Package Details: lib32-networkmanager 1.2.4-1

Git Clone URL: (read-only)
Package Base: lib32-networkmanager
Description: Network Management daemon, 32bit libraries
Upstream URL:
Licenses: GPL2, LGPL2.1
Submitter: oi_wtf
Maintainer: oi_wtf
Last Packager: oi_wtf
Votes: 37
Popularity: 2.696550
First Submitted: 2013-05-31 21:26
Last Updated: 2016-08-06 13:13

Latest Comments

alucryd commented on 2016-08-19 20:48

Moved lib32-libnm-glib to [multilib] but not lib32-networkmanager, will keep this around for now.

Jeroen0494 commented on 2016-07-11 11:53

This is still a problem, any ideas?

ERROR: test-link-linux - too few tests run (expected 46, got 4)
ERROR: test-link-linux - exited with status 134 (terminated by signal 6?)


Jeroen0494 commented on 2016-05-28 13:30

The packages fails to build with the following error:

make[1]: Leaving directory '/home/jeroen/.cache/pacaur/lib32-networkmanager/src/NetworkManager-1.2.2/vapi'
Makefile:645: recipe for target 'check-recursive' failed
make: *** [check-recursive] Error 1
make: Target 'check' not remade because of errors.

This errors occurs after the 'Testsuite summary for NetworkManager 1.2.2' which passes with 11/11 tests succesfully run.

Any clue as to why this happens and how to fix it?

WoefulDerelict commented on 2016-05-19 21:41

IwfY: I suspect this is a flaw in lib32-bluez. The file is part of bluez-libs and is not included in lib32-bluez-libs. The dependency goes unsatisfied when the package is installed as it was undeclared and is not pulled in by the other bluez packages.

UPDATE: I've corrected the issue in lib32-bluez-libs. The headers should now be pulled in when a users installs lib32-bluez-libs. Please try building again after updating lib32-bluez-libs.

IwfY commented on 2016-05-19 07:23

When trying to build this I get the following error:

make[5]: Entering directory '/.../lib32-networkmanager/src/NetworkManager-1.2.2/src/devices/bluetooth'
CC nm-bluez-manager.lo
CC nm-bluez-device.lo
CC nm-bluez4-adapter.lo
CC nm-bluez4-manager.lo
CC nm-bluez5-manager.lo
CC nm-bt-error.lo
CC nm-device-bt.lo
CC nm-bluez5-dun.lo
nm-bluez5-dun.c:24:27: fatal error: bluetooth/sdp.h: No such file or directory
#include <bluetooth/sdp.h>
compilation terminated.

Any idea what could help?

oi_wtf commented on 2016-05-19 00:37

Yes, I intended to add bluez as an optional dependency like NM in [extra].
So the lib32-NM gets as close to NM in [extra] as possible.

lib32-bluez builds fine now, so I've added it now.

WoefulDerelict commented on 2016-05-18 20:18

I was simply pointing out that lib32-bluez-libs was a possible dependency for this package and that it will build the a 32-bit plugin for bluetooth if it is detected. I did not express whether this was desired or useful behaviour, only that it existed.

If it is a useful component then I'd suggest adding lib32-bluez-libs as a make depend and lib32-bluez as an optional dependency of the lib32-networkmanager package like the maintainers have done with the networkmanager package in extra.

The build issues with the lib32-bluez package set have been addressed.

felipec commented on 2016-05-17 23:52

Personally I have no interest in getting bluez-libs to build. In my opinion right now either it should be added as a builddep, or the configure should have --disable-bluez5-dun. As the PKGCONFIG stands now it's kind of broken.

oi_wtf commented on 2016-05-17 23:31

One may get a clean package from an "unclean" enviroment, but you can never *expect" it to be.

Anyway, I'd rather have bluez-libs as dependency, so the focus should be on getting that one to build.

felipec commented on 2016-05-17 23:13

Yes you can. Just do --disable-bluez5-dun, the it doesn't matter if the environment is "unclean" (there isn't such thing).

All comments