Hello,
On Thu, Sep 05, 2013 at 04:38:46PM -0700, Vagrant
Cascadian wrote:
When called via ssh directly, it echoes the passphrase:
vagrant@local:~$ ssh server passwd
(current) UNIX password: dlkgfjsdgfkjsd
ssh user@server command doesn't allocate
pseudo terminal, therefore tcgetattr() +
Hello,
On Fri, Mar 29, 2013 at 11:54:21PM +0100, Wolfgang
Zarre wrote:
I think that in any case the right solution is
in just flushing the input queue before
returning to the caller which would not just
protect in case of hijacking but also of buggy
applications.
I've successfuly tested an
Hello,
On Wed, Mar 06, 2013 at 11:23:01AM -0600, Serge
Hallyn wrote:
As mdeslaur has pointed out in irc, one solution
would be to have interactive su use a new pty
for the session. Not trivial,
and not very portable too. The best pty handling
is done by Expect (Tcl/Expect) IMHO, but its code
Hello,
On Fri, Feb 08, 2013 at 07:12:26AM +0100,
Christian PERRIER wrote:
Quoting Moo Lambda (lambda...@gmail.com):
Now su - does not work to open x
applications on wheezy. Should I switch to
fedora?
IIRC in RH/Fedora/CentOS/SLX this is handled
by smth like pam-xauth module, which is
Package: gs
Version: 8.62.dfsg.1-3.2lenny5
Severity: normal
Hello,
gs ignores -dDuplex option when using
pswrite/psmono/pdfwrite output drivers, i.e. it
doesn't insert /Duplex true pagedevice entry:
xrgtn@ux230p:/tmp$ cat hello.ps world.ps
%!PS
/Courier 30 selectfont
100 400 moveto
Package: iceweasel
Version: 3.5.16-14
Severity: minor
Hello,
While adding security exception for internal
corporate site I noticed that Certificate Viewer
says *Could not verify this certificate for
unknown reasons* while the reason is known just
fine: Certificate is not trusted, because it
Hello,
After I installed
debian-live-6.0.4-i386-rescue.iso on a notebook,
the first reboot failed with cannot find root
device message.
I checked initrd.img-2.6.32-5-486 and
initrd.img-2.6.32-5-686 and found that
/conf/conf.d/cryptroot was missing.
Also, /etc/crypptab on root filesystem is
On Mon, Jul 09, 2012 at 01:58:46PM +0300,
Alexander Gattin wrote:
I used rescue mode of the CD, generated proper
entry for sda5_crypt in /etc/crypptab, and ran
dpkg-reconfigure cryptsetup to regenerate both
initrd images
only first (686) one was regenerated, so I ran
update-initramfs -u -k
severity 664463 grave
thanks
Raising severity to grave based on bug 554506
precedent (leaves system unbootable after
instellation)
-- xrgtn
signature.asc
Description: Digital signature
The GNU/Linux random(3) manpage says explicitly:
The random() function returns a value between
0 and RAND_MAX
The same is said in GNU libc docs (info libc BSD
Random):
This section describes a set of random number
generation functions that are derived from BSD.
There is no advantage to using
Hello,
On Wed, Jun 13, 2012 at 01:04:51AM +0200, Nicolas
François wrote:
On Wed, Jun 13, 2012 at 12:15:53AM +0400,
pashev.i...@gmail.com wrote:
But random() returns numbers in the range from
0 to 2^31-1 [2].
So, random()/RAND_MAX could result in a value 1.
Theoretically agreed.
Christian, hello!
29.04.2012, 10:16, Christian PERRIER bubu...@debian.org:
Quoting Jakub Wilk (jw...@debian.org):
The attached patch fixes the bug. However, most likely it also
reopens #605329.
Maybe I'll be able to make a fix that closes the 659957
and doesn't reopen 605329, but I'll need
I've compared Debian's lpr/lpd versions vs.
