Re: [PLUG] VLC 3.0

[Tomas K]

Thanks all for trying, I really appreciate it.

I am now confused, on the top of just not knowing.
I guess, I was naive to seek simple functional description after seeing
all those mountains of buzzwords on compiz.org. 

It gets even worse with more unknown stuff such as Aero and Mac OS -
GPU and openGL. Those are not good reference points to me. I guess, it
has something to do with 3D, triangles, polygons, vertexes, fills, etc.
as in when you want to fill a window frame in games before it gets
passed to compositor for screen frame assembly? Or am I completely off
track? I thought X does that window decorations and screen frame
assembly work.

Anyway, thanks again for trying - I thought that it would be easy to
just ask. It always is. :-)

It seems that I would have to spent some non trivial time learning
about stuff I am not currently that much interested in. I am server,
web service, engineering and data junkie kind of a guy - html5, svg,
gnuplot, R, Qt,  can display any data I can think of at the moment.

This is completely unfamiliar area to me.

Cheers, Tomas

On Tue, 2018-05-08 at 21:19 -0700, Ben Koenig wrote:
> Not exactly
> 
> Compiz is a window manager that enables desktop effects similar to
> those 
> found in Mac OS X and Windows Aero. It accomplishes this by drawing 
> application windows to the screen using OpenGL and you graphics card.
> It 
> is a component of your Desktop Environment.
> 
> Maximize, minimize, and window manipulation that we expect on a
> desktop 
> computer is handed by a "Window Manager". Compiz provides these
> features 
> using opengl on your GPU, so it can go nuts and arrange your desktop
> in 
> 3 dimensions. Windows Aero MacOS Expo/Mission Control...  or in
> our 
> case the Compiz Desktop Cube.
> 
> 
> Most DE's such as KDE and whatnot have the ability to run a
> different 
> WM, therefore drastically changing the way windows are managed.
> 
> 
> On 05/08/2018 09:00 PM, Tomas Kuchta wrote:
> > Thanks.
> > So is it a desktop environment like Gnome/kde/lxde/... ?
> > 
> > On Tue, May 8, 2018, 8:23 PM John Jason Jordan 
> > wrote:
> > 
> > > On Wed, 09 May 2018 02:55:17 +
> > > Tomas Kuchta  dijo:
> > > 
> > > > What is Compiz?
> > > > What is it good for?
> > > > Can some explain it in a single digestible sentence?
> > > 
> > > Compiz allows the user to add a bit of art to their desktop by
> > > altering
> > > the appearance of windows and objects. As a common example that I
> > > use
> > > myself, you can add a shadow around windows. I keep my shadow
> > > thin and
> > > not very dark so it is just a hint. If you're not artistically
> > > inclined, then compiz is not for you.
> > > 
> > > Sorry, that was more than one sentence. Hopefully they were
> > > digestible.
> > > ___
> > > PLUG mailing list
> > > PLUG@pdxlinux.org
> > > http://lists.pdxlinux.org/mailman/listinfo/plug
> > > 
> > 
> > ___
> > PLUG mailing list
> > PLUG@pdxlinux.org
> > http://lists.pdxlinux.org/mailman/listinfo/plug
> 
> ___
> PLUG mailing list
> PLUG@pdxlinux.org
> http://lists.pdxlinux.org/mailman/listinfo/plug
___
PLUG mailing list
PLUG@pdxlinux.org
http://lists.pdxlinux.org/mailman/listinfo/plug

Re: [PLUG] Ubuntu MATE 18.04 frustrations

[Tomas K]

I'd be rather careful with modifying users on existing system - unless
you know exactly what you are doing. It can have pretty nasty side
effects appearing/re-appearing and entertaining you for long time to
come, at the 'best' possible times.

Local users and file ownership in Linux is established by UID.
User name is just a name for UID on a particular system.
  * so if you change user name (not UID), nothing has really changed
beside the user name displayed on given host.
  * if you see file owned by dick - there must be a user for that UID
with name dick on your system, otherwise you would see familiar nobody.
  * so based on UID on a particular systems you could be rsteff on
both, but with different UIDs. So when you copy those files -
preserving UID, you lose ownership and the new owner becomes someone
else or Mr/Mrs nobody.

When you use rsync to copy stuff between systems - the ownership gets
translated.
When you use NFS mounts (v3 or newer) between systems, the ownership
gets translated too as long as there are matching user names on all
connected hosts.

So as you can guess - it is good idea to follow bunch of rules when
managing a herd of standalone linux hosts without centralized user
management:
1. Synchronize UIDs and user names across all hosts.
   You do not have to have all users on all hosts, but their UIDs and
user name pairs should match.
  It is good practice to replicate users though an disable them when
needed, so that their old files still have an owner.
2. Synchronize GIDs and group names across all hosts - for precisely
the same reasons as 1.

Forward looking thoughts:
-
Following these two rules will make your shepherd's (no pun intended)
life much easier. Although following these rules across your herd can
get pretty tedious over time - leading some shepherds to use central
user, group, host,  management such as LDAP or FreeIPA (mentioned
here a few weeks ago). That combined with networked storage makes
computers orders of magnitude more useful.

Having central user management (AKA information services) and
associated DNS, NFS, DHCP, etc. used to be pretty expensive proposition
from IT, electricity and man power cost. This is no longer the case
with the advent of RaspberryPi type of computers which cost almost
nothing in IT and electricity cost and are pretty easy to connect to
your network and forget about them. RPi type computers also enable
pretty cheap/easy redundancy and backup - which is a must with
centralized services - because when they go down everything tends to
get stuck. Pulling another RPi from your drawer with identical setup to
the old one, and saving the day, makes you feel pretty darm good about
yourself.

