PC-BSD

afab4 · posted: 2012-07-16 17:27

Look at the new"killer feature" PC-BSD is developing for their next release:

========= SCRAPER REMOVED AN EMBEDDED LINK HERE ===========
url was:"http://distrowatch.com/weekly.php?issue=20120716#news"
linktext was:"http://distrowatch.com/weekly.php?issue=20120716#news"
====================================

Wait a minute....antiX-12 already has this feature. It's called antixsnapshot.

chrispop99 · posted: 2012-07-16 20:51

Much as it pains me to say it, MS has had pretty much the same thing for years - System Restore.

Chris

Dave · posted: 2012-07-16 21:28

chrispop99 wrote:Much as it pains me to say it, MS has had pretty much the same thing for years - System Restore.

Chris

Have you ever actually have this work? Every time I have tried System Restore there would be new, far greater problems than those that have been solved. Running it again causes more problems, and from there these"restores" make a snowball of problems...... end result reinstall!

afab4 · posted: 2012-07-16 23:10

Dave wrote:
chrispop99 wrote:Much as it pains me to say it, MS has had pretty much the same thing for years - System Restore.

Chris
Have you ever actually have this work? Every time I have tried System Restore there would be new, far greater problems than those that have been solved. Running it again causes more problems, and from there these"restores" make a snowball of problems...... end result reinstall!

"Back in the day' when I ran W98/2000/xp most of my problems were viruses, spyware and similar afflictions. Running a system restore in Windows (if it worked) just brought me back to square one trying to figure out what was wrong....
I truly appreciate the dedication of you open source (read antiX) developers who provide and perfect the code to build a great application like antixsnapshot.
If I muck something up I can reinstall my antiXsnapshot in less than 10 minutes and be back in business.

meandean · posted: 2012-07-16 23:34

I truly appreciate the dedication of you open source (read antiX) developers....

well actually __{{emoticon}}__

snapshot is a great tool though...

afab4 · posted: 2012-07-17 01:07

meandean wrote:
well actually __{{emoticon}}__

snapshot is a great tool though...

Oh yeah...I guess snapshot is your baby, Dean. Originally refracta. I figured it out by loolking at the script. __{{emoticon}}__
Well kudo's to you on your excellent work. AND thank you!

anticapitalista · posted: 2012-07-17 10:29

Code: Select all

# adapted for antiX by anticapitalista@operamail.com 
# from version="refractasnapshot-9.0.1 (20110906)" by fsmithred@gmail.com (August 2011)
# which was based primarily on refractasnapshot-8.0.4  by Dean Linkous with ideas
# borrowed from dzsnapshot-gui.sh by David Hare, which was based on an
# earlier version of this script.
# fsmithred@gmail.com (August 2011)
# License: GPL-3
# This is free software with NO WARRANTY. Use at your own risk!

So kudos indeed to meandean, David Hare and fsmithred for an excellent script.

Since antiX is not built using Debian live and is more like how Knoppix used to build their live iso, I had to make some changes to the script otherwise it wouldn't work in antiX.

Now if only there was a way to control CPU usage when creating the squashfs file.

meandean · posted: 2012-07-17 16:30

Heck I think the basic bash commands were the easy part. The others did the complicated part... __{{emoticon}}__

BitJam · posted: 2012-07-18 08:03

anticapitalista wrote:Now if only there was a way to control CPU usage when creating the squashfs file.

Have you tried"env nice -n 20 mksquashfs ..." ?

anticapitalista · posted: 2012-07-18 10:57

BitJam wrote:
anticapitalista wrote:Now if only there was a way to control CPU usage when creating the squashfs file.
Have you tried"env nice -n 20 mksquashfs ..." ?

Doesn't have any effect.

BitJam · posted: 2012-07-18 15:46

========= SCRAPER REMOVED AN EMBEDDED LINK HERE ===========
url was:"http://cpulimit.sourceforge.net/"
linktext was:"http://cpulimit.sourceforge.net/"
====================================

cpulimit is a simple program which attempts to limit the cpu usage of a process (expressed in percentage, not in cpu time). This is useful to control batch jobs, when you don't want them to eat too much cpu. It does not act on the nice value or other scheduling priority stuff, but on the real cpu usage. Also, it is able to adapt itself to the overall system load, dynamically and quickly.

The following works well here (on Gentoo):

Code: Select all

# burnK7 &
# cpulimit -e burnK7 -l 50
Process 9558 detected
[1]  + suspended (signal)  burnK7

I stopped it with ctrl-C then"fg" command then another ctrl-C. This may not be the optimal way to control it. The cpulimit program is very tiny. It should be run as root. The burnK7 is supposed to use 100% of one AMD cpu. There are other burn programs for Intel chips.

meandean · posted: 2012-07-18 16:00

I was going to mention cpulimit earlier but I am pretty sure fsmithred already told anticapitalista about it at the refracta forum.

Of course if you have more than one processor then mksquashfs has the -processors option so tell it how many processors to use.

Besides that, I can't think of much...

You could squash a piece, give it a break, squash another, and so on but that wouldn't help lower processor usage when it is actively squashing.

BitJam · posted: 2012-07-18 17:05

On second (third?) thought the"nice" program may be the ideal solution. Mksquashfs will still consume all available CPU if no other processes are trying to use the CPU. What happens is that mksquashfs will become very polite and give way to the other processes more easily so you will only notice the effect of nice when you actually try to do something else. The system should become more responsive.

I can't tell here because my system stays responsive even with all cores running burnK7 without nice.

anticapitalista · posted: 2012-07-18 17:07

Yes, cpulimit was mentioned before, but I forgot about it __{{emoticon}}__

This seems to be working.

#cpulimit --exe mksquashfs --limit 40

and then run mksquashfs

The main reason to reduce CPU is due to heat on laptops. I think this option is better than using a bag of ice cubes __{{emoticon}}__

chrispop99 · posted: 2012-07-18 17:38

Dave wrote:
Have you ever actually have this work

Many, many times.

I used to run Windows ME... __{{emoticon}}__

Chris