site  contact  subhomenews

Alpha Litebook laptop

July 19, 2017 — BarryK
Puppy Forum member muggins (David) purchased one of these laptops:
http://news.softpedia.com/news/litebook-launches-cheap-chromebook-like-linux-laptop-powered-by-elementary-os-513578.shtml

However, he decided that he didn't want it, and offered it to me. I offered to pay him something, but he didn't want any money. That is extremely generous.

The reason behind the offer, is I posted awhile back that I wanted to replace my baby laptop with Cherry Trail CPU, as it has problems running Linux.

The Alpha Litebook has a Intel Celeron N3150 CPU, which is 14nm technology, 6W TDP and 1.6GHz (2.1GHz burst). This is a more Linux-froendly CPU.

It arrived this afternoon, and I eagerly fired it up. It comes with Linux, however, I am keen to run Quirky Linux on it.
No problem with configuring the UEFI-setup to boot from a Quirky Xerus 8.2 USB stick, but...

The kernel loaded, then just got a blank screen with a blinking cursor at lop-left. Rebooted with "loglevel=7" so I can see what is happening, and saw messages like this:

ata3: Link is slow to respond, please be patient (ready=0)
ata3: SRST failed (errno=-16)
ata3: Reset failed


Anyway, after 2 minutes and 20 seconds, spewing out the above messages repeatedly, it booted and I got a desktop. Well, after bootup, it all looks good, sound, drives all working.

I examined the GRUB config, can't see any special kernel boot parameters. So, the next thing to do is compare the kernel configuration of the Linux that the laptop came with. It is kernel 4.4.66.

Quite a nice laptop, hi-res screen, good specs for such a low price.

Read more...

Quirky Xerus64 8.2 coming soon

July 19, 2017 — BarryK
I thought that I had built 8.2 final this afternoon, but it failed one test. This is what can happen when one introduces changes at the 12th hour.

The problem is the introduction of the ntfs-3g driver into the initramfs, as I reported yesterday:
http://bkhome.org/news/201707/ntfs-3g-driver-in-initramfs.html

I had compiled this driver statically sometime ago, I thought that it was configured to use its internal fuse driver. The ntfs-3g driver can be configured to have its own internal fuse driver, or use the fuse driver of the Linux kernel.

Testing "8.2", the ntfs-3g driver reported fuse to be missing. There is no setup to have kernel modules in the initramfs, all drivers required for bootup are expected to be builtin to the kernel.

Instead of recompiling the ntfs-3g driver, I decided to recompile the kernel. Have bumped to the latest, 4.11.11, and just made one change, fuse driver now builtin. It is compiling right now.

So, hopefully, 8.2 will be out about another 24 hours from now. Will of course have to do more sanity testing before releasing.

Tags: quirky, linux

YASSM 3.0

July 18, 2017 — BarryK
As well as the "peasy" series, rcrsn51 also created Yet Another Share Mounter (YASSM) for Samba. Now up to version 3.0:
http://murga-linux.com/puppy/viewtopic.php?t=63090

Quirky has version 2.9, have updated to 3.0.

Tags: quirky, linux

ntfs-3g driver in initramfs

July 17, 2017 — BarryK
The guys on the Puppy Forum are testing Quirky 8.1.94, most recently an issue with ntfs:
http://murga-linux.com/puppy/viewtopic.php?p=961004#961004

There is something strange going on here. The Linux kernel has the ntfs driver builtin, and configured to allow writes when ntfs partitions are mounted.

Except that we can't. Get an error when try to write. It used to work, something has changed.

Anyway, writing to an ntfs filesystem is supposed to be less reliable using the kernel driver, and ntfs-3g fuse-based driver is preferred.

I didn't have ntfs-3g in the initramfs as compiled statically it is about 410KB. Have now put it in and modified the 'init' script to use it.

This is a good thing, as writing to ntfs filesystems will now be more reliable. Though, it may be possible to avoid writing from the initramfs, and remount with ntfs-3g when running in the main Quirky f.s.

The downside of ntfs-3g is that being fuse-based, it is slow.

