Package: gespeaker
Version: 0.8.6-1
Tags: patch
Dear Maintainer,
Per xqqy's "gespeaker fails to start" bug report below, I'm attaching
a patch that resolves the error for me.
On Tue, 05 Jan 2021 21:27:20 +0800 xqqy wrote:
> Install this package with'sudo apt install gespeaker' command and
Package: timidity
Version: 2.14.0-8
Followup-For: Bug #578859
Dear Maintainer,
As an addendum to the last email, I realized some of the (working)
behavior I experienced disappeared when I restarted my computer.
The attached Rosegarden-alsa startup script seems to work after a
fresh restart, while
Package: timidity
Version: 2.14.0-8
Followup-For: Bug #578859
Dear Maintainer,
Attached is a patch that allows timidity to play via JACK output in
v2.14.0-8. This feature was lost at some point in the last few years
(at least as far back as 2014). When timidity is broken, attempting to
use it p
Sunil Mohan writes:
> @Nick, we would like to comment on the following solution: [...]
>
> I propose that we take advantage of systemd's network name
> predictability scheme and fix this properly.
That all sounds really reasonable. What do you need from me for this?
pgpwtK3d8udjW.pgp
Descript
One last edit, attached: this build'll succeed even when the TTY
device is completely unavailable (move "open /dev/tty" into the try
block).
vmdebootstrap-build-without-tty-1.patch
Description:
> On Mon, 15 Dec 2014 01:32:00 +0000 Nick Daly
wrote:
> > 1. BuildBot Fixes: Attempts to build with a TTY device, and falls
> > back to headless mode if TTY device is unavailable. This change
> > allows BuildBot builds to complete.
>
> On Tue, Dec 23, 2014, 11:05 Neil
Hello Neil, apologies for the long delay on this. I've attached a patch
that solves this bug and two other bugs:
1. BuildBot Fixes: Attempts to build with a TTY device, and falls back to
headless mode if TTY device is unavailable. This change allows BuildBot
builds to complete.
2. Add customize
Hi, I'll try to reproduce this again this weekend, I've only been able to
reproduce this regularly when attempting to setup MS on my Jessie Raspberry
Pi (I believe it was an image produced by Bdale's Freedom-Maker), but I
recently hosed my Pi, so I'll have to start it over from scratch again.
Nick
As a follow-up, here's a much smaller version of the above patch.
Thanks go to Petter for pointing me to the maintainer scripts
documentation.
diff -r 9e5e6d9f91b1 debian/monkeysphere.postinst
--- a/debian/monkeysphere.postinst Sat Jun 28 14:54:59 2014 -0500
+++ b/debian/monkeysphere.postinst Sat
diff -r 9e5e6d9f91b1 debian/monkeysphere.preinst
--- /dev/null Thu Jan 01 00:00:00 1970 +
+++ b/debian/monkeysphere.preinst Sat Jun 28 14:55:33 2014 -0500
@@ -0,0 +1,9 @@
+#! /bin/sh
+
+# preinst script for monkeysphere.
+
+# Author: Nick Daly
+# Copyright: N/A
+
+preinstall-monkeysphere-version=`monkeysphere --version`
+export preinstall-monkeysphere-version
Hi folks,
What still needs to be done on this bug to resolve it? If I understand
correctly, Martin's patch [0] in message #49 resolves the broken
behavior, correcting the grave breakage in this bug. If the
hotplug-before-resolve bug still needs to be handled, could we open a
new bug for that, or
Antonio Terceiro writes:
> What's wrong with running chmod on the built image as part of your
> automated build?
That was my first thought. However, because root owns the image, I'd
have to either:
- Build the image as root to chmod the image, or,
- Give the building user sudo-chmod privileges
Hi, this is just a note to mention that the previous patch attached to
the 2014-05-17 email is somewhat incorrect and was less well tested than
it should've been. I'm currently revising and retesting the patch and
should submit a corrected patch in the next day or so.
Thank you,
Nick
--
To UNS
The attached vmdebootstrap patch should resolve Pere's concerns:
- We now attempt to run in interactive mode, however, if that fails, we
fall back to non-interactive mode.
