Le 2014-9-28 14:47, Joey Hess a écrit :
Christian Hudon wrote:
whoami="$(whoami)"
shorthostname="$(hostname -s)"
fullhostname="$(hostname -f)"
domainname="${fullhostname#*.}"
These comm
Le 2014-9-28 13:04, Joey Hess a écrit :
Christian Hudon wrote:
# Check hostname
if [ "$shorthostname" = "$host" -o
"$fullhostname" = "$host" -o \
"
Package: mr
Severity: wishlist
I'm using an improved version of the on() function in my .mrconfig,
which handles usernames, etc. Maybe you'd like to include it in the
defaults for mr, so people don't have to reinvent the wheel. Here it is:
# Checks whether the given host basename i
Package: mr
Severity: wishlist
I'm using mr to manage (among other things) my ~/Code directory, where
all the version control repositories for code I'm interested in live. I
currently have 35 repositories listed for the .mrconfig for that
directory, and more are added every month, so I don't want
Package: mr
Version: 1.11
Running "mr register" on a bzr checkout which has related branches
fails. My guess is that mr is confused by the related branches which
appear in the output of "bzr info". Running a similar "mr register" on a
bzr checkout without related branches succeeds.
Here is a
Package: dpkg
Version: 1.16.2
Severity: minor
I was trying (and failing) to start a Python paster instance inside a
virtualenv like this:
start-stop-daemon -u useridhere --chdir /srv/virtualenv_path --exec
bin/paster --start -- serve development.ini
It took me a while to figure out what the
Yes, this works great. Thanks!
I went through the relevant parts of the mr code for the bzr commands
and found something that I had missed. For the "record" command, when
working on a bazaar checkout, the command to run is "bzr commit
--local". Other than that, it would be nice to add a short note
Package: mr
Version: 1.04
Severity: minor
I noticed that mr uses "bzr clone" to retrieve a repository. This is a
deprecated (as of bzr 2.4) alias for the "branch" command. The "bzr
clone" calls should be replaced by "bzr branch".
Thanks,
Christian
--
To UNSUBSCRIBE, email to debian-bugs-d
Package: mr
Version: 1.04
Severity: wishlist
Bazaar supports a concept of "checkouts", which basically keeps a
designated remote repository in sync with the local one. For example,
running "bzr update" on a checkout will automatically pull from the
remote repository, and doing a commit will automa
Package: r-core-case
Version: 2.13.1-1
Severity: wishlist
Hello,
The Debian packages for R add a "/usr/local/lib/R/site-library"
directory into which the local sysadmin can install packages, which is
really useful. However, compiled R packages are not always compatible
between versions of R (e.g.
Package: adduser
Version: 3.102
Severity: wishlist
Hi,
The installed /etc/adduser.conf file features most of the variables that
can control adduser, at least in commented out form, together with an
explanatory comment. This is really nice for the sysadmin, as it saves
back and forth between the d
Package: cuetools
Version: 1.3.1-4
Severity: important
Tags: patch
Hi,
I thought I had submitted this bug to Debian too, but I might have
forgotten. When converting .toc files to .cue files, I get a "buffer
overflow detected" warning and a core dump.
The following patch fixes the buffer overflow
Package: apt
Version: 0.7.26
Severity: wishlist
Hi,
I've been wishing more than once for a version of the "install" command
that works as follows: it installs the new version of the package only
it said package was already installed on the computer. If the package
was not installed, it does nothi
Package: slurm-llnl
Severity: wishlist
Upstream version 2.1 of Slurm was released about a week ago. It
introduces useful improvements in the scheduler, among others:
https://computing.llnl.gov/linux/slurm/news.html#21
It'd be nice if the Debian packages were upgraded to this version.
Thanks!
