTED]> wrote:
> > On Wed, 2007-02-28 at 09:40 -0500, Ari Johnson wrote:
> > > Package: sxid
> > > Version: 4.0.5
On 2/28/07, Ben Collins <[EMAIL PROTECTED]> wrote:
On Wed, 2007-02-28 at 09:40 -0500, Ari Johnson wrote:
> Package: sxid
> Version: 4.0.5
> Severity: normal
>
> The sxid program e-mails a list of files every day that it claims have
> changed md5sums. The list appears
Package: sxid
Version: 4.0.5
Severity: normal
The sxid program e-mails a list of files every day that it claims have
changed md5sums. The list appears to include every suid or sgid file on
my system, suggesting that the md5sum comparison performed by sxid
simply does not work. I chose a sample f
15 matches
Mail list logo