Tomas

On Thu, 2018-05-17 at 16:26 -0700, Dick Steffens wrote:
> On 05/17/2018 04:21 PM, Rich Shepard wrote:
> > On Thu, 17 May 2018, Dick Steffens wrote:
> > 
> > > That part I know. What I've never looked into is how to change
> > > the 
> > > name of user dick to user rsteff, not from a file ownership 
> > > perspective, but from who is logged in. I'm sure I could create 
> > > another user named rsteff, but I don't think that's what I mean
> > > to 
> > > do, either.
> > 
> > Dick,
> > 
> >   Perhaps this will help:
> >  > 34075>
> > 
> >   For details, 'man usermod(8)':
> > 
> > NAME
> >    usermod - modify a user account
> > 
> > SYNOPSIS
> >    usermod [options] LOGIN
> > 
> > DESCRIPTION
> >    The usermod command modifies the system account files to 
> > reflect the
> >    changes that are specified on the command line.
> >   ...
> > 
> > -l, --login NEW_LOGIN
> >    The name of the user will be changed from LOGIN to
> > NEW_LOGIN.
> >    Nothing else is changed. In particular, the user's home 
> > directory
> >    or mail spool should probably be renamed manually to 
> > reflect the
> >    new login name.
> > 
> > HTH,
> > 
> 
> Thanks. That's probably what I need.
> 
___
PLUG mailing list
PLUG@pdxlinux.org
http://lists.pdxlinux.org/mailman/listinfo/plug

Re: [PLUG] Ubuntu MATE 18.04 frustrations

[Tomas K]

Great to hear that the file copy problem is out of the way.If you keep
that rsync command in your toolbox, it works incrementally. The next
time you run it, it will only copy new files on the other way. If you
change the source and destination, it will copy new files to the other
machine.

To see your local drive, you will need to mount it.
You said:
>  Can you see it mouted by: mount?
Yes. /dev/sda2 on /

That does not seem to be right - as that is your new installation root
directory.
Can you post output of following commands:
sudo fdisk -l
cat /etc/fstab
Thanks, T



On Tue, May 15, 2018, 8:10 AM Dick Steffens 
wrote:
> On 05/14/2018 11:54 PM, Tomas Kuchta wrote:
> 
> > I do not know what is Caja - G thinks that it is some JavaScript 
> 
> > framework. I doesn't matter anyway.
> 
> 
> 
> It's the GUI file manipulator. Caja is what MATE uses instead of
> Nautilus.
> 
> 
> 
> > I suggest that since you can share to your other machine, you copy
> the 
> 
> > your stuff by rsync from command line.
> 
> 
> 
> I'm more of a visual person, but I am an occasional command line
> user.
> 
> 
> 
> > This should work:
> 
> > cd yourMusicDir
> 
> > rsync -a --progress  rst...@192.168.0.xxx:/home/rsteff/Music/78s ./
> 
> >
> 
> > That should work and show the progress.
> 
> 
> 
> Yes, that is working. (Still has a ways to go, but looks like I
> would 
> 
> expect it to).
> 
> 
> 
> > When you get this out of the way,
> 
> > We can continue on the subject of your local drive access.
> 
> 
> 
> Thanks.
> 
> 
> 
> And yes, spell checkers can turn around and byte you from time to
> thyme.
> 
> 
> 
> 
___
PLUG mailing list
PLUG@pdxlinux.org
http://lists.pdxlinux.org/mailman/listinfo/plug

Re: [PLUG] VLC 3.0

[Tomas K]

Thanks Galen,

I believe that my lack of awareness of it is total indifference to 
any eye candy gimicks in UI. All I care about is practicality and
effectiveness of the interface together with good ergonomic. My hands
hurt with mouse and I hate distraction from the task at hand.

For that reason, I do not even know what to look for and it would not
surprise me if it is currently active on my desktop. I am quite
familiar with Knoppix and I still have no idea - I guess if it does not
have label or command line form - then it does not exist in my mind.
Well, since I am not aware of it, it must be good UI element - not in
the way of anything, not distracting me from writing this.

If I remember, I will ask you some day to point finger at it and show
me.

:-) Tomas

On Sat, 2018-05-19 at 12:04 -0700, Galen Seitz wrote:
> On 05/09/2018 02:18 AM, Tomas K wrote:
> > Thanks all for trying, I really appreciate it.
> > 
> > I am now confused, on the top of just not knowing.
> > I guess, I was naive to seek simple functional description after
> > seeing
> > all those mountains of buzzwords on compiz.org.
> 
> I'm late to the party, but an easy way to get a feel for what compiz
> can 
> do is to boot Knoppix.  Alternatively, I'm sure you can find compiz 
> demos on youtube.
> 
> (BTW, no compiz on my machines.  I've just encountered it with
> Knoppix)
> 
> galen
___
PLUG mailing list
PLUG@pdxlinux.org
http://lists.pdxlinux.org/mailman/listinfo/plug

Re: [PLUG] Ubuntu MATE 18.04 frustrations

[Tomas K]

I suspect that your local disk access issue may be permissions if you
changed uid between the installations.

That should be easy to check for:
ls -l /media/dick/a857c2f6-6e1a-4afe-9d6e-8109105a740
Do you own any files there?

If you do not own the files an directories and you think that you
should, you can do it this way:
sudo chown -R $USER /media/dick/a857c2f6-6e1a-4afe-9d6e-
8109105a740/someDirOrFile

Then, you can try if your access to the disk is any better. 

