Thanks Bastien. I'm not currently using this package or set up to upload.
On Mon, Oct 2, 2023, 1:15 PM Bastian Germann wrote:
> I am uploading a NMU to DELAYED/10 in order to fix this.
> Please find the debdiff attached.
No need to delay. Total possible impact one way or another would be really
low, and they can rebuild if they need to. I haven't used it in a decade,
and am surprised my deadline hacks have survived all this time. I don't
currently have a known working build system.
-Dtake
On Wed, Nov 2,
I have a similar bug on a laptop with both Intel UHD graphics and an NVidia
GPU. Without Nvidia loaded, behavior on unlock is as expected. With
nvidia modules loaded, even though not used for display, something is
disabling the display _after_ I unlock the screen (screen comes up from
sleep and
I've had this for a while too and been looking. Not all versions of clang
have -dwarf-column-info, and the clang version just changed. What worked
for me was just deleting it from intel-opencl-clang.
intel-grahics-compiler validates the flag and passes the errors back up,
but isn't where the flag
d...@debian.org should be forwarding to my drake.diedr...@gmail.com
account. I've uploaded empire-hub 1.0.2.2 several times since December,
with a Closes: 809625, but it still hasn't closed.
The two most recent upload attempts from .upload:
u empire-hub_1.0.2.2.tar.gz ftp.upload.debian.org
The bug is the duma script setting the wrong path to the installed duma
library (it's good enough to just make it unqualified and not specify ./
for current directory).
duma is also declared as a bash script, while bash isn't a package
requirement. The DYLD environment variables appear to be a
On Sun, Sep 26, 2010 at 2:17 PM, Joey Hess jo...@debian.org wrote:
Drake, I pulled this patch into a branch of my debmirror git repository,
and looked it over.
It seems to disable cleanup of obsolete files:
-cleanup_unknown_files() if ($post_cleanup);
+cleanup_unknown_files() if
On Mon, Sep 27, 2010 at 11:25 AM, Joey Hess jo...@debian.org wrote:
Ok. I don't really mind applying the patch to debmirror without cleaning
support, but there needs to at least be documentation/links on the man
page about it.
The patch looks ok to me, BTW. I would rather all the if
Brought forward onto debmirror 2.4.5.
-Drake d...@debian.org
diff --git a/debmirror b/debmirror
index 089cdac..ecf7dbf 100755
--- a/debmirror
+++ b/debmirror
@@ -69,7 +69,7 @@ Usage: $0 [--progress] [--verbose] [--debug] [--dry-run] [--help]
[--postcleanup|--cleanup|--nocleanup]
Package: debmirror
Version: 1:2.2.1
Severity: wishlist
Tags: patch
This patch implements full snapshots of each pull using hardlinks, and
a symlink to point to the most recent pull. Most of what this does is
making sure each new file is hardlinked to a similar location in a
numbered
We had an issue last week where existing kernel packages were
installing, but quietly failing to run update-grub. It turned out to
be that the build environment for the kernels was an older
distribution and the resulting kernel packages had a test -x $grubpath
in the generated postinst. The new
I was just bitten by this bug as well, patched it, then found this bug
report. For reference, this is the patch I'm using locally, using the
documented instruction in wait(2) to only call WEXITSTATUS() if
WIFEXITED() is true. Please count this as another request to put a
warning in the manpage
Turns out the latest upstream has a similar patch applied in 1.18.
return (pid == child_pid ? WEXITSTATUS(status) | WTERMSIG(status) : -1);
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
I'm another one who got bit (an ldap server with a - in the name).
- is a valid character in a hostname, but % and / are not. Here's a
patch:
--- debian/libnss-ldap.postinst.orig2006-07-11 13:41:04.481500312 -0700
+++ debian/libnss-ldap.postinst 2006-07-11 13:41:30.918481280 -0700
@@
I was about to file a similar wishlist and found that
there was already a well-evolved set of scripts in this one.
Rather than rely on partition types (luks is a partition type?), I had
my init-premount script use cryptsetup and an encrypted= cmdline, as
follows:
export ENCRYPTED=
for x in $(cat
Package: qt3-dev-tools
Version: 3:3.3.4-3
Severity: normal
Tags: patch
AMD64 began doing away with .../lib64/ directories a few weeks ago, and
this has broken qt3-dev-tools for one of my users. All that needs fixing is
the pathnames in /usr/share/qt3/mkspecs/linux-g++-64/qmake.conf.
---
16 matches
Mail list logo