- Also, we now run everything inside a chroot to prevent poorly written
customize scripts from breaking the host system.
The previously mentioned patch is attached to this email. Apologies for
the noise.
Nick
--- vmdebootstrap 2014-05-11 13:16:27.948314432 -0500
+++ /usr/sbin/vmdebootstrap 2014-05-10 14:26:22.0 -0500
@@ -100,6 +97,9 @@
self.settings.boolean(['sudo'],
Package: vmdebootstrap
Version: 0.2-2das-boot2
Severity: normal
Dear Maintainer,
When vmdebootstrap completes, the generated images are always owned by
the root user. This is very inconvenient for automated builds which
need to be able to operate on the built image file. The attached patch
add
Petter Reinholdtsen writes:
> How will this patch affect packages trying to ask for answers from the
> user using debconf? I suspect it will break in this case.
Should packages really be asking interactive debconf questions during
the (customize stage of the) build process? I thought packages
Package: vmdebootstrap
Version: 0.2-2
Severity: normal
Dear Neil, Lars, and Antonio,
When trying to build a Freedom-Maker [0] image with buildbot, the image
fails to build without access to `/dev/tty` [1]. If necessary, the
buildbot instructions can be provided. The `/dev/tty` access is
unneces
Package: debian-maintainers
Severity: normal
Tags: patch
Dear Maintainer Maintainers,
Please add me as a Debian Maintainer. Jetring changeset attached.
Thank you for your time,
Nick
-- System Information:
Debian Release: 7.3
APT prefers stable-updates
APT policy: (500, 'stable-updates'), (
g. If VMs failed to suspend, they
# were powered off. If the VMs didn't power off, that user's VirtualBox process
# was killed. If VirtualBox processes failed to die, all users' VirtualBox
# processes are killed. If even the root user couldn't guarantee all VMs were
# stop
Why is http://anonscm.debian.org/hg 404ing? It's linked from the main page.
I suspect this bug might be a duplicate of bug 649707. For me, there
were several (reproducible) instances where StumpWM would crash and be
rendered unusable, that seem to have been corrected by updating to the
latest Git checkout [0].
StumpWM would crash when I:
- Opened a new image in Gimp.
-
The fix noted in the loopback-email [0] is correct. I was unable to
install mysql-server [1]. However, the mysql-server package installed
correctly after running:
# ifdown lo
# ifup lo
# dpkg --configure -a
This issue occurred on Wheezy.
Should the mysql-server ensure the loopback
Hi, this bug has been unattended for 5 years now. Is there any reason
uw-imapd still conflicts with the harden-servers package?
The Panda-IMAP website suggests that UW-IMAP 2007f has one security fix
that vanilla 2007e does not have (but perhaps that fix was packaged in
the 4 intervening years?):
Subject: bittorrent: Should operate as a background system service
Package: bittorrent
Version: 3.4.2-11.3
Severity: normal
Tags: patch
*** Please type your report below this line ***
Dear BitTorrent maintainers,
It's not currently possible to run a BitTorrent tracker as a system
service, like Ap
Dear bug 555296 subscribers,
Recently, I ran into this bug, where a host system cannot resume from
hibernate-to-disk when a VirtualBox VM was running while VT-X is
enabled. It's fairly common [0].
So, I decided to patch it. This patch allows the system to successfully
suspend while VMs are runn
Hi Moritz and Matthew,
This is an issue with the current release of Debian Squeeze. The
issue is actually a boot-time load order issue. If the pcspkr module
is inserted into the kernel too early in the boot process (as it is by
default, at least on my system) the beep becomes the evil buzzing ho
Sébastien Delafond writes:
> So the bug is in 7.01g-1, then, and not in 7.4-2 ?