Hope it helps,Tomas

On Wed, 2018-05-16 at 09:00 -0700, Dick Steffens wrote:
> On 05/16/2018 01:48 AM, Tomas K wrote:
> > Great to hear that the file copy problem is out of the way.If you
> > keep
> > that rsync command in your toolbox, it works incrementally. The
> > next
> > time you run it, it will only copy new files on the other way. If
> > you
> > change the source and destination, it will copy new files to the
> > other
> > machine.
> > 
> > To see your local drive, you will need to mount it.
> > You said:
> > >    Can you see it mouted by: mount?
> > 
> > Yes. /dev/sda2 on /
> > 
> > That does not seem to be right - as that is your new installation
> > root
> > directory.
> 
> Sorry. Typo, or misread. The 482 GB drive is:
> 
> /dev/sdb1 on /media/dick/a857c2f6-6e1a-4afe-9d6e-8109105a740 type
> ext4 
> (rw,nosuid, nodev,relative,date=ordered,uhelper=udisks2)
> 
> > Can you post output of following commands:
> > sudo fdisk -l
> 
> Disk /dev/loop0: 86.7 MiB, 90923008 bytes, 177584 sectors
> Units: sectors of 1 * 512 = 512 bytes
> Sector size (logical/physical): 512 bytes / 512 bytes
> I/O size (minimum/optimal): 512 bytes / 512 bytes
> 
> 
> Disk /dev/loop1: 71.7 MiB, 75120640 bytes, 146720 sectors
> Units: sectors of 1 * 512 = 512 bytes
> Sector size (logical/physical): 512 bytes / 512 bytes
> I/O size (minimum/optimal): 512 bytes / 512 bytes
> 
> 
> Disk /dev/loop2: 86.6 MiB, 90759168 bytes, 177264 sectors
> Units: sectors of 1 * 512 = 512 bytes
> Sector size (logical/physical): 512 bytes / 512 bytes
> I/O size (minimum/optimal): 512 bytes / 512 bytes
> 
> 
> Disk /dev/loop3: 7.9 MiB, 8294400 bytes, 16200 sectors
> Units: sectors of 1 * 512 = 512 bytes
> Sector size (logical/physical): 512 bytes / 512 bytes
> I/O size (minimum/optimal): 512 bytes / 512 bytes
> 
> 
> Disk /dev/loop4: 7.5 MiB, 7884800 bytes, 15400 sectors
> Units: sectors of 1 * 512 = 512 bytes
> Sector size (logical/physical): 512 bytes / 512 bytes
> I/O size (minimum/optimal): 512 bytes / 512 bytes
> 
> 
> Disk /dev/loop5: 86.6 MiB, 90828800 bytes, 177400 sectors
> Units: sectors of 1 * 512 = 512 bytes
> Sector size (logical/physical): 512 bytes / 512 bytes
> I/O size (minimum/optimal): 512 bytes / 512 bytes
> 
> 
> Disk /dev/sda: 931.5 GiB, 1000204886016 bytes, 1953525168 sectors
> Units: sectors of 1 * 512 = 512 bytes
> Sector size (logical/physical): 512 bytes / 4096 bytes
> I/O size (minimum/optimal): 4096 bytes / 4096 bytes
> Disklabel type: gpt
> Disk identifier: 476788A2-28E0-4D9E-BF80-AEE9A112B59B
> 
> Device   Start    End    Sectors  Size Type
> /dev/sda1 2048    1050623    1048576  512M EFI System
> /dev/sda2  1050624 1953523711 1952473088  931G Linux filesystem
> 
> 
> Disk /dev/sdb: 465.8 GiB, 500107862016 bytes, 976773168 sectors
> Units: sectors of 1 * 512 = 512 bytes
> Sector size (logical/physical): 512 bytes / 512 bytes
> I/O size (minimum/optimal): 512 bytes / 512 bytes
> Disklabel type: dos
> Disk identifier: 0x000698ae
> 
> Device Boot Start   End   Sectors   Size Id Type
> /dev/sdb1  * 2048 960258047 960256000 457.9G 83 Linux
> /dev/sdb2   960260094 976771071  16510978   7.9G  5 Extended
> /dev/sdb5   960260096 976771071  16510976   7.9G 82 Linux swap /
> Solaris
> 
> > cat /etc/fstab
> 
> # /etc/fstab: static file system information.
> #
> # Use 'blkid' to print the universally unique identifier for a
> # device; this may be used with UUID= as a more robust way to name
> devices
> # that works even if disks are added and removed. See fstab(5).
> #
> #           
> # / was on /dev/sda2 during installation
> UUID=e1ff5590-d1ce-44be-9f10-7ce2bdd432c1 /   ext4 
> errors=remount-ro 0   1
> # /boot/efi was on /dev/sda1 during installation
> UUID=E8BB-3F15  /boot/efi   vfat    umask=0077  0   1
> /swapfile none    swap 
> sw  0   0
> 
> 
> I don't have the 482 GB drive in fstab. It was disconnected during
> the 
> install, since I put a bunch of stuff there I want to copy to the
> main 
> drive now that the install is finished. I just mount it as needed by 
> clicking on it in Caja's Devices list.
> 
___
PLUG mailing list
PLUG@pdxlinux.org
http://lists.pdxlinux.org/mailman/listinfo/plug

Re: [PLUG] [OT ? ?? ???] Linux and computer literacy

[Tomas K]

On Sun, 2018-06-17 at 20:39 -0500, Richard Owlett wrote:
> On 06/17/2018 05:50 PM, Tyrell Jentink wrote:
> > Oh... I guess I only answered one question...
> 
> 
> 
> > 
> > To the second question: When you want to do something that no one
> > else has ever done,
FTFY: or nobody seems to be interested in solving networking that way.

That being said - I noted people genuinely trying to advice you the
best they thought would help you solve your problem. There is little
point being too upset about the fact that ethernet and TCP/IP stack
rules networking these days.

It was definitely easier to build coal fired steam engine for
educational purpose or to use acoustic couplers for networking decades
ago than it is now. There sure was more accessible, and up to date
information, and useful hardware around. Though, coal fire steam
engines maybe coming back again.

> 
> Actually it has been done.
> It's just not popular ;/
> 
> > why would you expect anyone to know how to do it?
> 
> Actually I'm finding much information on "having it my way".
> I ask NARROWLY focused questions and am told "DON'T" ;<
> 
> > If you want to
> > be the first, then go do that... But people who need their hands
> > held don't
> > usually succeed in the undoable.
> 
> Not is it not only "doable". It *HAS* been done.
> An underlying motive is education.
> 
> > 
> > It's really not a "Computer literacy" issue... 
> 
> Agreed ;>
> It just surfaced at the same time and might be considered equally 
> 'off-topic".
> 
> 
> 
> > It's a "When you can't tell
> > who's in charge, it's probably you..." issue.
> > 
> > On Sun, Jun 17, 2018, 15:29 Richard Owlett 
> > wrote:
> > 
> > > I have two questions I wish to discuss.
> > > Both likely push the limits of being "On Topic".
> > > Is there a forum on which either (preferably *BOTH*) would be "On
> > > Topic"?
> > > 
> > > The more general question
> > > 
> > > On my last visit to our local public library I picked up a glossy
> > > 16
> > > page four color brochure titled _Free Computer Training and
> > > Professional
> > > Development Resources_ .
> > > 
> > > In that brochure the word "Mac" occurs twice. "Linux" *NEVER*
> > > occurs.
> > > Keyboarding skills are mentioned. They also ask the question
> > > "Does
> > > grammar matter?"
> > > 
> > > Can anyone point me to a "library like" resource  recognizing
> > > that
> > > Micro$oft has competition?
> > > 
> > > There has to be somebody with more "academic integrity".
> > > 
> > > My second question has only a temporal relationship to above.
> > > 
> > > For IDIOSYNCRATIC and weird local constraints I am working on
> > > networking
> > > *TWO* computers via a "USB MASTER-USB MASTER" cable. There are
> > > "cookbook"{sic/sick}" for "normal"(sic) users with ethernet.
> > > 
> > > I've been asking questions on a distro specific list.
> > > The predominant response is *DON'T*
> > > 
> > > Ignoring the *TRIVIAL* aspects of technical(sick) aspects
> > > I wish to do XYZ, That the rest of world does pqrz is
> > > *IRRELEVANT*
> > > I have already found *EDUCATIONAL* value in chosen path.
> > > 
> > > Where may I ask *MY* questions without being told to "get lost"?
> > > 
> > > TIA
> > > 
> 
> 
> ___
> PLUG mailing list
> PLUG@pdxlinux.org
> http://lists.pdxlinux.org/mailman/listinfo/plug
___
PLUG mailing list
PLUG@pdxlinux.org
http://lists.pdxlinux.org/mailman/listinfo/plug

