Package: db4.7-util
Version: 4.7.25-8
Severity: minor
Tags: patch
File: /usr/share/man/man1/db4.7_load.1.gz
I think standart input is a bit more meaningful.
--- db4.7_load.12009-11-22 23:45:38.0 +0300
+++ db4.7_load.1.patched2009-11-22 23:46:23.0 +0300
Package: libcap-ng0
Version: 0.6.2-3
Severity: normal
One line prior to last line of description states:
This package contains header files and libraries for libcap-ng.
Doesn't header files usually reffers to *.h include files?
--
sobtw...@sdf.lonestar.org
SDF Public Access UNIX
Package: debhlper
Version: 7.4.3
Severity: wishlist
Version 2.9.27 of libc6 has the following item in
changelog.Debian.gz:
Strip *.o files manually (dh_strip does not do it) to prevent
leakage of the build directory.
According to
This should solve it:
--- /etc/init.d/nut 2009-07-27 02:50:53.0 +0200
+++ debian/nut.init 2009-11-05 19:35:36.0 +0200
@@ -160,7 +160,7 @@
poweroff)
flag=`sed -ne 's#^ *POWERDOWNFLAG *\(.*\)$#\1#p' /etc/nut/upsmon.conf`
-wait_delay=`sed -ne 's#^
Package: iptables
Vesrion: 1.4.4-2
Severity: minor
File: /usr/share/man/man8/iptables.8.gz
That man page has an obscure single f letter:
length
This module matches the length of the layer-3 payload (e.g. layer-4
packet) f a packet against a specific value or range of
Package: vsftpd
Version: 2.2.0-3
Severity: wishlist
The name of the directory /usr/share/doc/vsftpd/examples is a bit
confusing. More so when there is also upstream EXAMPLE directory.
Please rename it to debconf or debconf-examples or something.
--
sobtw...@sdf.lonestar.org
SDF Public
Package: db4.7-util
Version: 4.7.25-8
Severity: minor
File: /usr/share/man/man1/db4.7_load.1.gz
--- db4.7_load.12009-10-18 12:28:52.0 +0200
+++ db4.7_load.1.fix2009-10-18 12:28:28.0 +0200
@@ -72,7 +72,7 @@
\fBkeys\fR is specified as set, the input must be paired
That section also has an example.
I think the patch that I provided is more clear and more common.
--
sobtw...@sdf.lonestar.org
SDF Public Access UNIX System - http://sdf.lonestar.org
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe.
On Sat, Sep 12, 2009 at 04:15:48PM +0100, Bradley Smith wrote:
Gah sorry, I misread the bug report..
1) The problem seem to appear because
ioctl(i8k_fd, I8K_BIOS_VERSION, args))
return value is less then 0.
$ grep -A2 I8K_BIOS_VERSION i8kutils-1.32/i8kctl.c
if ((rc=ioctl(i8k_fd,
Running i8kctl with the following patch yeilds:
$ i8kctl
ioctl failed: Invalid argument
ioctl failed: Invalid argument
ioctl failed: Invalid argument
1.0 (null) 6WNHA 44 2 2 -1 -1 0 0
The patch is:
--- i8kutils-1.32/i8kctl.c 2009-08-22 14:27:09.0 +0300
+++
Not just printing the same message 3 times. Tagging it (Error
or Warning) and specify the faulty ioctl.