NetBSD's (5.1):
xrgtn@ux380n:~/deb/lpr-2008.05.17$ ls -l lp[dr]/lp[dr].c
-rw-r--r-- 1 xrgtn xrgtn 21694 May 18 2008 lpd/lpd.c
-rw-r--r-- 1 xrgtn xrgtn 16709 May 18 2008 lpr/lpr.c
xrgtn@ux380n:~/deb/lpr-2008.05.17$ grep '\.c,v ' lp[dr]/lp[dr].c
severity wishlist
thanks
It looks like newer version of BSD lpr should
correctly handle :if: filter for remote machines
too, but the tldp's text is not specific enough
(http://tldp.org/HOWTO/Printing-HOWTO/setup.html)
I've installed NetBSD into kvm to play with its
lpd and it _does_ have :if:
Package: lpr
Version: 1:2008.05.17
Severity: normal
Hello,
lpd ignores Input Filter (:if=...:) printcap
setting for Remote Machine printers (:rm=...:).
For local and :lp=port@host: printers the :if:
works OK.
Thus, if remote machine can't handle some file
types it's hard to work around.
Hello,
On Mon, Jan 23, 2012 at 09:06:38PM +0200, Török
Edwin wrote:
PIE refers to -fPIE from GCC of course.
First, let me thank you for your report, Edwin.
Using that flag doesn't completely prevent the
exploit though.
How unfortunate,
Here is a good summary and discussions:
Hello,
On Mon, Jan 23, 2012 at 09:06:38PM +0200, Török
Edwin wrote:
PIE refers to -fPIE from GCC of course.
Using that flag doesn't completely prevent the exploit though.
...
Apparently packages should adopt hardening flags for wheezy:
http://wiki.debian.org/Hardening#State_of_implementation:
Hello,
On Mon, Jan 09, 2012 at 10:09:44PM +0100, Nicolas
François wrote:
On Mon, Jan 09, 2012 at 09:22:59AM +0100,
harald.dun...@aixigo.de wrote:
I had expected passwd is based on pam, isn't it?
OK. Right, if supported by PAM, then passwd
should work.
IIRC passwd's operation for NIS auth
Hello,
On Sun, Jan 08, 2012 at 06:57:27PM +0200, Shalom
Ben-Zvi wrote:
then tried to add myself to sudo group:
~# usermod -G sudo shalom
In Debian you should use adduser shalom sudo
instead of usermod,
since then I was periodically logged out of the
gnome session without notice.
I never
Hello,
I have the same problem with xvidtune, on
Debian/stable system with intel video adapter
(it's handled by i915.ko):
00:02.0 VGA compatible controller: Intel Corporation Mobile 945GM/GMS,
943/940GML Express Integrated Graphics Controller (rev 03)
00:02.1 Display controller: Intel
Package: x11-xserver-utils
Version: 7.5+3
Severity: normal
Hello,
I use UMPC with 4.5 1024x600 screen and need to
magnify part of screen in some applications while
being able to drag magnified zone across all the
desktop/desktops. I tried to use xrandr for that
and it's not fit for the job. For
Package: debfoster
Version: 2.7-1.1
Severity: normal
Hello,
debfoster -o UseRecommends=no xxx
currently behaves the same as
debfoster xxx
It's true for -o UseSuggests=yes too.
Uncommenting and setting these options in
/etc/debfoster.conf has no effect either.
I noticed this nehaviour when
Package: ppp
Version: 2.4.5-4
Severity: important
Hello,
Today I've upgraded ppp from /sarge to /stable and
updetach option started causing pppd to hang up
immediately upon establishing connection:
...
May 16 16:26:53 xrgtn-q40 pppd[30472]: rcvd [IPCP ConfReq id=0x2 addr
192.168.100.101]
Package: linux-2.6
Version: 2.6.32-30
Severity: normal
Hello,
I was experimenting with DoC 2000 chip
(MD2200-D48) on Netier NetXpress XL1000
thin client PC:
Jan 2 04:47:32 xrgtn-cf8g kernel: [ 2251.686031] DiskOnChip found at 0xd4000
Jan 2 04:47:32 xrgtn-cf8g kernel: [ 2251.686371]
Hello,
had the same situation several days ago, libc
became corrupt (no process could be started,
including /sbin/init).
