sputnik
-
Posts
122 -
Joined
-
Last visited
-
Days Won
9
Content Type
Profiles
Forums
Blogs
Posts posted by sputnik
-
-
Well, there's this: http://www.freedesktop.org/wiki/Software/ModemManager/Debugging/. Of course, instead of the systemd commands, replace with /etc/init.d/NetworkManager stop, etc.
Is it a service? It has an entry in /etc/init.d? Then it's easy, you can put it in rc-update, which is really just a wrapper for putting links in /etc/runlevels.
You could create a service file in /etc/init.d , it wouldn't be too difficult, just use existing files as a guide.
If not, then you can put it in your /etc/local.d/*.start file. Create one if you don't have one already, there is an explanation file in that directory. Something like this would work:
/usr/sbin/ModemManager /etc/init.d/NetworkManager restart
Or if there is a timing error with the startup of NetworkManager that way you can always do it with a startup file, say in your home directory. It'll have to be owned by root and you'll have to add it to /etc/sudoers by way of sudo visudo, an entry like
<username> ALL = NOPASSWD: /home/<username>/<filename>
. This file would be like:
#!/bin/bash killall networkmanager /usr/sbin/ModemManager /etc/init.d/NetworkManager start
Then start that file with sudo <filename> in your desktop startup stuff or .xinitrc if you use that.
Listed in order of preference. I don't have ModemManager, so I can only suggest general ways of handling it.
-
You can try:
eselect python set python2.7
prior to the emerge. Then remember to set it back to 3.3 when finished.
Else you might try python-updater, never hurts, often fixes obscure python-related problems.
Some ebuilds just don't automatically select the right version, I had that problem with compiz. Solved with this:
pkg_setup() { python_set_active_version 2 }And then you have to add python to the INHERIT line.
Just shots in the dark you understand, not saying I know the problem. Onnea
-
If it works don't fix it. Either the qt stuff we solved or the speex thing probably.
You're welcome, it was a fun challenge, I kinda like fighting unofficial ebuilds.
-
There are two subdirs in that /usr/lib64/binutils/x86_64-pc-linux-gnu/2.23.2 directory, include and ldscripts. Judging by their contents it looks like maybe libbfd uses them. I've checked and there's nothing with a conflicting name in /usr/lib. So you might try linking to those as well:
ln -s /usr/lib64/binutils/x86_64-pc-linux-gnu/2.23.2/include /usr/lib/include ln -s /usr/lib64/binutils/x86_64-pc-linux-gnu/2.23.2/ldscripts /usr/lib/ldscripts
VERY tacky, but you can remove those links once it gets installed. Then you just gotta make a note to yourself the steps involved if you ever want to reinstall/upgrade radium.
If that doesn't work, about out of ideas here. Best bet is try filing an issue there at Github, sounds like maybe you already did that.
-
I dl'ed the radium source you're using. Weak documentation. However, by deduction I believe it's looking for libbfd.so in /usr/lib. It's tacky, but you should be able to make a link like so:
ln -s /usr/lib64/binutils/x86_64-pc-linux-gnu/2.23.2/libbfd.so /usr/lib/libbfd.so (or /usr/lib64/libbfd.so if you prefer)
In the supplied README it talks about installing on Ubuntu 12. Check this out:http://packages.ubuntu.com/lucid/amd64/binutils-dev/filelist
So that's where Ubuntu's putting it. Worth a try, I'm pretty sure it will work.
A patch would be better, but grep -R libbfd didn't give a clue, don't think it specifically looks for libbfd.so, probably looking for all libraries to be in /usr/lib.
-
No, not a use flag.
Debian, Ubuntu, etc., they have the -devel versions because they are behind, usually in a leading edge distro like Funtoo we *HAVE* what they call the -devel version. In current x86_64 it's 2.23.2, one step back from the very latest, 2.24. So that's not it. Besides, it's looking for libbfd, which has been in binutils for a few years at least. It's just not finding it.