$ i8kctl
Warning: ioctl I8K_BIOS_VERSION failed: Invalid argument
Warning: ioctl I8K_GET_SPEED failed: Invalid argument
Warning: ioctl I8K_GET_SPEED failed: Invalid argument
1.0 (null) 6WNHA
Package: i8kutils
Version: 1.32
Severity: minor
Tags: patch
I think the following is the more common structure in debian/rules
for debug,noopt,nostrip alternatives:
--- debian/rules2009-09-17 20:36:51.0 +0100
+++ debian/rules2009-09-17 20:40:15.0 +0100
@@ -1,12
Package: i8kutils
Version: 1.32
Severity: Normal
Tags: Patch
In view of the current content of www.debian.org/~dz/i8kutils,
I suggests the following patch:
--- README.i8kutils 2009-09-12 16:57:22.0 +0100
+++ README.i8kutils.mod 2009-09-12 17:11:41.0 +0100
@@ -43,7 +43,7 @@
Package: i8kutils
Version: 1.32
Severity: Normal
The proc file does know the version, namely A14:
$ cat /proc/i8k
1.0 A14 6WNHA 45 2 2 -22 -22 -1 0
Why the utility doesn't report it? It states (null):
$ i8kctl
1.0 (null) 6WNHA 45 2 2 -1 -1 0 0
Is the last '0' in that
Just reporting that I have the same problem with 0.4.11.11-1+b2.
--
sobtw...@sdf.lonestar.org
SDF Public Access UNIX System - http://sdf.lonestar.org
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact
Qouting debconf-devel.7
Avoid outputting anything to stdout in your postinst, since
that can confuse debconf, and postinst should not be verbose
anyway. Output to stderr is ok, if you must.
--- portmap.postinst2009-08-10 01:18:51.0 +0100
+++ portmap.postinst
I was upgrading from 6.0-10 to 6.0-11, and had exactly the same
situation that was decribed by the original submitter of the bug. It
could be that I had it on earlier upgrades, but only noticed at this
upgrade.
--
sobtw...@sdf.lonestar.org
SDF Public Access UNIX System -
Package: debconf-doc
Version: 1.5.27
Severity: wishlist
It took me a while to get the suggestion to debug with debconf(1)
right. Hopefully, the following will help others:
--- debconf-devel.7 2009-08-22 01:53:08.0 +0100
+++ debconf-devel.7.new 2009-08-22 03:12:22.0 +0100
It took me a while to see what this is all about. I hope I am not
the only one that have that difficulty. For the benefit of others,
the problem is that there should be an `i' character instead of a `1'
character. The affix for the package name is `i8k', not `18k'.
--
Package: i8kutils
Version: 1.31
Severity: normal
Tags: patch
Files: /usr/share/doc/i8kutils/README.i8kutils
I understand that the following statement was written by the
prevoius maintainer. I find highly inappropriate to keep it as it is:
--- README.i8kutils.orig2009-08-20
I read the file more carefully. The header is for 1.28. The footer
has the email address of the prevoius maintainer. Since he doesn't
wish to recieve any messages for the utilities I think that that
pragraph must make it clear that the statement only refer to the
single person whose email
Package: i8kutils
Version: 1.31
Severity: minor
Tags: patch
--- etc/default/i8kmon2009-05-26 17:42:52.0 +0100
+++ etc/default/i8kmon 2009-08-20 03:51:21.0 +0100
@@ -1,4 +1,4 @@
# /etc/default/i8kmon
-# Change to one enable i8kmon
+# Change to one to enable i8kmon
2.4.1-3 has:
# re process nut.conf MODE so that it can be sourced
NUT_MODE=`sed -ne 's#^ *MODE = *\(.*\)$#\1#p' /etc/nut/nut.conf`
sed s/^MODE\(.*\)/MODE=$NUT_MODE/ /etc/nut/nut.conf
/etc/nut/nut.conf.new
mv /etc/nut/nut.conf.new /etc/nut/nut.conf
Those lines are buggy.
I
The prevoius patch contains spaces around shell variable settings.
This can confuse new users:
--- nut.conf.5 2009-08-16 01:29:51.0 +0100
+++ devel/debian/nut/nut-2.4.1/man/nut.conf.5 2009-08-16
01:28:42.0
+0100
@@ -70,19 +70,23 @@
# /etc/nut/nut.conf. See nut.conf(5)
--- debian/nut.preinst 2009-07-15 22:29:32.0 +0100
+++ debian/nut.preinst.new 2009-08-16 03:12:48.0 +0100
@@ -2,7 +2,7 @@
if [ $1 = upgrade ]; then
# remove the obsolete udev file
-[ -f /etc/udev/rules.d/025_nut-usbups.rules ] rm -f
Perhaps you should just grep the whole line, and then use tr,
instead of sed, for the actual fix?
