bug#32170: python-cairocffi error "dlopen() failed to load a library: cairo / cairo-2"

2018-07-16 Thread Ben Sturmfels
On 17/07/18 07:54, Danny Milosavljevic wrote: > Hi Ben, > > thanks for the report. > > I've fixed it in a545090ce5e6b9da1f473f558dabda69f317e461 on guix master > by allowing absolute paths as python-cairocffi's dlopen's parameter > and also substituting the sonames by the absolute paths in the ca

bug#32180: French manual installed unexpectedly

2018-07-16 Thread Chris Marusich
Hi Guix, I recently upgraded two different GuixSD systems by running "guix pull" to get the latest and greatest. In both cases, when I invoke "info guix" after the upgrade, it brings me to the French manual: GNU Guix Cette documentation décrit GNU Guix version 0.14.0.6214-4

bug#32167: Kernel 'build' directory in the store is a broken symbolic link

2018-07-16 Thread pkill9
Yes I agree with you, since it is a lot of space then it's probably best to just delete the symlink. The reasoning behind my suggestion of keeping it is mostly for convenience in compiling/testing an external kernel module, i.e. just downloading the source and then compiling it with the current

bug#32167: Kernel 'build' directory in the store is a broken symbolic link

2018-07-16 Thread Mark H Weaver
Danny Milosavljevic writes: > On Mon, 16 Jul 2018 18:55:11 +0100 (BST) > wrote: > >> It would be good to keep the build directory though, since it's >> expected to exist, and it's easier to just download a module's >> source and compile it and test it. > > I agree. > > /run/booted-system/kernel/

bug#32170: python-cairocffi error "dlopen() failed to load a library: cairo / cairo-2"

2018-07-16 Thread Danny Milosavljevic
Hi Ben, thanks for the report. I've fixed it in a545090ce5e6b9da1f473f558dabda69f317e461 on guix master by allowing absolute paths as python-cairocffi's dlopen's parameter and also substituting the sonames by the absolute paths in the callers. python-cairocffi's dlopen is re-exported. That mean

bug#32167: Kernel 'build' directory in the store is a broken symbolic link

2018-07-16 Thread Danny Milosavljevic
On Mon, 16 Jul 2018 18:55:11 +0100 (BST) wrote: > It would be good to keep the build directory though, since it's expected to > exist, and it's easier to just download a module's source and compile it and > test it. I agree. /run/booted-system/kernel/lib/modules/4.17.3-gnu is in the store any

bug#32167: Kernel 'build' directory in the store is a broken symbolic link

2018-07-16 Thread pkill9
It would be good to keep the build directory though, since it's expected to exist, and it's easier to just download a module's source and compile it and test it.

bug#31831: CVE-2018-0495 Key Extraction Side Channel in Multiple Crypto Libraries

2018-07-16 Thread Leo Famulari
On Mon, Jul 16, 2018 at 01:14:30PM -0400, Leo Famulari wrote: > libtomcrypt version 1.18.2 includes a fix; we would need to adapt this > to the bundled copy in Dropbear. I can take a look at this today. Dropbear's bundled libtomcrypt includes a variety of whitespace and comment changes that make i

bug#31831: CVE-2018-0495 Key Extraction Side Channel in Multiple Crypto Libraries

2018-07-16 Thread Leo Famulari
On Mon, Jul 16, 2018 at 08:53:56AM +0200, Gábor Boskovits wrote: > Are there any more packages needing attention? libtomcrypt version 1.18.2 includes a fix; we would need to adapt this to the bundled copy in Dropbear. I can take a look at this today. NSS was fixed in Guix commit 7c3bea7e6299e1026

bug#32170: python-cairocffi error "dlopen() failed to load a library: cairo / cairo-2"

2018-07-16 Thread Ben Sturmfels
Hi Folks, I'm having problems running Python code that depends on the "python-cairocffi" package. Running "from cairocffi._ffi import ffi" on Trisquel 8 works fine, but on GuixSD I get: $ guix environment --container --ad-hoc python cairo python-cairocffi $ python3.6 -c "from cairocffi._ffi impor