You should have it here: /usr/lib64/binutils/x86_64-pc-linux-gnu/2.23.2/libbfd.so
So you may have to get fancy. In the radium tarball, like most open-source packages, you'll find files like INSTALL and README in the root dir. You may have to look in there where it will tell you how to point the install to certain key directories, it'll be something like CXXFLAGS="/usr/lib" emerge radium (example, not for use). You can put this kind of thing in package.env and env, so 3 months from now when you've forgotten all about it it will happen automatically.
That's the best I've got for now, will take another look tomorrow. Hopefully somebody more knowledgeable can chime in.
Oh, BTW, it's looking in /usr/x86_64-pc-linux-gnu/bin/ld, which is a pointer link (here) to: /usr/x86_64-pc-linux-gnu/binutils-bin/2.23.2/ld. There my comprehension of the issue gets hazy, gotta brush up on that ld stuff someday.
-
It appears it can't find speex.h:
externals/unauthorized/speexout~.c:69:56: fatal error: speex/speex.h: No such file or directory #include <speex/speex.h> /* speex codec stuff */ ^ compilation terminated.
I have that header file here in /usr/include/speex, it was installed by media-libs/speex. Perhaps you need to install that?
-
-
That's the kind of info needed you see.
I would try this first. If you don't have FEATURES="buildpkg..." in your /etc/portage/make.conf file, use quickpkg to build binaries of qtgui and qtcore, so that you can quickly & easily reinstall them if need be.
Then emerge --unmerge qtgui qtcore. Remove them.
Then emerge radium. They should get reinstalled just fine. Else nothing lost but a few minutes effort. emerge -K qtgui qtcore and everything's back the way it was if it fails.
Edit: you can probably do emerge -avK radium, that way just the binaries of those two will quickly install.
2nd edit: Whoops. Meant emerge -avk radium
-
The problem is we could be "what ifing" all day. Better to just do itTM and post any error messages.
Probably you've already checked out the variables http://devmanual.gentoo.org/ebuild-writing/variables/
This line
SRC_URI="http://archive.notam02.no/arkiv/src/${P}.tar.gz"is the pointer to the version, using the name of the ebuild to define ${P}. So if you make an ebuild in your local repository named after the version you want, that's the one it will go fetch.
After creating this custom ebuild in your local repository any patch files will also need to be in that directory. Often you can get away with just creating a link to the patches "file" dir in the official Funtoo repository, if not just copy the files dir there.
Do ebuild <name & version>.ebuild manifest and give it a shot.
-
I also tried to live with systemd. I was using archlinuxarm on my embedded device, they had no choice but to follow suit with Big Daddy arch when they switched (hmm, common, eh?).
Generic run-of-the-mill daemon starting is easy enough, just like rc. For me, the problem was the "in between the cracks" stuff. Oh, Lennart Poettering is quick (and vocal) to point out the MOUNTAINS of documentation about his baby. Also to finger people who don't care for it for whatever reason "Haters". I'm all for spending time learning about something new if it has benefit to me, but this didn't. And mountains of documentation, all well and fine, but wading through that to do something I could do in seconds in rc didn't seem like the direction I wanted to go in learning.
Specifically, I had an issue using dtach. On a limited memory device, dtach is much better than screen if you run it 24/7, since it is MUCH smaller. It involved sockets. At the time there was very little to go on, although I have seen a lot more now. I spent a couple of days trying to work it out, finally decided to switch to Funtoo as I was quite familiar with it running on my "real" computers.
By default it also ran the systemd log, journald. On my embedded device there is a total of 128M ram, it sucked up 70M or so. Even when I figured out how to defeat that, there was still some memory footprint of it, I never got the full 70M back.
It keeps growing and growing in scope, 1st you don't need a system log anymore, journald replaces it (poorly IMO). Then cron, it has it's own implimentation, login? Nope, only systemd login needed now. What's next? This isn't the GNU/linux way to me.
Finally, there are a lot of political implications, I won't go into that here, it's easy enough to find. I do find it quite interesting that the U.S. military is Red Hat's biggest customer.
I'm no expert on systemd, it sounds like you've put some effort into learning about it, this is just my personal experience with trying to use it for 8-9 months.