Re: [PLUG] Quick question on linux man page coverage

[Tomas K]

Did you check if you could install some of the missing man/manuals you
care about? Not all man pages are selected for default install.

on my opensuse 42.3:
installed:
sudo zypper search man | awk '$1 !~ /^i/ && $2 ~ /(^|-)(man|manual)$/
{print} $1 ~ /^i/ && $3 ~ /(^|-)(man|manual)$/ {print}' | grep -c '^i'
1
not-installed:
sudo zypper search man | awk '$1 !~ /^i/ && $2 ~ /(^|-)(man|manual)$/
{print} $1 ~ /^i/ && $3 ~ /(^|-)(man|manual)$/ {print}' | grep -cv '^i'
47

I looked for the sl utility - no joy finding it, SAD!
  no xsnow either, SAD!
  found xeyes, huraaahhh!

Tomas

On Sun, 2018-06-17 at 15:50 -0700, logical american wrote:
> A brief question on sparse linux documentation for executable
> programs
> 
> I did a survey of the 15,000+ executables on my openSuse Leap v42.3
> OS 
> which has about 8869 software packages loaded according to zypper,
> but 
> 76% of them have no man page at all. I did find dozens of programs 
> running, such as gvfs, which are intrinsic to the OS and some
> apparently 
> embedded in the kernal, most running under systemctl control, but
> with 
> no documentation.
> 
> For openSuse Leap v42.3, it appears that the linux developers just
> want 
> the product out the door and have not documented the /etc folder very
> well.
> 
> Should we be concerned that 3/4 of the programs running on a linux OS
> do 
> not have a man page?
> 
> I was a bit surprised to find this rather high ratio. Is it
> surprising?
> 
> Randall
> 
> ___
> PLUG mailing list
> PLUG@pdxlinux.org
> http://lists.pdxlinux.org/mailman/listinfo/plug
___
PLUG mailing list
PLUG@pdxlinux.org
http://lists.pdxlinux.org/mailman/listinfo/plug

Re: [PLUG] New computer

[Tomas K]

For what it is wort it 42.2 will not boot your new HW, you either
install 42.3 or do not waste your time. I know, I have tried that on
the same CPU and chipset before.

I am not trying to start any flame wars, I just do not know much about
this deep into Ubuntu - 42.3 has backported your CPU and chipset
features - that is probably why it booted, installed and worked while
the other distros, do not out of the box.

At this point, I would just try to replicate what we did and what
worked - and that was opensuse 42.3. Opensuse works regardless of the
secure boot enabled/disabled as it is properly signed by MS, so UEFI
trusts it.

Anyway, good luck, time to start a log book, so that you can remember
what you tried and how.

Tomas

On Tue, 2018-02-20 at 22:36 -0800, John Jason Jordan wrote:
> On Tue, 20 Feb 2018 18:37:29 -0800
> Tomas  dijo:
> 
> > 
> > If this is still relevant, unresolved, understanding you Ubuntu or
> > whatever the distro you want as requirement.
> 
> Still unresolved. Ultimately I want Xubuntu 18.04, but since it is
> not
> released yet my current plan is to install 17.10 and then do a
> dist-upgrade in a couple months. But right now I just want to get
> *something* that will boot.
> 
> The OpenSUSE that you installed will not boot. It starts, and then
> hangs. I burned the ISO to a DVD and tried to use it to reinstall. It
> got as far as 
> > 
> > > 
> > > > 
> > > > OpenSUSE Leap 42.2 installation programv5.0.87
> starting udev...
> And then it hangs.
> For the version you installed there are other options, including a
> recovery mode. I tried it but it hung on:
> 32.9125001 ata10.00: status: { DRDY }
> 32.9125091 ata10: hard resetting link
> 
> I have also tried Xubuntu 17.10, 17.04, 16.04, and they all hang.
> 
> I tried Knoppix 7.7 and it starts, but then hangs on 'Starting
> dbus...'
> 
> > 
> > Try to walk through this list in order:
> > 1. Is your bios up to date? If not, update it first.
> >    That often helps with new shiny things PC.
> 
> The BIOS has a flash utility that will go out to the internet and
> flash
> the BIOS. I used it and t finished saying '"The BIOS is up to date."
> But it made no difference, still can't get anything to boot.
> 
> > 
> > 2. Unless you have done something to the bios settings, boot
> >    priorities or started another installation, you should be
> >    able to get the openSuse installation going.
> 
> It hangs. See above comment.
> 
> > 
> >    - I would walk back the bios changes you might have made
> >    and get that going for piece of mind and to become familiar
> >    with the bios settings - particularly related to UEFI and
> >    various legacy settings.
> >    It is not given that disabling UEFI makes magically things work.
> >    These days UEFI is more reliable than disabling it because
> >    MS Win needs it, and needs it in locked down state.
> >    - UEFI typically depends on number of bios settings - and often
> >    the PC does not boot until the only combination is right.
> >    - You can always use the "Reset CMOS" jumper to set BIOS to
> >    defaults, if there is no Bios way resetting it to defaults.
> 
> This computer will never boot any version of Windows, so that is
> irrelevant.  
> Your comment about resetting the BIOS is something I had not yet
> tried,
> so just now I did it. And I know it was successful because I got a
> different start screen. But I still can't get anything to boot.
> 
> > 
> > 3. Another thing is to try, after you are back to default bios
> >    settings would be to choose the SSD as boot in the boot menu
> > (F11).
> >    Bios sometimes sets the last used boot device as default for
> >    the next boot.
> 
> This BIOS has an option to get the list of drives and what is on them
> by pressing F11 when it boots. I can select either optical drive or
> the
> SSD, and the SSD shows OpenSUSE twice - once as 'secure boot' and the
> other blank. No matter, neither will boot. If there is a bootable USB
> stick that will also appear in the list. 
> 
> > 
> > 4. Failing all the above, try to make yourself opensuse 42.3 usb
> > boot
> >    stick. It just worked, so the chances are that it will just work
> >    again. It would not achieve your favorite distro setup, but it
> >    could take you to the right track and turn the brick to useful
> >    device until you have new Ubuntu in a few months.
> 
> I already tried that, except I burned it to a DVD because I'm running
> out of USB sticks. And I used 42.2 because I already had it
> downloaded.
> It hangs like everything else.
> 
> There is a glimmer of hope: You installed Leap 42.3, so it must have
> booted from your USB stick. As far as I know that is the only time
> when
> any distro has actually booted. Somehow I need to get back to the
> state
> it was in when you did that, but clearing the CMOS didn't do it. 
> 
> It's my bedtime. Maybe in the morning I'll think of something else.
> In
> the meantime, thanks for the 

Re: [PLUG] Gaming graphics cards?

