Package: wnpp
Severity: normal
I intend to orphan the diffstat package.
The package description is:
This program is a simple filter that reads the output of the 'diff' program,
and produces a histogram of the total number of lines that were changed.
It is useful for scanning a patch file to
Package: wnpp
Severity: normal
I intend to orphan the libpam-pwdfile package.
The package description is:
This PAM module lets you use an arbitrarily-named text file similar in
structure to /etc/passwd to authenticate users. Both DES and MD5
hashed passwords are supported.
--
To
If it helps at all, I see the following entry in /var/log/kern.log each
time the problem occurs.
rred[28536]: segfault at bf56cf84 ip b74958af sp bf56cf88 error 6 in
libc-2.10.1.so[b742e000+141000]
I have apt 0.7.24 and libc6 2.10.1-5 installed.
--
To UNSUBSCRIBE, email to
On Sun, Jul 20, 2008 at 11:33:22AM -0400, Joe Nahmias wrote:
It would be nice if diffstat supported compressed diffs natively (-z
option?) instead of having to do `zcat foo.diff | diffstat`.
This is already in place, assuming that the diff filename has the usual
.gz/.bz2 suffix. For example:
My apologies for the lengthy delay... I've been somewhat overwhelmed by
offline issues recently. :(
Unfortunately I've been unable to reproduce your issue, even after
duplicating the pam configuration you provided, so I'm not sure what's
going on at the moment. I'll keep trying, and let you
tags 488248 + patch
quit
The attached patch implements this request via the -q option, and
seems to work reliably here. I'm going to forward the beastie upstream
before applying it to the Debian package, however, since I'd rather not
diverge unnecessarily if Thomas opts to implement it
On Sun, Mar 02, 2008 at 08:23:36AM +0800, Jerome BENOIT wrote:
I have noticed that libpam-pwdfile outputs the following kind of messages in
/var/log/syslog:
Feb 20 01:37:01 rainbow CRON[18708]: PAM unable to resolve symbol:
pam_sm_acct_mgmt
Feb 20 01:37:01 rainbow CRON[18708]: PAM unable
it would be nice if diffstat provided a way to work with gzip'd patches,
as it's pretty common that they come in this flavor (for example, the
diff.gz files from debian source packages). while it's possible to do
zcat foo.gz | diffstat, it'd be a lot nicer if one could do diffstat
- -z
Here's the relevant information from ~boinc/stdoutdae.txt, using the
original ca-bundle.crt with http_debug enabled.
2007-10-09 20:59:04 [World Community Grid] Sending scheduler request:
Requested by user
2007-10-09 20:59:04 [World Community Grid] Reporting 1 tasks
2007-10-09 20:59:04
Package: boinc-client
Version: 5.10.8-2
Severity: normal
With the supplied ~boinc/ca-bundle.crt, which is a symlink to
/etc/ssl/certs/ca-certificates.crt, boinc-client is unable to
communicate with the World Community Grid project (which requires SSL).
The logfile shows the following error
On Thu, 2007-09-13 at 22:49 -0400, Ken Wolf wrote:
I too am experiencing the same problem described in this thread.
Is there a fix or workaround to this issue?
As a workaround, downgrading the following four packages should get
gnucash working again... worked for me, anyway. I believe the
I've sent the attached patch, which implements this request vi the -b
option, upstream for review. If I haven't heard back in the next week
or two, I'll probably go ahead and include it in the next upload.
diff -urN diffstat-1.44.orig/diffstat.1 diffstat-1.44/diffstat.1
---
On Wed, Apr 18, 2007 at 10:20:08AM -0500, Greg Norris wrote:
You probably just need to double up the quotes the name, so that the
remote shell sees them. Something like:
scp 'leghorn:/mnt/rdbackup/home/sachin/Daily report April -07.ods' .
Actually, that should've been:
scp leghorn
Package: samba-common
Version: 3.0.24-3
Severity: minor
I recently noticed the following error message, after rebooting for a
kernel update.
invoke-rc.d: unknown initscript, /etc/init.d/samba not found.
It turned out to be caused by /etc/dhcp3/dhclient-enter-hooks.d/samba,
since I don't have
Package: fuzzyocr3
Version: 3.5.1-2
Severity: normal
I've got fuzzyocr3 installed with ocrad, rather than gocr, and get the
following warning when syncing the bayes journal.
$ sa-learn --sync
[19425] warn: FuzzyOcr: Cannot find executable for gocr
bayes: synced databases from journal in
tags 379736 patch
quit
I also haven't been able to locate any version of netcat which supports
this feature. The attached patch therefore updates the manpage with an
equivalent example using socat, which is definitely available.
diff -ur openssh-4.3p2.orig/ssh_config.0
On Fri, Nov 17, 2006 at 03:22:59AM +0100, Lorenzo Martignoni wrote:
Why do you say that? The init script in the package does reference
/etc/default/shorewall.