$ for str in MODE = standalone MODE= standalone MODE
=standalone; do printf $str\t\t; printf $str\n | tr -d ;
done
MODE = standalone MODE=standalone
MODE= standalone
The original submitter wrote:
3) The permissions I set on the serial port, that is connected to
the UPS, were modified
A possible reason is udev.
$ zcat /usr/share/doc/udev/NEWS.Debian.gz | head
udev (0.140-1) unstable; urgency=low
Starting from this release the
Package: nut
Version: 2.4.1-3
Severity: Normal
Tags: patch
udev files that are added on preinst are not removed on purge:
--- debian/nut.postrm 2009-08-15 03:11:28.0 +0100
+++ debian/nut.postrm.new 2009-08-15 03:14:22.0 +0100
@@ -12,6 +12,12 @@
;;
purge)
+[
With rm -f, there is no need to test for existance:
--- debian/nut.postrm 2009-08-15 03:14:22.0 +0100
+++ debian/nut.postrm.new 2009-08-15 03:56:36.0 +0100
@@ -5,19 +5,15 @@
remove)
# remove sample file if those were previously (wrongly) installed
for file
Package: bugs.debian.org
Severity: normal
the following was in a response from the bts:
found 539230 1:9.6.1.dfsg.P1-1
Bug #539230 [bind9] debian/control: s/libdb-dev (4.6)/libdb-dev
( 4.6)/
There is no source info for the package 'bind9' at version
'1:9.6.1.dfsg.P1-1'
Package: bind9
Version: 1:9.6.1.dfsg-2
Severity: normal
Tags: patch
Files: /etc/init.d/bind9
The init script uses a hard coded PIDFILE. It overrides possible
setting of pid-file by the local administrator. This patch
somewhat remedy this problem:
--- debian/bind9.init 2009-07-29
Package: pbuilder
Version: 0.189
Severity: normal
The warning below is indented by me.
...
Setting up cdebootstrap (0.5.4) ...^M
Setting up pbuilder (0.188) ...^M
I: Setting DEBBUILDOPTS=
- Attempting to satisfy build-dependencies
- Creating pbuilder-satisfydepends-dummy package
Package:
On one hand:
$ zgrep 212659 /usr/share/doc/bind9/changelog.Debian.gz
* upstream now uses a bind subdir. Closes: #212659
and on the other:
$ grep -A3 2486 CHANGES
2486. [func] The default locations for named.pid and lwresd.pid
are now
1. It could be that the logging problem that was reported by shaul Karl
was beacuse of tcpd.
2. upsd.c:check_command:
2.1 I suggest to add a upsdebugx call when hosts_access returns 0.
2.2 upsd.c:server_load add 127.0.0.1 when no LISTEN addres has
been specified. In that case,
The following implements my suggestions from the prevoius message:
--- server/upsd.c 2009-07-26 19:19:22.0 +0300
+++ server/upsd.c.new 2009-07-27 01:16:34.0 +0300
@@ -69,6 +69,7 @@
/* default is to listen on all local interfaces */
static stype_t *firstaddr =
Package: nut
Version: 2.4.1-3
Severity: normal
Tags: patch
With
$ ps axf | grep /sbin/ups
9442 pts/4S+ 0:00 \_ grep /sbin/ups
16096 ?Ss 0:01 /sbin/upsd
16099 ?Ss 0:00 /sbin/upsmon
16100 ?S 0:01 \_ /sbin/upsmon
I get
#
Package: gdb
Version: 6.8-3
Severity: wishlist
I think that
http://sources.redhat.com/gdb/onlinedocs/gdb.html#SEC_Top
is great to get someone quickly into gdb. I wish it would be
specifically mentioned in /usr/share/doc/gdb/README.Debian .