Package: wnpp
Severity: wishlist
Blcr is a process checkpointing solution for Linux. More information is
available at: https://ftg.lbl.gov/CheckpointRestart/CheckpointRestart.shtml
Extract from the web site:
Future Technologies Group researchers are developing a hybrid
kernel/user implementa
Package: apt
Severity: wishlist
Hello,
Sometimes, I need to cut and paste the contents of a Depends field (from
running "dpkg -s", etc.) or from a Build-depend line (from the control
file of a .deb whose source I just unpacked) to install packages
required by a package that I downloaded separ
Package: mount
Version: 2.13.1.1-1
Severity: wishlist
It'd be nice if the mount package included a copy of the
sharedsubtree.txt file from the Linux kernel documentation in
/usr/share/doc/mount. Currently, it's impossible to understand what the
--make-private, etc. options do from the mount ma
Package: grep
Version: 2.5.3~dfsg-6
Severity: minor
The manpage for grep contains the following about the --color option:
--color[=WHEN], --colour[=WHEN]
Surround the matched (non-empty) strings, matching
lines,
context lines, file names, line numbers
Package: mdadm
Version: 2.6.7-1
Severity: wishlist
It'd nice if the help printed out when running mdadm -G --help also
mentioned the "max" parameter that can be passed to the -z option. Just
a one-line mention would be fine. (I know it's already in the manpage...
I found it trying to come up b
Dirk Eddelbuettel wrote:
Well, first off, congrats for finding it :) Second, you didn't have too,
this is all documented in help(Startup) [ though I agree that finding
help is not always easy; OTOH doing help.search("Rprofile") would have
lead you there, along with some false positives ]
Debian Bug Tracking System wrote:
This is an automatic notification regarding your Bug report
which was filed against the r-cran-cairodevice package:
#498527: Cairo should be default output device in R instead of X11 when
r-cran-cairodevice installed
It has been closed by Dirk Eddelbuettel <[E
Package: r-cran-cairodevice
Version: 2.8-1
It'd be really nice if the r-cran-cairodevice package was set up so that
the R output device would default to 'Cairo' instead of 'X11' when
r-cran-cairodevice is installed. That way, things would Just Work,
instead of having to hunt for how to enable
Package: r-base-core
Version: 2.7.1-1
Local R customizations by the sysadmin to the R startup are supposed to
be done in the Rprofile.site file. However, R currently looks for that
file under /usr (more precisely: /usr/lib/R/etc/Rprofile.site). It'd be
nice to have the same setup as for other
Jurij Smakov a écrit :
> It appears that the intention of the maintainer was to build on most
> (all?) architectures with -O0 to avoid problems (as there is a longish
> trail of icon bugs on various architectures, related to its inability
> to work properly when compiled with -O2). However, due
Jérémy Bobbio escribió:
> Package: icon
> Version: 9.4.2-2.7
> Severity: serious
> Justification: no longer builds from source
>
> Hi!
>
> icon currently fails to build from source in sid [1]:
>
Which architecture are you building on?
Christian
--
To UNSUBSCRIBE, email to [EMAIL PROTECTE
Filipus Klutiero escribió:
> Hi, the maintainer of ethereal requested more information from you about this
> bug. Please see his mail and reply to his question.
>
Hi,
I tried to reproduce the bug by removing gnome-icon-theme, but I
couldn't. I don't know what the problem was when installing it
I went to have a look at the Info-zip FAQ page, and the method used in
Chris Chiappa's patch is mentionned for handling zip files less than 2GB
that create uncompressed data files between 2GB and 4GB. From
http://www.info-zip.org/FAQ.html#limits:
In the short term, it is possible to improve Zi
Package: tdsodbc
Version: 0.63-2
Installing tdsodbc is not possible when answering yes to the debconf
question about registering ODBC stuff, unless the odbcinst1 package is
also installed. tdsodbc should depend on odbcinst1 if it's using
programs from that package in its postinst.
I also thi
Michael Meskes wrote:
Could you please add your local config to the bug report? The default
Subject should be "Over quota", but can be configured differently using
/etc/warnquota.conf. So unless you have a meaningful entry in
warnquota.conf but still no subject, I do not see a bug here.