I had no busybox-static on this system, only
sash, but it wasn't very helpful -- no
tar/gzip/lvm2 tools available from sash on this
broken system anyway.
I booted off a USB
On Wed, Jun 23, 2010 at 10:18:28PM +0300, Alexander Gattin wrote:
had the same situation several days ago, libc
became corrupt (no process could be started,
including /sbin/init).
here's excerpt from dpkg.log:
2010-06-20 02:14:41 upgrade libc6 2.10.2-9 2.11.1-3
2010-06-20 02:14:41 status half
Hello,
On Tue, Jun 22, 2010 at 10:15:59PM +0200, Thibault Manlay wrote:
On 21/06/10 01:01, Aurelien Jarno wrote:
If you are able again to produce an image with libc6 crashing (I mean
already crashing, not that will crash after an upgrade), that's probably
something we can already study.
Hello,
On Tue, Mar 17, 2009 at 06:50:46AM +0100, Christian Perrier wrote:
Quoting Alexander Gattin (xr...@yandex.ru):
P.S. Christian, what's your opinion on the issue?
I should have one? :-)
It would be nice if you do :). Maybe you faced
similar discussions in the past?
I generally trust
Hello again,
On Sun, Mar 08, 2009 at 03:19:09PM +0100, Pierre Habouzit wrote:
you absolutely don't get it. I give the two most used
configurations
(i.e. pdnsd-recurse.conf and pdnsd-resolvconf.conf)
for free, namely:
* local cache, slave to your ISP or dhcp servers, through the
Hello,
On Sat, Mar 07, 2009 at 10:31:13AM +0100, Pierre Habouzit wrote:
On Fri, Mar 06, 2009 at 09:21:15AM +, Alexander Gattin wrote:
you have totally screwed up the Policy's
definitions and intentions: regardless of
whether it's a conffile or Config File, local
changes must
I do not understand why did you introduce the
AUTO_MODE at all? Probably, you wanted to help
users configuring pdnsd by making it
accomplishable through dpkg-reconfigure pdnsd?
If so, then amount of behavoiur which is exposed
to dpkg-reconfigure pdnsd is obviously
insufficient (server_ip?
On Sun, Mar 08, 2009 at 03:19:09PM +0100, Pierre Habouzit wrote:
Oh boy, you absolutely don't get it.
of course I do,
I give the two most used configurations for free
yes, and while it's perfectly OK regarding your
intentions, the manner in which it's implemented
isn't.
If you're not in
On Sun, Mar 08, 2009 at 11:03:17PM +0100, Pierre Habouzit wrote:
It's not, many packages do the same,
?
In the next upload I'll make it very clear in
the /usr files that those are not meant to be
edited by the user, and will be overwritten by
an upgrade
a large billboard that says in
Hello, Pierre,
you have totally screwed up the Policy's definitions and intentions:
regardless of whether it's a conffile or Config File, local changes
must be preserved during a package upgrade.
--
WBR,
xrgtn
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a
Hello,
On Fri, Nov 28, 2008 at 12:13:19PM +0100, Nicolas François wrote:
There is one feature in login, to change the terminal in all-caps mode
when you enter your login in capital letters only.
IIRS there's also a feature in getty to switch
terminal to all-caps mode when user enters
non-ascii
Hi,
On Wed, Jan 30, 2008 at 05:47:07PM +0100, Folkert van Heusden wrote:
What about parallel logins?
Number of parallel logins is limited to:
1. number of nproc/nofile for root user
2. max number of PIDs on a system
3. speed of CPU context switches
4. in case of login-KILL attempts,
the
Hi,
On Fri, Sep 07, 2007 at 12:06:06AM +0200, Raúl Sánchez Siles wrote:
Could you please try upgrading to latest bochs sid packages just to test that
this is not a duplicate of #417416
(http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=417416)
I checked with latest bochsbios and this
really
Package: qemu
Version: 0.9.0-2
Severity: grave
Hi,
after upgrading qemu and libc:
1.
the WinXP image stoped booting. I can see only
BIOS logo during boot, then qemu hangs showing
no signs of boot progress. (this behaviour
regardless of -no-kqemu rmmod kqemu).