Apologies for the late reply, but you are right. This is a bug in 7.01
and not in 7.4. Doc-view-mode doesn't provide the function that 7.01
was trying to use, while image-mode does. That's the necessary
correction
Subject: org-mode: Cannot Store Org-Links from DocView Mode
Package: org-mode
Version: 7.4-2
Severity: normal
Tags: patch
*** Please type your report below this line ***
Trying to store a link to a PDF file with Squeeze's org-mode causes
org-mode to error instead of storing the link.
Sid's versi
Francois Marier writes:
> Hi Nick,
>
> Thanks for contributing some ponies. They'll be in the next cowsay upload
> (which should happen in a couple of minutes).
>
> I'm assuming that you are releasing your new "cows" under the same licensing
> terms as the rest of cowsay. If that's incorrect, let
Apparently PGP signed messages aren't well tolerated. I'm re-sending
this without signing it in hopes that it'll be readable on the bug's
page.
Attached are a pair of files that should make a good start for a pony
(one small and one large). They were created with Emacs's Artist mode.
Also attac
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
- --=-=-=
Content-Transfer-Encoding: quoted-printable
Attached are a pair of files that should make a good start for a pony
(one small and one large). They were created with Emacs's Artist mode.
The only major difference is that the small pony is a c
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
I was actually working off of old documentation without realizing that
console-setup exists at all. That probably takes better care of those
needs in a more extensible way better than my proposed patch, so feel
free to consider this bug closed.
I'll
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
- --=-=-=
Because I'm smart, I'll forget to do things like actually attach the
file I claim to be submitting.
However, I've remembered this time.
Enjoy,
Nick
- --=-=-=
Content-Type: application/octet-stream
Content-Disposition: attachment; filename
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Subject: console-data: Add a new keymap, lisp-emacs-us.kmap.gz, to
/usr/share/keymaps/i386/qwerty
Package: console-data
Version: 2:1.07-11
Severity: wishlist
Tags: patch
*** Please type your report below this line ***
In order to reduce Emacs-based
Kurt, thanks for looking into this. I'm able to make the bug
disappear only by removing starttls (so "emacs-jabber" is forced to
use "gnutls-cli" instead of "starttls").
If I could figure out how to configure emacs-jabber to use
gnutls-cli instead of starttls, I could work around the problem. As
To configure Emacs-Jabber to use "gnutls-cli" instead of
"starttls",
add the following to your ``~/.emacs``::
(setq starttls-gnutls-program "gnutls")
Evaluate the line and you can connect as normal, via
``M-x jabber-connect``.
Of course, this still doesn't solve any of the underlying issu
Package: libssl0.9.8
Version: 0.9.8g-15+lenny5
Severity: grave
Justification: renders package unusable
*** Please type your report below this line ***
Intro
=
I've been using Emacs and Emacs-Jabber for a while now, quite
flawlessly and wonderfully. However, since I've installed the
starttl
Subject: pida: FileManager doesn't respect user's gtkrc colors.
Package: pida
Version: 0.5.1-5
Severity: normal
*** Please type your report below this line ***
pida/services/filemanager/filemanager.py doesn't respect user's colors
as defined in ~/.gtkrc-2.0. This is odd, as it does a great job a
This patch /seems/ to fix the NoneType error without unpleasant side-effects.
It looks like they missed filtering an event before the gui was displayed.
Nick
---
/usr/lib/python2.5/site-packages/pida/services/optionsmanager/optionsmanager.py
2008-07-23 00:04:02.0 -0500
+++ o
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
The best-practice method for using Pida (on Lenny) at this point seems to be:
1. Install pida normally.
2. Install emacs-snapshot from:
http://emacs.orebokech.com/
3. update-alternatives --config emacs
Choosing 'emacs-snapshot'
Much easier than tampering with the source, thanks Anarcat.
To patch pida myself, I did the following steps. I think it works, as
I'm writing this email in Pida with Emacs.
0. Install pida as normal to satisfy all the run-time dependencies.
# apt-get install pida
1. Create a pida directory to hack on later.
$ mkdir ~/pida
$ cd ~/pida
2. Down
According to (what I understand of) Pida's Emacs Support page[0], the
necessary patches for Emacs have been adopted upstream (and have been
released as of Emacs 22.3), though this version hasn't yet percolated
into Debian.