You're correct, of course... color me embarrassed. Not sure how I
managed to pull off that particular disconnect. :(
signature.asc
At the very least, README.Debian should be updated to reflect this
change in behaviour. Currently the file says:
1. AUTOMATIC STARTUP
In order to avoid the startup of the firewall on an unconfigured
machine, automatic startup, on boot, is disabled by default.
Could you post a complete copy of /etc/hosts.deny as an attachment? I'm
thinking that perhaps it contains a non-displayable character which is
confusing libwrap... a long shot, admittedly, but I've seen this sort of
thing in the past.
Also, what platform is the affected server?
I just checked this using version 1:4.3p2-5.1 from sid, and it appears
to be working as expected.
[EMAIL PROTECTED] tail -2 /etc/hosts.deny
ALL EXCEPT sshd: PARANOID
sshd: 127.0.0.1
[EMAIL PROTECTED] ssh localhost
ssh_exchange_identification: Connection closed by remote host
On Fri, Oct 27, 2006 at 09:19:14AM +1000, Alexander Samad wrote:
it should be looking at TMP or TMPDIR
[EMAIL PROTECTED] dir `which ssh-agent`
-rwxr-sr-x 1 root ssh 65836 Oct 24 16:06 /usr/bin/ssh-agent
Since ssh-agent has the setgid bit set, it won't even see the TMPDIR
variable... this
On Thu, Sep 14, 2006 at 01:40:25AM +0200, Helge Hafting wrote:
Unfortunately, ssh always check the hostkey against the IP
address only, and so it thinks there is a man-in-the-middle
attack when I try the second pc instead of the first.
Have you tried setting the HostKeyAlias option in
Package: console-tools
Version: 1:0.2.3dbs-65
Severity: normal
Tags: patch
The console-screen.sh scripts sets the NUM_CONSOLES variable only if
setting a global default font, but it's also referenced if setting the
keyboard LEDS (which is run independently). The patch below simply
relocates the
Package: xfce4-utils
Version: 4.2.3-1
Severity: wishlist
Tags: patch
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Please consider adding support for user-defined environment settings by
xinitrc, similar to how kde processes the $HOME/.kde/env directory.
This is particularly useful when
Package: login
Version: 1:4.0.15-2
Severity: wishlist
Version 2.3.3-8 of amavisd-new fixes the syntax issue with su. Please
update the Conflicts entry to allow it's installation.
Thanx!
-- System Information:
Debian Release: testing/unstable
APT prefers unstable
APT policy: (500,
The easiest way to do this, if I understand you correctly, is to set
your forced command to a script which interrogates the
SSH_ORIGINAL_COMMAND variable. Assuming that the requested operation is
legitimate, it can then exec whatever process is appropriate.
signature.asc
Description: Digital
Unfortunately, I seem to be unable to reproduce your issue... even when
using the exact configuration you provided. Any chance you could strace
the ssh daemon, and capture a pair of connections which exhibit the
problem?
$ strace -fF -o sshd.strace /usr/sbin/sshd -D
signature.asc
This seems to be working as expected here (same version of ssh on a
fully up-to-date sarge box), regardless of the X11UseLocalhost setting.
If you'd be so kind as to post a copy of /etc/ssh/sshd_config, I'll
compare it against mine and try to determine if there are any relevant
differences.
On Thu, Jan 26, 2006 at 02:56:24PM -0600, Marcos Pinto wrote:
hosts.allow has ALL: *
I want everyone to have access except for those in hosts.deny. Is
that not the proper way?
The settings in hosts.allow have precedence over hosts.deny, so this
entry is overriding your denial rules. Since
On Thu, Jan 26, 2006 at 04:44:19PM -0600, Greg Norris wrote:
This is documented in the hosts_access manpage.
More specifically, that should be hosts_access(5)... I hadn't noticed
that there are multiple manpages.
signature.asc
Description: Digital signature
I have a sarge box (fully updated as of 2 days ago) running the same
version of ssh, and it honours the /etc/hosts.deny settings. Do you
have any entries in /etc/hosts.allow?
signature.asc
Description: Digital signature
tags 234627 patch
stop
Here's a patch which implements the requested change. Please note that
I defaulted to the RSA public key (id_rsa.pub) rather than DSA, as it
appears to be the preferred format now that the relevant patent has
expired. Should you prefer to go with DSA instead, only
Package: installwatch
Version: 0.6.3-2
Severity: normal
I haven't touched installwatch in quite some time, and no longer have
any real interest in the beastie, so I've decided to orphan this package.
At one point Matt Hope had expressed interest in taking it over (more
recent versions are bundled
Package: kernel-package
Version: 10.026
Severity: normal
The --revision argument seems to be ignored by make-kpkg.
$ fakeroot -- make-kpkg --revision=9:sasami.1 binary
[ SNIPPED ]
$ cd ..