When downgraded to 0.8.2-4, the
,
So maybe we should not include it in /usr/bin?
But we could still bundle it in /usr/share/doc/ncc
Actually I thought just that, to ship it
under /usr/share/doc/.
--
With best regards,
Alexander Gattin
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble
Package: cvs
Version: 1:1.12.13-8
Severity: normal
I've hit this problem on production CVS server,
when tagging (cvs tag XXX a b c/ d/) a part of
distribution of some proprietary s/w for release.
One directory was not tagged and hence didn't
get into the export (cvs export -r XXX some_sw).
I've
Please, see also the similar bug (#15050)
reported against upstream:
http://savannah.nongnu.org/bugs/index.php?15050
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Hi,
On Wed, Mar 21, 2007 at 09:46:02PM +0100, Christoph Berg wrote:
tags 362239 + moreinfo
thanks
Re: Alexander Gattin 2006-04-12 [EMAIL PROTECTED]
I keep sources in CVS and build from working directory.
That way there's a CVS/ directory in every subdirectory
of mutt's source
Hello,
I disagree on 2 points:
1. The bug should be assigned to libc6 instead of libc6-i686,
because it's equally applicable to all architectures,
including i386 of course.
2. The severity isn't minor in any way, because:
* the bug potentially affects a lot of packages that
use
Package: ncc
Version: 2.6-4
Severity: normal
Hi,
I've found that the package source has useful
script gengraph.py for converting ncc output
files to .dot format.
And while source has it and /usr/share/doc/ncc/NCC.gz
mentions it, binary package doesn't include
neither gengraph.py nor its
Package: devscripts
Version: 2.9.27
Severity: normal
Hello,
Looks like pattern parameter expansions:
${parameter#word}, ${parameter##word},
${parameter%word}, ${parameter%%word},
${parameter/pattern/string} and ${parameter//pattern/string}),
parameter length expansion ${#parameter},
error
Hello,
On Tue, Jan 30, 2007 at 10:07:50AM +, martin f krafft wrote:
also sprach Alexander Gattin [EMAIL PROTECTED] [2007.01.30.0930 +]:
${parameter#word}, ${parameter##word},
${parameter%word}, ${parameter%%word},
these are not bashisms.
I assume you know POSIX better than me
I have added support for new utag parameter,
and enforced propagation of tag parameter down
to cvs-buildpackage. The patch is attached.
Now, my .cvs-autoreleasedeb/conf looks like this:
?xml version=1.0 encoding=ISO-8859-1?
!DOCTYPE cvs-autoreleasedeb [
!ELEMENT sources (server)+
!ELEMENT server
Package: cvs-autoreleasedeb
Version: 0.12-1
Severity: normal
Tags: patch
Hello,
I've found 2 problems with handling of CVS tags by
cvs-autoreleasedeb. One is related to validating tag
property in .cvs-autoreleasedeb/conf, it's quite
obvious and the patch for fixing it is attached.
Another
On Sun, Jul 02, 2006 at 08:38:31AM +0200, Christian Perrier wrote:
Tomasz, others, do you think that it would be possible to use PAM to
expire passwords ?
Impossible AFAIK.
--
WBR,
xrgtn
Package: aide
Version: 0.11a-4
Severity: important
Looks like upgrade from 0.11a-3 to 0.11a-4 broke my
aide:
cherokee:~# aide --check
Couldn't open file /tmp/empty/aide.db for reading
cherokee:~# echo $?
18
// ENXIO
ATM I was unable to find older .deb on snapshot.debian.net,
so I'll rebuild -3
Hi!