I loo
Package: quota
Version: 3.12-6
Severity: wishlist
Right now when the quota system sends emails, it leaves the subject
empty. It'd be nice if it supplied a descriptive subject line instead,
so I'd see something like "Over quota on /home/whatever" or "Over quota
on /foo, last warning!", which wo
Package: apcupsd-doc
Version: 3.10.17-2
apcupsd-doc does not register its texinfo documentation using
install-info, which means it does not show up in info's top-level (dir)
page.
Registering the texinfo documentation would make it much easier to find.
Christian
--
To UNSUBSCRIBE, email
Package: matplotlib
Version: 0.82-5
A new version of matplotlib (0.85) is available from upstream.
Christian
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Roger Leigh wrote:
On Fri, Sep 16, 2005 at 11:15:05AM +0200, Marc 'HE' Brockschmidt wrote:
During the Debian QA meeting hold during Sept. 09th till 11th, we
decided that looking at packages that haven't been uploaded for a very
long time could cover up some QA problems.
I've done this now an
Dan Jacobson a écrit :
Well, in
SYNOPSIS
pppd pty 'pppoe [pppoe_options]' [pppd_options]
pppoe -A [pppoe_options]
we see that just plain pppoe with no arguments is illegal.
So someone should add it to the SYNOPSIS, at least.
It wouldn't hurt to also mention what it does.
Actua
Jeroen van Wolffelaar wrote:
>On Mon, Aug 15, 2005 at 01:14:45AM +0200, Matthias Klose wrote:
>
>
>A number of libc5 related packages still depend on it though, and
>should be removed together.
>
>ld.so: The Linux dynamic linker and library for libc4 and libc5.
>libc: libc5
>libdb: libdb1
>libg
Package: libxmlrpc-c3
Version: 0.9.10-4
This version of the library is getting rather old... Version 1.0 was
released on 2004-10-25. The current version is 1.03.01, released a
couple of days ago. An updated Debian package would be nice.
Thanks,
Christian
--
To UNSUBSCRIBE, email to [EMAIL P
Jurij Smakov wrote:
Removing capabilities with lcap on a 2.6.11 kernel without capability.ko
module loaded does not produce any warning or error. In fact, it erroneously
indicates that the capability was successfully removed, creating a false
impression of security.
Can you please post the .c
Marco d'Itri wrote:
On May 16, Marc Dequènes <[EMAIL PROTECTED]> wrote:
I forgot to decrease mtu of one of my boxes to 1492 (because i'm using
IPv6), thus causing pppoe to die with the following error when doing a
simple "apt-get update" from that box :
Interesting, but you should really us
Package: python-rpy-doc
Version: 0.4.1-2
The way the python-rpy-doc package is done now, it's not much more
useful than the rpy source package. There's just the raw texinfo source
together with a Makefile. It'd be nice if the texinfo source was
installed and registered with install-info so info
Package: coreutils
Version: 5.2.1-2
Severity: minor
In the ls manpage (and in its info manual too, I presume), option -f is
described as:
-f do not sort, enable -aU, disable -lst
However looking in the neighborhood of option -s, we see:
-s, --size
print size of each f
Package: python-numarray
Version: 1.1.1-3
FYI, the latest upstream version of numarray is 1.2.3.
http://sourceforge.net/project/showfiles.php?group_id=1369&release_id=223264
Christian
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Package: isakmpd
Version: 20041012-1
Severity: minor
The description for the -L option in the manpage does not make it clear
where the capture output is send:
-L Enable IKE packet capture. When this option is given, isakmpd
will capture to file an unencrypted copy of the neg
Reuben Thomas wrote:
It's easily possible that there are still bugs in Zile's SIGWINCH
handling, but it's not as simple as this. It works fine for me (now
using 2.0.2 of course, but there's no change to this code since
2.0.1). If the reporter could give more precise indications of the
problem t
Package: zile
Version: 2.0.1-1
When resizing a terminal window with a zile editing session running
inside it, it'd be really nice if zile reacted to the SIGWINCH signal
sent by the terminal on a resize and resized itself also to fit in the
terminal window's new size.