[Tomas K]

I think that you are against the physics here, not against any
particular piece of a software.

Normal RGB pixel is 3*8b=24b
1920*1080*24b*30fps=1,492,992,000
Yes, that is 1.5Gb/s just in raw pixels no image frame or network
overhead. Gamers typically prefer 60fps to feel fast shooters and need
to respond pretty fast.

Those display cables carry pretty awesome bandwidth and with pretty low
latency. Just imagine what that network speed and latency will do to
the game, MS+magic or not. That traffic must come out of Dfx card, to X
then through network stack back to X and another Gfx card. 

If you are thinking about saving money by throwing 10Gb network at it,
I would sleep on it. Cheap PC with cheap graphics card can give you
better and cheaper gaming experience.

Alternatively you could do it other way around and get decent
workstation and run server workload on it when not gaming. That would
work just fine.

Just saying, Tomas

On Sat, 2018-02-17 at 22:47 -0800, Tyrell Jentink wrote:
> On Feb 17, 2018 20:55, "Tomas Kuchta" 
> wrote:
> 
> In regards to your server room VM setup idea - how are you planning
> to
> display the stuff from far away server room?
> 
> 
> If Linux ends up sucking for this, I will cry a small tear, then
> install
> Windows in the KVM Virtual Machine, and use RemoteFX (
> https://en.m.wikipedia.org/wiki/RemoteFX)
> 
> But SURELY X is flexible enough to outdo anything Microsoft built
> into
> RDP... I just haven't made it that far in my research yet.
> ___
> PLUG mailing list
> PLUG@pdxlinux.org
> http://lists.pdxlinux.org/mailman/listinfo/plug
___
PLUG mailing list
PLUG@pdxlinux.org
http://lists.pdxlinux.org/mailman/listinfo/plug

[PLUG] Found unopened WinXP install disk + license - any takers?

[Tomas K]

I am trying to rid of some old junk - and found unopened, unused,
shrink wrapped WinXP SP1 installation disk with the installation guide,
etc. books Anybody needs/wants it?

This proposition is/sounds pretty bad, particularly on this email list
- but there seems to be some old stuff collectors around.

"This offer will expire next weekend"

Tomas

___
PLUG mailing list
PLUG@pdxlinux.org
http://lists.pdxlinux.org/mailman/listinfo/plug

Re: [PLUG] How to restore external drive

[Tomas K]

It seems that it would be safer to have the DAS array attached
permanently to your NAS and access it over the network. That is if your
NAS has eSATA port.

That way it would be permanently attached like internal drives and that
should avoid these type of synchronization errors - Especially when you
use it as RAID0.

RAID arrays, especially in striping configuration, do not like to be
detached or put to sleep without proper sync and umount.

If attaching it permanently to NAS is not an option, autofs with
reasonable timeout avoiding the DAS power saving mode, instead of mount
in fstab, would probably help too.

Hope it helps, Tomas

On Sun, 2018-10-14 at 18:26 -0700, John Jason Jordan wrote:
> On Sun, 14 Oct 2018 13:45:14 -0700 (PDT)
> Rich Shepard  dijo:
> 
> > On Sun, 14 Oct 2018, Ben Koenig wrote:
> > 
> > > Please copy/paste the following command into your terminal and
> > > post
> > > the output here so that we can offer sensible advice.
> > > 
> > > ls -l /dev/sd*  
> > 
> > Ben,
> > 
> >   Might lsblk provide John with the same information?
> 
> It does, and I already used it last night. It found /dev/sdc, but not
> the partition /dev/sdc1. Gparted had a utility to search a disk for a
> filesystem, which I left running overnight. It found nothing, so this
> morning I bit the bullet, recreated and formatted the partition, and
> then started copying everything from the backup Synology NAS. I
> started
> the copying at about 9am, and it is now working on letter 'O.' I hope
> to be done before bedtime tonight. 
> ___
> PLUG mailing list
> PLUG@pdxlinux.org
> http://lists.pdxlinux.org/mailman/listinfo/plug
___
PLUG mailing list
PLUG@pdxlinux.org
http://lists.pdxlinux.org/mailman/listinfo/plug

Re: [PLUG] How to restore external drive

[Tomas K]

A few comments...

> However, I have long been impatient with the speed of the Mediasonic;
> moving or renaming a file can take over a minute while the drives
> whir and the lights dance.

Renaming or moving files on the same drive should take fraction of a
second regardless of the file size or the speed of connection to it. If
this what you really experience, you should consider using mv on a
command line.

> During the restore from backup the GUI (Nautilus) gave me a popup
> with continuous information on the progress, and it reported a steady
> 115-116 Mbps transfer rate, so that is apparently the max that my
> home ethernet can do.

That 100-120MBs is about maximum you can observe over 1Gbs ethernet to
a NAS for sequential, large file transfers. Recent Synology NAS`s are
capable of that without problem even for RAID5 arrays.

Please note - sequential, large file transfers - is the keyword.
Once we talk about random or small file access - you are into 40MBs-ish 
territory even for fast SSDs - no what the speed of connection.

See some benchmarks such as:
SSDs: https://www.anandtech.com/show/12485/the-plextor-m8v-sata-ssd-rev
iew/5
NVMEs: https://www.anandtech.com/show/13254/the-toshiba-xg6-1tb-ssd-rev
iew-first-96l-3d-nand/5

> Thinkpad P72 (not yet available in the US) or a P71 that I can buy
> right now. Both come with Intel Thunderbolt 3 USB-C ports that are
> capable of 40Gbps, compared to the 5Gbps of my USB 3.0 connection or
> 3GBps of eSATA.

So, connecting to HDD or SSD from single user, single computer at 5Gbs,
10Gbs or even 40Gbs is rather irrelevant. Especially when you consider
a laptop which is very limited in what it is capable of compared to a
desktop - especially in high bandwidth usage applications.