On Thu, Jun 08, 2006 at 04:58:12PM +0200, Michael Banck wrote:
--- lib/commonio.c.orig 2006-06-08 15:12:21.0 +0200
+++ lib/commonio.c2006-06-08 15:39:49.0 +0200
@@ -47,13 +47,22 @@
int lrename (const char *old, const char *new)
{
+#ifdef PATH_MAX
Hi!
On Mon, May 29, 2006 at 05:33:19PM +0200, Christian Perrier wrote:
since the LC_* settings were moved from /etc/environment to
/etc/default/locale, they are missing in 'su -' sessions.
And from `su` sessions too, BTW.
Perhaps it should be configured similar to /etc/pam.d/login?
That
Package: setcd
Version: 1.5-4
Severity: wishlist
Tags: patch
Hi!
setcd -x as well as hdparm -X wasn't able to slow down
my DVD+RW drive during DVD playback (and it was
rocketing as a Ferrary car (thanks G-d I had earphones)).
While investigating the issue I've found (thank to
Goga777/Sky777):
Hi!
On Wed, May 10, 2006 at 10:13:40AM +0200, Christian Perrier wrote:
Could this be investigated again?
IMHO the bug is related to utmp corruption.
Right now I'm not quite ready to debug the issue deeper
as I don't know how exactly is the utmp is used in
Unices.
I'll try to come up with some
Hi!
On Wed, Apr 26, 2006 at 08:36:51PM +0200, Nicolas François wrote:
On Wed, Apr 26, 2006 at 04:10:36PM +0100, Reuben Thomas wrote:
On Debian, the constraints on the username are lowered:
Usernames must neither start by a dash ('-') nor contain a
colon (':') or an end
Package: mutt
Version: 1.5.11+cvs20060403-2
Severity: normal
Tags: patch
I keep sources in CVS and build from working directory.
That way there's a CVS/ directory in every subdirectory
of mutt's source.
debian/rules does not expect this to happen, and has
some places where /* is shell expanded
Package: mutt
Version: 1.5.11+cvs20060403-1
Severity: normal
Tags: patch
Hi!
When using NTLM(SASL) for authenticating against M$
Exchange Server (its IMAP), mutt fails to auth (SASL
authentication failed.) and then hangs when closing
connection to the server.
Debuglog says:
imap_auth_sasl:
Hi!
On Mon, Apr 03, 2006 at 01:42:09PM +0100, Greg Matthews wrote:
I have no simple way of testing this as I have no host with this
version.
Yeah, I understand you very well indeed. :)
E.g. I have a problem with apcupsd-cgi on one
Debian/testing host while everything is OK on 2 Sarge
hosts.
Hi!
On Tue, Mar 28, 2006 at 06:55:46PM +0300, Alexander Gattin wrote:
never had too much problem setting up either start_TLS or ldaps security
altho I've always used RSA I think.
I've got answer in openldap-software maillist, and
impressively quickly, what a miracle! %)
On Wed, Mar 29, 2006
Hi!
On Tue, Mar 28, 2006 at 03:20:31PM +0100, Greg Matthews wrote:
On Mon, 2006-03-27 at 23:30 +0300, Alexander Gattin wrote:
(I used slapd -d 65535, s_client's debug, tcpdump,
then ssldump...).
I forgot to mention strace and RTFS, of course. :/
never had too much problem setting up
Hi!
On Mon, Mar 06, 2006 at 09:55:23AM +, Greg Matthews wrote:
yes, you can have a number of different CA certs depending on what you
are connecting to. Dropping them into a directory means the ldap tools
will be able to use them (after the symbolic links have been set up).
Today I have
Package: unionfs-source
Version: 1.1.3-1
Severity: important
Hi!
I can't build newest unionfs-modules against 2.4.32
Previous version of unionfs was OK. Last modules that
I've successfully built, are:
unionfs-modules-2.4.32-grsec_1.0.14-1+6_i386.deb
Usually I build from script (which builds
Hi!