--
sobtw...@sdf.lonestar.org
SDF Public
Package: libsensors4
Version: 1:3.1.1-2
Severity: wishlist
There is already a sensors.d directory under /etc/. Having the
main configuration in it seems to me a less clattered /etc. And
while at it, /etc/sensors.d might be renamed /etc/sensors.
--
sobtw...@sdf.lonestar.org
SDF Public Access
I have modified a bit the patch that the original submitter proposed:
--- lsof.8.orig 2009-07-20 04:58:52.0 +0300
+++ lsof.8.new 2009-07-20 05:16:02.0 +0300
@@ -127,31 +127,31 @@
by user ``foo''.
The exceptions are:
.IP
-.TP \w'1) 'u
+.TP \w'1)\ 'u
1)
the `^' (negated) login
Package: console-data
Version: 1.09-1
Severity: normal
Tags: patch
debian/attic/console-data.config uses
dpkg --print-installation-architecture
, which is obsolete and prints a warning to the user.
$ dpkg --print-installation-architecture
dpkg: warning: obsolete option
Package: nut
Version: 2.4.1-2
Severity: normal
According to dh_clean(1), -k is deprecated, use dh_prep(1) instead.
There are 2 instances of dh_clean -k.
--
sobtw...@sdf.lonestar.org
SDF Public Access UNIX System - http://sdf.lonestar.org
--
To UNSUBSCRIBE, email to
Package: www.debian.org
Severity: normal
One of the assorted links at the bottom of www.debian.org/devel
is `Debian OID hierarchy'. That link is broken. It pints to a
404 Not Found.
--
sobtw...@sdf.lonestar.org
SDF Public Access UNIX System - http://sdf.lonestar.org
--
To UNSUBSCRIBE, email
Package: manpages-dev
Version: 3.21-1
Severity: normal
Tags: patch
File: /usr/share/man/man2/vfork.2.gz
--- vfork.2.orig2009-06-21 03:53:54.0 +0300
+++ vfork.2 2009-06-21 03:54:52.0 +0300
@@ -83,7 +83,7 @@
in that the parent is suspended until the child terminates
Package: ftp.debian.org
Severity: normal
It looks to me that chagelogs of packages not getting updated when
new versions are accepted. Are you aware of that? Is that intended?
For example, reportbug 4.4 is in testing, while
/changelogs/pool/main/r/reportbug/current points to 4.2
--
Package: manpages-dev
Version: 3.21-1
Severity: normal
File: /usr/share/man/man3/strcpy.3.gz
strcpy mentions strpcpy. Does strpcy has a manpages?
--
sobtw...@sdf.lonestar.org
SDF Public Access UNIX System - http://sdf.lonestar.org
--
To UNSUBSCRIBE, email to
--- environ.7 2009-05-16 16:21:37.0 +0300
+++ environ.7.mod 2009-05-16 16:30:18.0 +0300
@@ -52,6 +52,10 @@
This array of strings is made available to the process by the
.BR exec (3)
call that started the process.
+The last element of the array is a NULL pointer.
+There
Package: manpages
Version: 3.21-1
Severity: normal
Quoting environ.7 BUGS, 1st paragraph:
Clearly there is a security risk here. Many a system command has
Isn't `Many a system' bad English?
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of
Package: nut
Version: 2.4.1-2
Severity: normal
Tags: patch
The *.sample files that are written to /etc/nut by the deb are marked
as conffiles even though the application does not read them.
I suggest to:
1. Remove the .sample suffix
2. Have the distributed files also in
--- Makefile.in 2009-02-17 11:24:10.0 +0200
+++ Makefile.in.new 2009-05-12 10:42:52.0 +0300
@@ -236,8 +236,8 @@
upscli_splitname.3 upscli_strerror.3 upscli_fd.3 upscli_get.3 \
upscli_list_next.3 upscli_list_start.3 upscli_ssl.3 \
upscli_upserror.3
49 matches
Mail list logo