Anyway - large files are mainly media - Encoding a film with x264
proceeds at glacial 5Mbs-ish speed these days. So even if you could
encode 5 streams at the same time, you would probably be fine with 1Gbs
ethernet to a NAS.

Backups are incremental, and do not slow you down, especially if your
NAS writes to your DAS directly or between two NAS`s. I assume that you
are not watching the backups to make them go faster. :-)

--- If we would be talking (= not dreaming) about low cost 4-6
drive Synology NAS with 2.5/5/10Gbs ethernet + $150-ish 8 port
2.5/5/10Gbs switch to go with it - that would be different cup-a-tea.
---

Storage is slow, and who likes politics or taxes.

Hope it helps, Tomas

On Mon, 2018-10-15 at 08:39 -0700, John Jason Jordan wrote:
> On Mon, 15 Oct 2018 01:27:31 -0700
> Tomas K  dijo:
> 
> > It seems that it would be safer to have the DAS array attached
> > permanently to your NAS and access it over the network. That is if
> > your
> > NAS has eSATA port.
> > 
> > That way it would be permanently attached like internal drives and
> > that
> > should avoid these type of synchronization errors - Especially when
> > you
> > use it as RAID0.
> > 
> > RAID arrays, especially in striping configuration, do not like to
> > be
> > detached or put to sleep without proper sync and umount.
> 
> The Synology NAS does not have an eSATA port, but it does have a USB
> port, so I could probably attach the Mediasonic to it. And I agree
> that
> doing so might make the connection more stable - the NAS is across
> the
> room, while the Mediasonic is on my desk where stuff gets moved
> around
> a bit. However, I have long been impatient with the speed of the
> Mediasonic; moving or renaming a file can take over a minute while
> the
> drives whir and the lights dance. I lust for SSDs to replace the two
> WD
> Red Pro drives in it, but the cost of 16TB of SSDs always makes me
> discard the notion. If I attach the Mediasonic to the Synology I
> wonder
> about access time compared to its current USB 3 connection. 
> 
> During the restore from backup the GUI (Nautilus) gave me a popup
> with
> continuous information on the progress, and it reported a steady 115-
> 116
> Mbps transfer rate, so that is apparently the max that my home
> ethernet
> can do.
> 
> I have been shopping for a new computer and I'm currently looking at
> a
> Thinkpad P72 (not yet available in the US) or a P71 that I can buy
> right now. Both come with Intel Thunderbolt 3 USB-C ports that are
> capable of 40Gbps, compared to the 5Gbps of my USB 3.0 connection or
> 3GBps of eSATA. Then again, the WD Red Pro drives are SATA-3, which
> are
> rated at 6Gbps, so until I go to SSDs there wouldn't be much to gain.
> And the only faster enclosures available are Mediasonic rated at
> 10Gbps. What would be really cool is 16TB of SSDs mounted inside the
> laptop itself. 
> 
> > If attaching it permanently to NAS is not an option, autofs with
> > reasonable timeout avoiding the DAS power sa

Re: [PLUG] Indecipherable rsync error message SOLVED

[Tomas K]

It appears that you either did not have the directory mounted or it was
in power saving mode (disks spun down, etc.)

I would recommend that you to either manually make sure that the source
and destinations are both available or setup some wait loop in your
script to allow and trigger mounting.

Ages ago, I wrote/used this simple bash function which waits for disk
to spin up and become available on the file system.

function waitForDir () {
  # waits for directory to not be empty
  local maxWait=30
  local c=0
  if (( $# != 1 )); then
>&2 echo "ERROR: Please give directory as cmd argument."
exit 0
  fi
  while (( $(ls $1 | wc -l) == 0 )); do
if (( $c > $maxWait )); then
  >&2 echo "ERROR: Directory $1 missing content after $maxWait
seconds"
  exit 0
else
  c=$(( $c + 1 ))
fi
sleep 1
  done
}

Then I used it this way:
  waitForDir /media/jjj/Movies
Note: that it needs some content to be in the directory.

Best regards, Tomas

On Sat, 2018-09-29 at 11:01 -0700, John Jason Jordan wrote:
> On Fri, 28 Sep 2018 22:39:53 -0700
> John Jason Jordan  dijo:
> 
> Problem solved:
> 
> cd ..
> cd /media/jjj/Movies
> 
> Now the rsync command works as expected, as do other commands.
> ___
> PLUG mailing list
> PLUG@pdxlinux.org
> http://lists.pdxlinux.org/mailman/listinfo/plug
___
PLUG mailing list
PLUG@pdxlinux.org
http://lists.pdxlinux.org/mailman/listinfo/plug

Re: [PLUG] Indecipherable rsync error message SOLVED

[Tomas K]

Try hdparam to control sleep time.

Here is Ubuntu link:
https://askubuntu.com/questions/39760/how-can-i-control-hdd-spin-down-t
ime

If that does not work, and you cannot change your DAS settings, you
might need to look at powertop to check if your USB is not going to
sleep.

Synology NAS have settings to set disk sleep timer. It is a great way
to prolong disk life and reduce heat and noise. By default it is off.
 
Tomas

On Sat, 2018-09-29 at 16:16 -0700, John Jason Jordan wrote:
> On Sat, 29 Sep 2018 14:28:43 -0700
> Tomas K  dijo:
> 
> > It appears that you either did not have the directory mounted or it
> > was
> > in power saving mode (disks spun down, etc.)
> 
> The Movies USB device is always mounted, but yes, it seems to go into
> sleep mode all the time. Any disk activity brings it out of sleep
> mode,
> although it can take a full minute before it is available. The
> Synology
> is NAS, and it doesn't appear to suffer from being in sleep mode.
> 
> Why couldn't I just make a little bash script:
> 
> cd ..
> cd /media/jjj/Movies
> 
> 
> That seems simpler, simple enough that even I understand it. :)
> 
> Having said that I suffer from this drive having to refresh itself
> all
> the time. I mean, dozens of times a day I sit here twiddling my
> thumbs
> while I wait and listen to the discs rumble and watch the lights
> blink. I use this disk all day long, yet it seems to go into sleep
> mode
> after five minutes of inactivity. 
> ___
> PLUG mailing list
> PLUG@pdxlinux.org
> http://lists.pdxlinux.org/mailman/listinfo/plug
___
PLUG mailing list
PLUG@pdxlinux.org
http://lists.pdxlinux.org/mailman/listinfo/plug

Re: [PLUG] Rsync again

[Tomas K]

The answer is - Do not start rsync until your NAS is mounted and
working.