On Thu, Mar 02, 2006 at 09:56:03AM +, Greg Matthews wrote:
sorry for the long silence,
It's me who had to be sorry, actually, as I didn't have
enough time to work on the bug.
just trying to reproduce this bug and the symptoms seem to have changed,
I dont get a segfault but su is
Hi!
On Sat, Mar 04, 2006 at 07:08:20PM +0100, Nicolas François wrote:
Anyway, I will add the x, because it fails with ash otherwise.
Yes, I had big concerns about not having x in front
of expanded password. However, because I didn't succeed
trying to exploit this using bash's test or GNU test
On Sat, Mar 04, 2006 at 07:08:20PM +0100, Nicolas François wrote:
Your patch also makes root_password to return 0 when the /etc/passwd
passwd is set to ! (and it does not check if the /etc/shadow passwd is set
to !). What is it used for?
Please ignore my previous comments WRT this issue. Now
I
Hi!
On Mon, Feb 27, 2006 at 07:20:45AM +0100, Christian Perrier wrote:
P.S. The requred infrastructure will be ready soon.
And now, one month later?
And now, *two* months later? :-)
Oh, yeah, now it's 2 months closer to completion ;)
Actually, a lot of different and I'd say boring job
Hi!
On Wed, Feb 22, 2006 at 10:08:46AM +0100, Samuele Giovanni Tonon wrote:
Alexander Gattin wrote:
Today I found that apcupsd-cgi reports status of
perfectly working hosts as ONLINE NO BATTERY.
Downgrading to stable version fixes the problem.
pls can u try to upgrade all of three
Package: apcupsd-cgi
Version: 3.12.1-2
Severity: normal
Hi!
Today I found that apcupsd-cgi reports status of
perfectly working hosts as ONLINE NO BATTERY.
Downgrading to stable version fixes the problem.
Here is my /etc/apcupsd/hosts.conf:
MONITOR vespasian vespasian (chassis #01)
MONITOR
Hi!
On Thu, Jan 26, 2006 at 03:53:35PM -0500, Jeffrey Sheinberg wrote:
This worked fine under sarge, but now no longer works under etch.
Just a first thought:
If you `chmod u+s /bin/login` (return it to suid root),
does it start working again?
--
WBR,
xrgtn
--
To UNSUBSCRIBE, email to
, Inc.
.\ August 2004: Updated by Thomas Hood [EMAIL PROTECTED]
+.\ January 2006: Updated by Alexander Gattin [EMAIL PROTECTED]
.\
.\ This man page is free documentation; you can redistribute it and/or modify
.\ it under the terms of the GNU General Public License as published by
@@ -20,9 +21,12
tags 334803 unreproducible
thanks
Hi!
On Fri, Oct 21, 2005 at 02:19:18AM +0300, Alexander Gattin wrote:
First `sudo su - newuser` in new xterm/shell is
unsuccessful, I send SIGCONT to su process, exit from
awaken shell and then next `sudo su - newuser` is OK
WRT `suspend -f`...
I can get
Hi!
On Sun, Jan 08, 2006 at 12:11:41AM +0200, Alexander Gattin wrote:
Forgot to say that I have committed the patch. Probably
I'll improve it further, but it's quite usable right
now.
Marc, last shadow release (1:4.0.14-3) includes this
patch. It should give us more clue about what really
Hi!
On Sun, Jan 15, 2006 at 10:03:57AM +0900, Junichi Uekawa wrote:
The main problem is that you are calling '/usr/bin/env'.
As Nicolas has discovered yesterday, -p is ignored
_only_ when su starts a _shell_. Bug itself was in
shell() function.
--
WBR,
xrgtn
--
To UNSUBSCRIBE, email to
Hi!
On Sat, Jan 14, 2006 at 10:22:58PM +0900, Junichi Uekawa wrote:
Methinks we will need to evaluate what needs to be fixed, and how
difficult the fix is, and then proceed further.
Maybe, there is smth. wrong with .*rc of target user's shell?
For example, I see that environment is preserved
Package: dhcp3-server
Version: 3.0.3-5
Severity: wishlist
Hello!