Hah, here's a quote on wikipedia/systemd (http://en.wikipedia.org/wiki/Systemd):
In an August 2014 article published in InfoWorld, Paul Venezia writes about the systemd controversy, and attributes the controversy to violation of the Unix philosophy, and to "enormous egos who firmly believe they can do no wrong."[39] The article also characterizes the architecture of systemd as more similar to that of Microsoft Windows software:[39] While systemd has succeeded in its original goals, it's not stopping there. systemd is becoming the Svchost of Linux ? which I don't think most Linux folks want. You see, systemd is growing, like wildfire, well outside the bounds of enhancing the Linux boot experience. systemd wants to control most, if not all, of the fundamental functional aspects of a Linux system ? from authentication to mounting shares to network configuration to syslog to cron.
I take issue with the "succeeded in its original goals" statement.Haters Central: http://boycottsystemd.org/ -
Apologies, low on sleep here. Glad you got it.
-
Hello and welcome.
It's probably not your fault, you lucked out and installed at the same time perl was being updated to a new version. It can be kind of tricky even for established installs to cope with that.
Yes, the fact that perl-cleaner didn't finish sounds like a good bet. Hopefully someone more knowledgeable will come to the rescue here, but until then, I wonder if perl-cleaner complains about blockers? That's often the case. If so, make a note of the blockers and uninstall them, then it should finish and you should get over the hump. Then you can reinstall them ( I think you should, but I never do), do emerge --depclean && revdep-rebuild. You often have to pull out perl stuff to get it to work.
Otherwise, perhaps post your results from perl-cleaner to give us more to go on.
-
Good eyes.
Hmm, arch & debian already fixed...nothing new here yet:
# eix app-shells/bashapp-shells/bashAvailable versions: 4.2_p10 (~)4.2_p10-r1 (~)4.2_p20 (~)4.2_p24 4.2_p37 4.2_p45 4.2_p45-r1{tbz2} (~)4.2_p47{tbz2} -
This phoenix has risen from the ashes about 3-4 times. Nothing a little shake n bake in the oven can't fix (BGA video chip, last laptop I'll even own). Actually, I use a heat gun.
-
Like everyone else, I know my desktop is the most beautiful on the planet: http://imgur.com/9baiBBn
Compiz | Lxpanel | SpaceFM | Docky - a no-desktop desktop
-
Latest crossdev command line I have used here is this:
crossdev --b 2.23.1 --g 4.8.1-r1 --k 3.13 --l 2.13-r5 -t armv5tel-softfloat-linux-gnueabi -P -v
Binutils and glibc are at that version in the Funtoo tree for my armv5 device, perhaps different for you.
You'll have to change the tuple for your device, probably at the very least you'll need to change the 1st and 2nd portions - yours is a hardfloat I think.
I prefer to add the -P -v so that portage will show me what it's up to, rather than waiting until the very end. Crossdev isn't very verbose otherwise.
To use the gentoo gcc ebuilds, download one with a different minor version from the link I gave above, try to pick one that isn't in the Funtoo tree. Then set it up in your local repository (usually /usr/local/portage). Run the above crossdev command (revised for your device) for a few seconds, then ctl-c, cancel it. Go to where crossdev has set up the portage links (again, usually /usr/local/portage), look for the cross-arm*-*float-linux-gnueabi directory. In there it simply links to the normal portage tree ebuilds for binutils, gcc, etc. Check to make sure the sys-devel/gcc link points to your local gentoo gcc ebuild, if not, delete the link and replace it pointing to the gentoo ebuild in your local repository.
/usr/local/portage/cross-armv5tel-softfloat-linux-gnueabi $ ls -l total 0 lrwxrwxrwx 1 root root 47 Jan 29 2014 binutils -> /usr/portage/sys-devel/binutils lrwxrwxrwx 1 root root 32 Aug 25 16:26 gcc -> /usr/local/portage/sys-devel/gcc lrwxrwxrwx 1 root root 42 Jan 29 2014 gdb -> /usr/portage/sys-devel/gdb lrwxrwxrwx 1 root root 43 Jan 29 2014 glibc -> /usr/portage/sys-libs/glibc lrwxrwxrwx 1 root root 53 Jan 29 2014 linux-headers -> /usr/portage/sys-kernel/linux-headers
Now run the crossdev command again without stopping it this time. That's it.