Tags: quirky, linux

JWMDesk 2.3

July 17, 2017 — BarryK
Forum member radky has created JWMDesk, a very comprehensive GUI desktop manager, for configuring JWM and many other things:
http://www.murga-linux.com/puppy/viewtopic.php?t=101460

It overlaps configuration tools already in Quirky, however has the advantage of being an integrated manager.

So, I have decided to give it a go in Quirky. The latest version is 2.3, and I have created a PET named 'JWMDesk-2.3-1.pet'. The "-1" is because I have made a couple of changes -- I removed the 'pinstall.sh' file, and created /usr/local/jwmdesk/window_width with "510" in it.

The problem with pinstall.sh is that it has absolute paths in it, so not suitable for woof* builder. Also, I scanned through it, and I couldn't see anything that is really required. Admittedly, my look was brief, but a quick check (also brief), it seems to work without having run pinstall.sh.

JWMDesk will be in the next Quirky.

Read more...

UrxvtControl 1.3

July 17, 2017 — BarryK
Ha ha, I was lamenting the proliferation of GUI popup message box utilities in Quirky:
http://bkhome.org/news/201707/bcm-popup-messages.html

I have now removed one of them: 'yad'.

It is only required by 'urxvtset', a urxvt terminal emulator configuration utility:
http://www.murga-linux.com/puppy/viewtopic.php?t=68624

However, I have removed urxvtset, replaced it with radky's UrxvtControl 1.3:
http://murga-linux.com/puppy/viewtopic.php?t=97337

Tags: quirky, linux

0setup script fails in Woof

July 09, 2017 — BarryK
The '0setup' script in Woof* downloads package database files and converts them to Puppy-standard-format.

Sometime ago, yes I have a date, 160411, I applied a fix to the 0setup script in woofQ. UTF-8 multi-byte characters were breaking the script for Ubuntu-based builds.

These characters only occur in the description fields of the Ubuntu database files.

Back then, I inserted this code into 0setup, around line 200:
      #[ $RETSTAT -eq 0 ] && mv -f $xDLFILE ${PKGLISTFILE}pre

#160411 filter out utf-8 chars...
if [ $RETSTAT -eq 0 ];then
iconv -c -f utf-8 -t ascii $xDLFILE > ${PKGLISTFILE}pre
rm -f $xDLFILE
fi
<

Today, the script was again failing. But it wasn't obvious that it failed. The Ubuntu "updates" DB files were not getting completely converted to Puppy-format, it was stopping partway through.

I wouldn't have known, except that discovered a DEB that I knew to be in the updates db, wasn't in the build, only the original version.

It took me hours, staring at the screen and wondering what was wrong. Eventually, the penny dropped, the above code also has to be inserted further down, around line 398.
That's where the update database files are being downloaded.

Well, those line numbers are in woofQ, but I just now looked in woof-CE and the line number for insertion will be 272 ...oh, it has become very different from my script, that is the only place that I can see to insert.

Note, the failure occurs at the assignment of "MANIPULATED1".

Woof-CE page:
https://github.com/puppylinux-woof-CE/woof-CE/blob/testing/woof-code/0setup

Note also, another way to do the conversion:
# perl -i.bk -pe 's/[^[:ascii:]]//g;' filename
where filename.bk will become the original file.

Tags: linux

Intermittent hang at bootup

July 07, 2017 — BarryK
There is a very old bug, the PC hanging at bootup. Puppy Forum member ETP reported it. I looked at the code where the hang occurs. It only happens occasionally, well rarely, for me, so I have ignored it. But, ETP is getting it more frequently. It seems to be hardware related, and I have implemented what is probably a fix.

My response to ETP is here:
http://murga-linux.com/puppy/viewtopic.php?p=959974#959974

hdparm man page:
https://linux.die.net/man/8/hdparm

I see that woof-CE still has that line with "hdparm" (line 761), so maybe they should consider removing it also:
https://github.com/puppylinux-woof-CE/woof-CE/blob/testing/woof-code/rootfs-skeleton/etc/rc.d/rc.sysinit

Read more...