For more than a week I'm fighting with DHCP-DNS update
in our local network. We have Windows machines, *Linux
and *BSD ones. On *Linux I mostly have pump as DHCP
client (for reasons not to be disclosed here I prefer
pump, udhcpc
so far, after minimal tesing, I can confirm that my hosts
ultimately started to register in desired zone.
This claim was wrong. What actually happened is that DHCP request
is recognized as FQDN by dhcp3-server.
WinXP clients send similar requests, although they send both
HN and FQDN options
Hi!
I made preliminary patch which adds support for all 5
modes (5th is made default) through cmdline options and
configfile directives.
so far, after minimal tesing, I can confirm that my hosts
ultimately started to register in desired zone.
Currently patch deals only with sending FQDN/HN,
Hi!
On Sat, Jan 07, 2006 at 09:46:13AM +0100, Christian Perrier wrote:
So, Alex, please go and commit this fix. Don't forget commenting the
quilt patch so that we later remember whether it's worth being
submitted upstream (it probably is anywayas it improves logging)
I think it's not
Forgot to say that I have committed the patch. Probably
I'll improve it further, but it's quite usable right
now.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Hi!
On Fri, Jan 06, 2006 at 06:45:01PM +0100, Christian Perrier wrote:
Alexander, have you made any progress in trying to reproduce this bug
report?
This is su segfaults using encrypted LDAP which needs some LDAP
setup to be worked on...
nope, not yet.
The only additional information I
Hi!
On Fri, Jan 06, 2006 at 06:50:07PM +0100, Christian Perrier wrote:
I absolutely don't know whether someone is still investigating it. I
even don't know if it happens with recent releases of shadow.
I suspect it does. Sometimes. Actually I think the bug
is caused by /var/tmp/utmp file being
Package: pump
Version: 0.8.24-1
Severity: wishlist
Tags: patch
Hi!
I needed to register hostnames in different DNS zones
depending on domain requested by DHCP clients. DHCP
server daemon is dhcp3.
Using interim ddns-update style of dhcp3-server the
problem can't be solved with pump as client (A
Package: i2c-source
Version: 1:2.9.2-1
Severity: normal
Tags: patch
Hi!
i2c-source fails to install/pack modules package
for 2.4.32 because it tries to find smth. under
/lib/modules/2.6.10-debp4/build (or source?) which
is host kernel, not target...
-
make[3]: Leaving
Hi!
On Sun, Dec 18, 2005 at 03:04:00AM +0100, Michael Banck wrote:
On Sun, Dec 18, 2005 at 12:01:41AM +0200, Alexander Gattin wrote:
WRT to the patch you sent -- alternative way to achieve
the same effect would be to make login package
non-required on The Hurd, wouldn't it?
Totally
Hi!
On Sun, Dec 18, 2005 at 12:25:13PM +0100, Michael Banck wrote:
On Sun, Dec 18, 2005 at 12:47:03PM +0200, Alexander Gattin wrote:
On Sun, Dec 18, 2005 at 03:04:00AM +0100, Michael Banck wrote:
On Sun, Dec 18, 2005 at 12:01:41AM +0200, Alexander Gattin wrote:
WRT to the patch you sent
Hi!
On Fri, Dec 16, 2005 at 11:55:58PM +0100, Michael Banck wrote:
On Fri, Dec 16, 2005 at 11:37:40PM +0100, Nicolas François wrote:
We probably also need to remove the passwd dependency on login (loginpam
substvar).
The Hurd package provides login, so the dependency gets resolved
(because
Hi!
On Sun, Nov 06, 2005 at 12:29:43PM +0100, Helge Kreutzmann wrote:
I replaced one terminal of mine with:
9:23:respawn:/sbin/mingetty tty9
and it at least fixes this bug.
yes, mingetty is:
1) for virtual consoles only (i.e. no serial lines
support and thus no bitness-autodetection)
2)
Hi!
On Tue, Oct 25, 2005 at 04:40:53AM -0700, Steve Langasek wrote:
Sorry, what problem are you trying to solve?