$ ls -1 *.deb
linux-doc-2.6.15_yggdrasil.1_all.deb
Package: firefox
Version: 1.5.dfsg-2
Severity: normal
Tags: sid fixed-upstream
The first time save file as is selected, the context menu is
immediately redisplayed in always-on-top mode (often overlaying the file
picker). It also causes all tabs (and windows, IIRC) to redraw, shifting
position
Package: gdm
Version: 2.6.0.8-1
Severity: normal
When I configure gdm to use the pam_tmpdir.so module via /etc/pam.d/gdm,
it appears to hang just after password entry. The password-entry field
is grayed out and the screen stops responding to input, but does not
appear to start any of the
On Wed, Aug 31, 2005 at 09:47:26AM +0200, Florian Weimer wrote:
Isn't the default in the binary itself still no, but the default
Debian configuration file includes a yes?
Yeah, I think you're correct. In that case, maybe a mention in
README.Debian or NEWS.Debian would be more appropriate.
On Wed, Aug 24, 2005 at 07:39:00PM +0200, Carlos González wrote:
Package: openssh-client
Version: 4.1p1-6
The saved keys of this ssh version in the known_hosts don't have host
name and ip. For each key there is 2 lines in the file like this:
[snipped for brevity]
This is a documented
tags 324891 patch
thanks
On Wed, Aug 24, 2005 at 06:36:51PM -0500, Greg Norris wrote:
It looks like the manpage needs to be updated, however, as it
incorrectly states that the parameter defaults to no (which is the
upstream default).
I've attached a trivial patch to correct the ssh_config
On Mon, Aug 15, 2005 at 08:58:25PM -0500, Greg Norris wrote:
Thomas Dickey has just informed me that version 1.40 has been released.
If all goes well, I'll upload the packaged version sometime this
weekend.
Apparently there will be a 1.41 bugfix release in the near future, so
I'm holding off
On Sun, Aug 14, 2005 at 11:49:04AM +, W. Borgert wrote:
The package description says:
This program is a simple that reads the output of the 'diff' program,
and produces a histogram of the total number of lines that were changed.
It is useful for scanning a patch file to see which files
On Wed, Apr 06, 2005 at 12:37:50PM -0700, Tyler MacDonald wrote:
I was just trying to copy a folder over with a colon in it's name. Of
course, scp thought that the colon was a hostname separator and responded
with ssh: The: Name or service not known
I ran into the same issue recently. You can
On Wed, Apr 06, 2005 at 12:42:02PM -0700, Tyler MacDonald wrote:
[i just accidentally submitted this with a bad email addy as an addition to
wontfix bug 198275; resubmitting since its actually a different bug...]
My reply went to #198275 as well... resubmitting here for completeness.
I was
On Wed, Mar 09, 2005 at 05:08:44PM -0600, Micah Anderson wrote:
Now that I look some more, I wonder... are you changing the window
size when you are attaching? If so, this is a problem with the
underlying curses library or how its being used, as this message comes
from a curses call failing.
On Sun, Mar 06, 2005 at 07:57:06PM +0100, Julien BLACHE wrote:
The bug is fixed in the newer upstream version anyway, so it isn't
really important. If needed I can try to recover the diff from the svn
repo.
Ah, I didn't realize that Thomas had made a new release (a couple of
them, actually)...
On Mon, Feb 14, 2005 at 11:31:55PM +0100, Tomas Pospisek wrote:
Please package in order to be able to package checkinstall. The latest
installwatch is included in the checkinstall sources [1].
Last I heard, Matt Hope (the checkinstall maintainer) was planning to
take over the installwatch
On Fri, Feb 04, 2005 at 04:09:39PM -0600, Micah Anderson wrote:
Can you try the newer version of bittornado to see if this problem
still exists?
Yes, it's still present in 0.3.10-1. The only visible difference from
the prior version seems to be the line number listed for
`SocketHandler.py'.
On Mon, Jan 24, 2005 at 01:46:30AM +0100, Santiago Vila wrote:
The recommended practice has always been not to submit bug reports for
things that would result in a lot of bugs being filed. That is massive
bug filing, so it should be discussed first.
Just to clarify, I never intended (and still
On Sun, Jan 23, 2005 at 10:29:12PM +0100, Jeroen van Wolffelaar wrote:
Greg: Ease of adding, and potentional negative benefits would be very
nice to have, and if it's going to be in policy, for lintian a way to
check for it.
Purpose:
PT_GNU_STACK is used to mark binaries which require an
On Sat, Jan 22, 2005 at 02:51:23PM +0100, Santiago Vila wrote:
I closed this bug because there must be literally hundreds of packages
like this and I consider premature to submit bugs for all of them.
I disagree on this point (being premature, not your estimate of the
affected packages), but I
Package: diff
Version: 2.8.1-9
Severity: minor
The binaries appear to have been built without the PT_GNU_STACK header,
which makes the 2.6.10 kernel enable read-implies-exec behaviour. This
in turn causes problems under SELinux, because executable stacks are not
typically allowed for legacy
51 matches
Mail list logo
