Arch Linux

this may be using a sledgehammer to crack a walnut but try deleting ~/.cache/dmenu_run

There's probably a nicer way to have it regenerated though :-) (and it's just a guess)

have you looked in any of the logs to see if there are any hints?

the pacnew file is the default vanilla one that comes with the package.  Typically, when I see one I do this to view the differences:

$ sdiff -s <original file> <file>.pacnew

There's no right answer to say "always do this" or "always do that" you've got to take it on a case-by-case basis.  Sometimes you may have modified the original file (like in the case of /etc/shadow) - when that happens, you should discard the pacnew one.

Sometimes (often with pacman.conf) there are new features that you need to merge in to your file (like the package signing stuff a while back)

If you're sure the original is totally vanilla you can copy the pacnew version over the top

For your specific situation, delete shadow.pacnew.  If you've already overwritten /etc/shadow, look for a backup (it might be called shadow-)

also, typically cron (not 100% sure about cronie) won't source your .bash* files.  An easy way to verify some stuff is to redirect all output from the cronjob

* * * * * /path/to/my/script > /tmp/script.log 2>&1

added TOC link to article and modified the TOC

nice guide... hope you don't mind, but I wifi-fied it at http://wiki.archbang.org/index.php?titl … _on_a_RAID

Totally agree on lack of need to mirror swap BTW

my very basic understanding is that your device driver (from the kernel) plumbs the interface - lets say eth0.

You then have tools to manipulate that such as dhcpcd/dhclient or ip/ifconfig

But no-one wants to run something like 'ifconfig eth0 inet 192.168.0.2 netmask 255.255.255.0 -trailers up' every time they log in so there are tools (netctl/networkmanager/netcfg) to make it all easier for you

You can edit /etc/systemd/journald.conf to forward all messages to syslog.

It adds overhead but it's anecdotally negligible and the benefit is you have your files in /var/log

what about regular 'xprop' (start if from the command line and then point the + at the qbitorrent window)

interesting... good info to know

what about 

${if_match ${exec laptop-detect}!=0}${battery_percent}${else}Power plugged${endif}

the -v option will specifically print the output too

$ laptop-detect -v
We're a laptop (ACPI batteries found)

$ echo $?
0

this definitely would have saved me a few times in the past.  Now I'm kind of paranoid and use GNU screen whenever I can (specifically for the logging and sharing capabilities too)

interesting app...  have you experienced any issues with screen re-drawing when grabbing an existing pty?

As long as it's straight in your head you don't need to change the names... but if it fails again, you can run:

systemctl status <name>

and it will tell you why it failed

Typically, you'll see an exit code (like I mentioned a few posts ago) that will help you pin down why it failed.

You can see a list here:
http://www.faqs.org/docs/abs/HTML/exitcodes.html

The echo part really needed the ? at the end :-) 

For example (the 0 means a match is found; the 1 means no match was found):

# grep root /etc/passwd; echo $?
root:x:0:0:root:/root:/bin/bash
0

# grep sdfdsfdsfdsfsd /etc/passwd; echo $?
1

Bringing this back to your problem, the exit code would tell you whether the rc.local script is executing properly or not.  Assuming it is, then Mr Green is probably right and you need to specify a shell at the top

do you have anything in journalctl?  Maybe your copy of agetty is messed up somewhere - there should be some kind of notification that it is trying to spawn

exit code 127 typically means 'file not found' indicating a typo or PATH problem.  Is your NM config completely vanilla?

I may be *way off* here but your volumeicon config is controlling alsa and it looks like you have pulseaudio running.  I *think* you either need to get rid of PA or use a PA volume control icon

The latest libgsf is 1.14.26-1 and the repos don't keep older versions so it is an update issue.  Your 'extra' repo is out of date - maybe the mirror is not synced up

I love the name... and thats half the battle for me.  I know it's totally irrational, but there are distros I wouldn't even try because I don't like the name.

Did you compile chmsee from source? 

Can you run:

bash$ ldd $(which chmsee)

The error message is saying it can't find /home/charles/libxpcom.so... quite why it's looking in your home dir is unclear to me unless you specified that in your compile-time options.

FWIW, chmsee is in the community repo... you can probably save yourself a lot a headaches with:

$ sudo pacman -S chmsee

Try setting things how you like and running alsactl

$ sudo alsactl store

Also, take a look at /etc/conf.d/alsa - there's an option for saving volume levels when stopping. 

I remember the live CD used to set the volume levels on bootup - not sure if that got pushed to the real install or not.  I'll take a closer look when I'm at a real laptop if the 'alsactl store' fails.

very nice...  I'm very partial to blue themes (no innuendo implied - especially on a work computer)

I'm going off on a bit of a tangent here but what does frustrate me is being attacked for things that were rectified *years* ago.  Virtually every thread on the arch forum that mentions archbang says a variant of "they use --force" by default with pacman."   This was true for *one* release a long time ago.  It may even have been the first thing that Mr Green fixed -  I don't remember.

My dilemma is whether we should try and correct that misinformation.  On one hand I think 'yes, people should be corrected for repeating bad information' but on the other hand I think 'they're just towing the party line' so why bother.

Right now I consider 'arch' to be the boss that you don't really connect with but you respect because you're a professional :-)  As long as they let me do my work I'm not going to get overly  involved or upset.

It seems that ubuntu wants to be it's own entity that happens to use a linux kernel... like webos or android.

My general rule of thumb is (so caveat emptor, your mileage may vary, etc etc) if I've never modified the original file then I stick the new file in it's place... but only after making a backup!

If I have modified the original, I try to recreate those changes in the new file

I ran 'pacman -Qo /etc/rc_maps.cfg' and it's owned by v4l-utils which is *not* essential for booting so it's nothing you can't recover from (but make it easy on yourself and back it up :-) )

10 print "aquarius rules"
20 goto 10
run

Good times... until you got the annoying "syntax error line xxx"

I guess only the opensource drivers have the xf86-video prefix so it's the expected output.  pacman -Qq is just going to return the package name so if the package name is nvidia then you won't see it in your output :-)

Not that that helps though - maybe look in the arch forums for lockups with nvidia

I think bridge linux also has an xfce4 version

you are prolific!

do you get an error or the message "there is nothing to do"?

qt4 is the new name for qt because qt5 is now in the repo.  You can safely say to replace qt with qt4 but anything from AUR that relies on qt[4] will have to be rebuilt

getting close to 500,000 signatures now

ArchBang should be required to have 15 pieces of flair

(sorry, can't resist)

It's definitely a tough ask to get rid of gtk[23] since tint2 relies on it (gtk2) as well as a number of useful apps (lxappearance, parcellite etc)

i see you're running jack... does it work as root?  Are you in the audio group?

patch -R (I think)

probably easier to install via ABS and not apply it to begin with though