Simple if statement in your backup script should take care of it.
You can do it number of ways - here is one - checking that there is any
content in the directory before starting rsync (in bash):

#!/bin/bash
backupTargetDir=/media/jjj/Synology
cd $backupTargetDir
# wait 30 seconds for the directory to mount
sleep 30
filesInDir=$(ls | wc -l)
if (( $filesInDir > 0 )); then
  echo "INFO: Backing up files to $backupTargetDir"
  rsync 
else
  echo "WARNING: No files in $backupTargetDir"
fi 

Another way would be to use mount to verify that your NAS is mounted


Tomas

On Sat, 2018-10-06 at 11:37 -0700, John Jason Jordan wrote:
> On Sat, 6 Oct 2018 10:17:05 -0700
> John Jason Jordan  dijo:
> 
> I need some advice with respect to the following command, which makes
> my Synology NAS device a mirror of my external USB storage enclosure:
> 
>   rsync -rptog --progress --stats --delete
>   --exclude-from=/media/jjj/Movies/rsync_exclusions
> /media/jjj/Movies/ /media/jjj/Synology
> 
> The first problem is that sometimes I get an error message that it
> can't find /media/jjj/Movies. I found that a simple 'cd ..' followed
> by
> 'cd /media/jjj/Movies' solves the problem. I plan to convert the
> rsync
> command above to a simple script, adding the cd commands in front of
> the rsync command.
> 
> But there is a second, even worse problem: It sometimes can't find
> the
> Synology NAS device. And if it can't find the device it dumps
> everything into the local hard drive where /media/jjj/Synology is
> (supposed to be) mounted. The hard drive has a couple hundred GB of
> free
> space, but since the external USB device has over 7TB of data on it
> the
> local disk is quickly filled up, and then the rsync command fails
> with
> 'the device is full.'
> 
> The Synology is mounted with this line in fstab:
> 
>   192.168.1.115:/volume1/Synology /media/jjj/Synology nfs
>   auto,user 0 0
> 
> I need to add another command before the rsync command that checks to
> be sure the Synology is mounted where it's supposed to be and mount
> it
> there if it is not already mounted. 
> 
> And finally, I need to add something after the rsync command that
> checks to be sure the command executed as it is supposed to and give
> me
> a popup that it did, or that it did not. I have gxmessage installed
> and
> I could probably figure out how to make it pop up the messages, but I
> don't know how to check that the rsync command executed properly.
> Could
> I just check the two filesystems to see if they are equal in all
> respects?
> 
> Suggestions?
> ___
> PLUG mailing list
> PLUG@pdxlinux.org
> http://lists.pdxlinux.org/mailman/listinfo/plug
___
PLUG mailing list
PLUG@pdxlinux.org
http://lists.pdxlinux.org/mailman/listinfo/plug

Re: [PLUG] NO JANUARY PLUG GENERAL MEETING

[Tomas K]

I am be interested :-)

On Tue, 2019-01-01 at 19:40 -0800, Russell Senior wrote:
> Excellent! Moar days to procrastinate!
> 
> If people want to hear about my temperature sensor array re-
> engineering
> project, an update to the talk I did in 2013, including zeromq and
> software
> defined radio, maybe we could do an informal thing at Lucky Lab NW
> this
> Thursday. It is a work in progress. Let me know if you are
> interested!
> Otherwise I'll huddle at home and continue poking at it.
> 
___
PLUG mailing list
PLUG@pdxlinux.org
http://lists.pdxlinux.org/mailman/listinfo/plug

Re: [PLUG] VirtualBox on Slackware

[Tomas K]

This is turning into general discussion - still I see the original
point of this email tread in a need to reuse existing VirtualBox
virtual machine image/installation.

So it makes little sense to try do anything but VirtualBox.

Often, the decision to use vBox was made decade or even longer ago.
There was very little of KVM back then, so the choices were Xen, vBox
or VMware. I would argue that Xen versus vBox 10-15 years ago was
really matter of taste, remember vBox was being actively opened by Sun
Microsystem.

I do not think that it is good idea doing fresh long term virtual
machine build in VMware or vBox today - Given the almost infinite
desire/need for longevity of these installations; and our experience
with fully commercial software over the course of multiple
acquisitions. If VMware or vBox changes their mind about their license
or support - one would find it extremely difficult, if not impossible,
to install that VM on your next hardware 3-5 years from now.

Tomas

On Sun, 2018-12-02 at 13:43 -0800, King Beowulf wrote:
> On 12/2/18 6:13 AM, Chuck Hast wrote:
> > I see all of this and wonder if you have given thought to VMWare
> > Player,
> > for personal use it is
> > free, I use it all of the time when I cannot get something to run
> > under
> > Wine, it sees the USB
> > stuff quite well.
> > 
> > I have not had issues with it seeing devices. I use it to program
> > radios, I
> > am running Win 10
> > on a Ubuntu Mate 18.04 host. The radios look like serial USB
> > devices to the
> > OS. Most of them
> > work just fine under wine, but one which seems the program is all
> > buggered
> > up and even a
> > trick to run on Windows. Oh, I also use it for some Garmin stuff
> > that will
> > not work on Wine.
> > Garmin should be ashamed, they sure use Linux behind the GPS
> > screens on
> > their products
> > but cannot come up with a decent piece of software for Linux to
> > talk to
> > their products.
> > 
> > Here is a bit from them about running workstation (player is WS
> > without the
> > license)  under
> > Slack:
> > https://kb.vmware.com/s/article/113
> > 
> > 
> 
> Chuck,
> 
> It all depends on one's linux zealotry and personal preferences.
> 
> qemu - fully open source
> Virtualbox - partially open source (uses qemu code in places)
> VMWare - fully proprietary.
> 
> qemu uses KVM modules already present in the Linux kernel.  VB and
> VMware have their own mystery binaries.
> 
> The link you posted has some factual errors.  Slackware uses a BSD
> style
> init system, which is NOT "unusual" and is also fully compatible with
> System V init scripts.  See also http://www.slackware.com/config/init
> .php
> 
> Secondly, the statement
> 
> "Keep in mind that Slackware is not an explicitly supported
> distribution, mostly because of its unusual file layout."
> 
> bespeaks of a great deal of ignorance regarding Linux file systems
> and
> layouts.  I personally prefer qemu (fewer extra dependencies but
> harder
> CLI configuration), with Virtualbox as a second choice (nice gui).
> 
> -Ed
> 
> 
> 
> 
> ___
> PLUG mailing list
> PLUG@pdxlinux.org
> http://lists.pdxlinux.org/mailman/listinfo/plug
___
PLUG mailing list
PLUG@pdxlinux.org
http://lists.pdxlinux.org/mailman/listinfo/plug