Excuse me for the confusion. I talk here about another
issue, similar to what we have in bug #333138.
Now I tend to agree with Robert that this is a bug,
because fact of
Hi!
On Wed, Oct 19, 2005 at 05:36:29PM -0700, Steve Langasek wrote:
Yes, I see this (pre-0.80 version):
1. fprintf(stderr, %s, prompt);
2. tcsetattr(STDIN_FILENO, TCSAFLUSH, ...);
3. read(STDIN_FILENO, ...);
4. tcsetattr(STDIN_FILENO, TCSADRAIN, ...);
OK. Even having 2, 1, 3, 4
tags 334803 confirmed
thanks
Hi!
On Thu, Oct 20, 2005 at 06:49:35AM +0200, Christian Perrier wrote:
Quoting Julian Gilbey ([EMAIL PROTECTED]):
Package: login
Version: 1:4.0.12-6
The issue described in #314727 appears to have reappeared in this
version of the package (currently in
On Fri, Oct 21, 2005 at 01:59:47AM +0300, Alexander Gattin wrote:
Well, this is really broken. Sometimes is hangs,
sometimes suspends.
If I use `sudo su - newuser`, suspend -f works all the
time AFAIS.
Looks like it starts working after first success.
First `sudo su - newuser` in new
is there a way to disable .bz2 at all in apt
in favor of .gz?
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
I have just downgraded the apt pkg from 0.6.41 down
to the 0.5.28.6. This worked even without also
downgrading apt-utils and co.
To the moment, this seems to work, so I'll just pin
apt to stable until the issue is fixed -- sorry, I
don't have enough time to debug it.
--
To UNSUBSCRIBE, email
Hi!
On Sun, Oct 16, 2005 at 03:37:57PM -0700, Steve Langasek wrote:
Anyway, this seems to be fixed upstream in Linux-PAM 0.80,
They took getpass() considered obsolete message to
heart and implemented their own getpass()? :) (shit, I
don't like all these re-inventions of wheel...)
Hi!
On Sun, Oct 16, 2005 at 12:05:18AM -0700, Steve Langasek wrote:
On Sat, Oct 15, 2005 at 06:34:03PM +0200, Christian Perrier wrote:
found that the prompting for a new password is not done by passwd
itself but by the pam_unix PAM module
pam_unix -- misc_conv -- IMHO getpass()
oops, I forgot to mention,
On Mon, Oct 17, 2005 at 01:22:00AM +0300, Alexander Gattin wrote:
Hi!
On Sun, Oct 16, 2005 at 12:05:18AM -0700, Steve Langasek wrote:
On Sat, Oct 15, 2005 at 06:34:03PM +0200, Christian Perrier wrote:
found that the prompting for a new password is not done
Hi!
On Sat, Oct 15, 2005 at 08:40:59AM +0200, Christian Perrier wrote:
No, this is NSS/libc guys' problem in first place, then
ours (with Tomasz).
...
From the above, I understand that this bug should be reassigned to
passwd,glibc.
Yes, we should have 2 bugs one depending on another.
For
Hi!
On Fri, Oct 14, 2005 at 07:09:43PM +0200, Christian Perrier wrote:
This is again a request for more information, to the bug submitter.
Is this bug still relevant in last version of shadow?
I *will* close it in a few weeks if we don't get more news, as
unreproducible.
We had an
Hi!
On Sat, Oct 15, 2005 at 06:34:03PM +0200, Christian Perrier wrote:
While looking at bug #182606 and then to passwd sources, I finally
found that the prompting for a new password is not done by passwd
itself but by the pam_unix PAM module
As a consequence, my first reaction is saying
Hi!
On Thu, Oct 13, 2005 at 12:12:37AM +0200, Piotr Engelking wrote:
BTW, this is _not_ _required_.
I was referring to chapter 12.7 of the Debian Policy Manual: If an
upstream changelog is available, it should be accessible as
/usr/share/doc/package/changelog.gz in plain text.
Well, this
1 - 100 of 209 matches
Mail list logo