-
There was some kind of funny stuff with the Gentoo afternoon update, I had a weird message about downgrading libreoffice for no good reason. Happens. Good to say "oh, well, I'll update tomorrow" when you see stuff like that. Maybe there's something else to the xfce stuff, I just don't know, but considering the other funnyness I'd wait and see. I re-synced later and the libreoffice downgrade was gone, although the xfce stuff you mention is still missing.
Also, as a BTW, I used to have a lot of trouble with eix-sync, lunched my Portdir 2 or 3 times, so I quit using it. In Funtoo, when I was using Gentoo it was no problem. Just FYI, in case you have trouble. Could have just been something here locally. Now I emerge --sync && eix-update, same thing, a little more typing.
-
Sorry for the late posting, but I wanted to correct something for posterity.
The link I gave for archlinuxarm kernels, while correct if you want to build them, is not the best. You can always get the latest here:
http://os.archlinuxarm.org/arm/core
The rollback machine is handy also for older kernels:
http://rollback.archlinuxarm.org/packages/l/
@kindofblue (if you haven't found your answer yet)1st of all, you cannot choose the same minor version of gcc that you are running on the host machine, you will always get the collision errors. You have to step down to 4.8.1. It's close enough for distcc.Secondly, as per the Funtoo devs, crossdev does not work in Funtoo. However, it works great for me as long as I use gentoo gcc ebuilds. You can find them here: http://sources.gentoo.org/cgi-bin/viewvc.cgi/gentoo-x86/?hidI use them only for cross-compile environments, I only use official Funtoo ebuilds for everything else.
-
I don't use luks, but I'm guessing it's the fstab it's having trouble with, have you considered using UUID's or labels? Most recommend that anyway.
I do use LVM but in fstab I use labels.
-
Rather than trying to decipher the meaning of some cryptic kernel panic message (good luck with that!) I would suggest starting with a major known working kernel like debian, and then modifying from there. Of course, doing a bunch of reading at places like pappy's kernel seeds is good too.
That way you have something that works right away, and when you are done modifying, it's still completely YOUR own custom kernel. Make small changes, compile, boot. If it doesn't boot you know what small changes you just made, easy to fix.
You can just emerge debian-sources, zcat /proc/config.gz > /<somewhere>/.config. Now you have the exact config they are using and can start from there.
Good to make a listing of lspci output too, then you know exactly what devices are on your box, you can start cutting out the extra junk you don't need. Like if you're running Intel, you can get rid of anything that starts with AMD. Etc.
Although I make my kernels very lean, I don't really believe it makes much difference having a bunch of unused drivers, other than increasing the size somewhat, speed, I think it makes little to no difference these days.
-
I don't have a beaglebone black, but this doc should cover your questions: http://www.funtoo.org/Funtoo_Linux_Installation_on_ARM
Regarding the kernel config, you might consider seeing how archarmlinux does theirs, I've stopped bothering compiling my arm kernels and just grab theirs, it actually works better for my device than the ones I do :wacko: .
https://github.com/archlinuxarm/PKGBUILDs/tree/master/core
I don't see the beaglebone there, but I know they support it, maybe some other Cortex-A8? I know one of the devs has one, have seen him talk about it on IRC.
I also don't see a Funtoo stage3 seed for your device so you may have the opportunity to learn about making one with Metro and posting it for those that follow you. B)
I hope I have helped you with all these I dunno's.
-
Perhaps Devices>show>settings?
This lxde-specific question might be better answered at the lxde forum, they are really good there: http://forum.lxde.org/
I am a big fan of the author's work. He has really stopped development on pcmanfm and concentrates on spacefm, which I use here for a desktop manager. It is the same thing only updated, more powerful. He also maintains a good question/answer board: http://ignorantguru.github.io/
Ohhoh, he is now taking an indefinite vacation it says there, so everything's unmaintained. We'll have to see how that goes.
-
I highly recommend this Gentoo-specific instructional page at the VideoLan.org:
http://www.videolan.org/vlc/download-gentoo.html
It's quite in depth about the USE flags needed, not only for vlc, but other support packages.
Werks fer mi
No pulseaudio
in Installation Help
Posted
um, there is no libpulse?