You could always submit a bug to the Emacs team and ask them to
package 22.
me to
design it prettily.
This program requires the "wajig" package to be installed.
Thanks,
Nick Daly
apt-busted.tar.bz2
Description: Binary data
Thanks for your very informative and extremely quick response on this bug!
The network issue seems completely unrelated. It turns out it was caused by 2
still running copies of openarena which refused to release their ports. These
were finally killed with signal 9 (after refusing to cooperate
am.
Thanks so much for your time and help on this issue,
Nick Daly
-- System Information:
Debian Release: lenny/sid
APT prefers testing
APT policy: (500, 'testing')
Architecture: i386 (i686)
Kernel: Linux 2.6.26-1-686 (SMP w/1 CPU core)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF
Subject: sear-media: Sear-Media requires update for Sear > 0.6.1
Package: sear-media
Version: 0.6-20060204-1
Severity: important
*** Please type your report below this line ***
Sear requires an updated sear-media package, as sear currently reports:
[Error] System: Error opening script file: ${SEA
Subject: openrpg: OpenRPG 1.7.1 Released (2007-03-09)
Package: openrpg
Version: 1.6.3-1.1
Severity: wishlist
*** Please type your report below this line ***
OpenRPG 1.7.1 was released on March 9, 2007, and it'd be nice to have
the most updated version available in Debian.
This version also seems
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
From: Nick <[EMAIL PROTECTED]>
To: Debian Bug Tracking System <[EMAIL PROTECTED]>
Subject: miro: Miro Segmentation Fault - Gone?
Message-ID: <[EMAIL PROTECTED]>
X-Mailer: reportbug 3.44
Date: Thu, 07 Aug
I've also had this problem. I noticed however, that it only occurs when the
"Miro Guide" is loaded from the internet.
If you start Miro while offline and then click off the Guide to another channel
(your library for example), then go online, Miro doesn't segfault. Of course,
you then have to
I finally got around to not breaking the autohiding originally in Pypanel.
This version of the mouse-aware autohiding will hide CLOCK_DELAY seconds after
the mouse leaves the panel.
I'll be honest, a 3-second delay really _is_ much nicer than immediately. :)
This patch was already submitted upst
me, I may work on a further patch to add this feature, but that'll
be in the future.
Thank you for your time,
Nick Daly
-- System Information:
Debian Release: lenny/sid
APT prefers testing
APT policy: (500, 'testing')
Architecture: i386 (i686)
Kernel: Linux 2.6.22-3-686 (SMP
cted token
near" the package name... I don't understand how the Debian menu on my
system is created at this point... I only stumbled upon this problem because I
recently switched from Gnome to fluxbox, and then openbox when I found that the
fluxbox menu didn't work. Running
e at python and bash, but it
seems possible...
I do realize the main problem is my shoddy hardware, and pipe dream though it
may be, my dream is still to have a system that runs well despite, or at least
that can recognize, its failures. Your point however, is well taken.
Nick Daly
--- &quo
ble.
After each run, "apt-busted" would provide a list of packages and
files that didn't verify correctly, and then offer the user the option
to reinstall the mentioned, possibly broken packages/files.
I know this is a crazy pie-in-the-sky idea, but it would give a
beautiful verifia
So sorry for submitting a duplicate report, I realize how valuable a
developer's time is.
Didn't think of looking upstream. The obvious workaround is just to take the
URL it out
of the URL object and format it yourself (usually in angled brackets). Thank
you so much
for looking into this bug
o imply that lyx itself bombs out when reading the
file and passes the pages that were parsed correctly, before the page
with the footnote, to the previewer. The citations are "?" because the
bibliography is never processed (on the 8th page).
Thank you very much,
Nick Daly
-- System Inf
Package: xevil
Version: 2.02r2-4
Severity: grave
Justification: renders package unusable
*** Please type your report below this line ***
When run with no commandline options, after the license screen, regardless of
options selected, xevil always crashes with the following output:
*** glibc dete
59 matches
Mail list logo