Christian
--
To UNSUBSCRI
Package: ethereal
Version: 0.10.9-1
I set up my ssh to run with "ForwardX11Trusted no". All the programs
I've tried run fine when run over a X11 connection forwarded via ssh,
except ethereal, which fails with the following error message:
The program 'ethereal' received an X Window System error.
Package: fam
Version 2.7.0-5
Today on one of our machines famd was taking up >97% of the CPU. I
attached to it with strace and I saw a continuous stream of:
write(2, "*** overflow famqueu"..., 51) = 51
rt_sigreturn(0xbfffecd0)= 0
--- SIGRT_1 (Real-time signal 0) @ 0 (0
Package: gzip
Version: 1.3.5-9
Severity: minor
The --rsyncable option appears when you do gzip --help, but it isn't
mentioned anywhere in the manpage. It'd be nice to have it appear in the
manpage too.
Christian
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Tro
Package: ipython
Version: 0.6.5-1
Severity: major
The new python2.3 split off the profiler module into its own package
(into non-free). When trying to run ipython with that new python package
installed, I get:
Traceback (most recent call last):
File "/usr/bin/ipython", line 26, in ?
import I
The "2.0a1" in the version number stands for alpha, so I'd be in no big
hurry to get a Debian package of that. But the latest stable version is
1.2, not 1.0. It'd be nice to have an update of the Debian package.
Christian
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubsc
Package: python2.4
Version: 2.4-3
Python2.4 recommends the packages python2.4-iconvcodec,
python2.4-japanese-codecs and python2.4-korean-codecs, however these
packages do not exist.
I seem to vaguely remember the CJK codecs being folded into the core
python package for python 2.4, so that coul
Package: python2.4
Version: 2.4-3
Python2.4 recommends the packages python2.4-iconvcodec,
python2.4-japanese-codecs and python2.4-korean-codecs, however these
packages do not exist.
I seem to vaguely remember the CJK codecs being folded into the core
python package for python 2.4, so that could
Package: lvm2
Version: 2.00.32-1
I'm using a setup with lvm over RAID1 (using the Linux software md
driver). I set up the machine using an old version of lvm2, etc. a year
ago and everything worked fine. But now (with upgraded tools) I can't
seem to create a VG on the new RAID array I just insta
Package: libcupsys2-gnutls10
Version: 1.1.23-2
Because all the conffiles from cupsys retain the $Id$ CVS strings from
upstream, pretty much every cupsys upgrade asks me "conffile x has
changed, what do you want to do?" even when the only change is that the
cvs id went from 1.8 to 1.9.
Please re
Package: gv
Version: 3.5.8-36
I create a Postscript file using (say) a2ps: a2ps foo.txt -o foo.ps
When I view the Postscript file using "gv foo.ps", gv selects the
correct paper format (letter) that the Postscript file was created in.
When I view the same Postscript file by piping it to gv ("cat
Package: eject
Version: 2.0.13deb-7
We have a directory /cdrom, which is intended to be a mountpoint for
CD-ROMs. When trying to eject an audio CD, the eject program's heuristic
picked /cdrom as the filename to open and then to try its various ioctls
to eject stuff. Needless to say, this doesn't
Package: libnspr-dev
Version: 1.7.3-5
Severity: wishlist
It'd be really useful when debugging programs based on NSPR to have the
debug version of the libraries available, either in libnspr-dev or in a
separate -dbg package.
Christian
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject
Package: subversion
Version: 1.1.1-2
Severity: minor
A new upstream version (1.1.2) is available, which includes several
bugfixes that would be nice to have...
Christian
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Patrick Caulfield wrote:
It might be that you are using LVM1 format metadata - which still has this
restriction.
vgconvert will upgrade that metadata to LVM2 and remove the limit.
(BTW the default of 4MB PE size is years old on LVM1 (25/10/2002),
and has never been a limitation in LVM2)
Hmm. I
Package: lvm2
Version: 2.00.24-1
I guess 256GB was a nice large number the for default max vg size when
lvm1 was started, but now that you can buy a single hard drive that is
twice this size, it seems to me to be a bit on the small side for a
default max vg size value. Especially given that ther
59 matches
Mail list logo