Re: [PLUG] Problem with Google Earth Install

[Tomas K]

I was kind of curious what is Google Earth. So 

I tried GEarth on bunch of Ubuntu installs both 16.04 and 18.04
I experienced no problems with GoogleEarth after doing following:
1. Download:
   wget -Nc https://dl.google.com/dl/earth/client/current/google-earth-
pro-stable_current_amd64.deb
   You might need to visit https://www.google.com/earth/versions/ in
browser
2. Install:
   sudo dpkg -i google-earth-pro-stable_current_amd64.deb
   (the installation added repository to
/etc/apt/sources.list.d/google-earth-pro.list)
   So, it should be updating itself normally when running apt upgrade
3. Start it:
   google-earth-pro &
It works as expected - according to Google's web screenshots.

My Ubuntu installations are mostly default without any of odd bits and
pieces of customization often discussed here.

In terms of your issues I'd try to:
* remove it and reinstall by:
  sudo apt remove google-earth-pro-stable
  sudo apt update
  sudo install google-earth-pro-stable

If that does not help - are you using X or Wayland? I believe that I am
 using X on both 16.04 as well as 18.04.

I cannot offer much help with GUI type problems as I am not a GUI kind
of guy. It just work for me in the default shape and form and I do not
feel the need to fight little changes and annoyanceswith
trivial workarounds - I do not mind adjusting self as things change.

Hope it helps, Tomas

On Fri, 2018-12-07 at 07:37 -0800, Michael Barnes wrote:
> On Fri, Dec 7, 2018 at 6:24 AM Rich Shepard  >
> wrote:
> 
> > On Fri, 7 Dec 2018, Michael Barnes wrote:
> > 
> > > I think that may have been the problem with my previous posts. I
> > > tried to
> > > include a screenshot and I don't think this list allows pictures.
> > > Not
> > 
> > sure
> > > where to post the screenshot.
> 
> 
> Okay, screen shot is at http://nwgrr.org/SS-GE.png
> 
> You can see my desktop is shown in the primary screen, the toolbar
> thing is
> on the left, and the actual map is a small box in the upper left
> corner. I
> can manipulate the small map, zoom, scroll, etc. but I can't really
> do
> anything useful as it is so small.
> 
> Thanks for any ideas.
> Michael
> ___
> PLUG mailing list
> PLUG@pdxlinux.org
> http://lists.pdxlinux.org/mailman/listinfo/plug
___
PLUG mailing list
PLUG@pdxlinux.org
http://lists.pdxlinux.org/mailman/listinfo/plug

Re: [PLUG] LibreOffice Writer: delete character

[Tomas K]

Ctrl-d is EOF - I'd be careful here.

On Thu, 2018-12-06 at 09:29 -0800, Rich Shepard wrote:
>    In LibreOffice-6.1.2, Tools -> Configure -> Keyboard allows user-
> assigned
> actions to Ctrl-D for deleting everything but a single character. Is
> there a
> way to assign the same function as the Del key to Ctrl-d?
> 
> TIA,
> 
> Rich
> ___
> PLUG mailing list
> PLUG@pdxlinux.org
> http://lists.pdxlinux.org/mailman/listinfo/plug
___
PLUG mailing list
PLUG@pdxlinux.org
http://lists.pdxlinux.org/mailman/listinfo/plug

Re: [PLUG] Slackware -- programs missing

[Tomas K]

Are you running it as root or using sudo?

just checking ...

On Sat, 2018-11-24 at 18:48 -0800, Dick Steffens wrote:
> On 11/24/18 6:40 PM, Ben Koenig wrote:
> > Skip the second argument then, just run "chroot /mnt/hd"
> > Always make sure you are root and check for spaces and whatnot.
> 
> bash-4.4# chroot /mnt/hd
> chroot: failed to run command '/bin/bash': No such file or directory
> 
> I don't know if this means anything, but
> basn-4.4# which bash
> /usr/bin/bash
> 
___
PLUG mailing list
PLUG@pdxlinux.org
http://lists.pdxlinux.org/mailman/listinfo/plug

Re: [PLUG] Slackware -- programs missing

[Tomas K]

If I'd to make a bet - I'd be betting on some /lib versus /lib64 snafu
while running from live cd/dvd/usb

If that is so perhaps binding the missing lib or fixing broken link
could fix it.

Tomas

On Sat, 2018-11-24 at 19:03 -0800, Ben Koenig wrote:
> The problem is that even though you appear to have the program it
> doesn't
> want to find it. WHAT???
> I think your install might be broken, but before I declare that post
> the
> output of the following commands:
> 
> ls -l /mnt/hd/  (making sure your filesystem looks like a root
> filesystem)
> ldd /mnt/hd/bin/bash (checking binary links for the bash program on
> your
> internal drive to see if it behaves like an executable file)
> uname -m (checking the arch that you are currently running in case of
> a
> 32/64bit mixup)
> 
> On Sat, Nov 24, 2018 at 7:01 PM Dick Steffens 
> wrote:
> 
> > On 11/24/18 6:55 PM, Tomas K wrote:
> > > Are you running it as root or using sudo?
> > > 
> > > just checking ...
> > 
> > Running as root.
> > 
> > --
> > Regards,
> > 
> > Dick Steffens
> > 
> > ___
> > PLUG mailing list
> > PLUG@pdxlinux.org
> > http://lists.pdxlinux.org/mailman/listinfo/plug
> > 
> 
> ___
> PLUG mailing list
> PLUG@pdxlinux.org
> http://lists.pdxlinux.org/mailman/listinfo/plug
___
PLUG mailing list
PLUG@pdxlinux.org
http://lists.pdxlinux.org/mailman/listinfo/plug

Re: [PLUG] Slackware -- programs missing

[Tomas K]

Dick, If you are confused about my comment - please ignore my post and
follow Ben's advice to post output from:

ls -l /mnt/hd/
ldd /mnt/hd/bin/bash
uname -a

That should give enough info to crack this.

My apology for adding second voice ...
T

On Sat, 2018-11-24 at 19:19 -0800, Ben Koenig wrote:
> On Sat, Nov 24, 2018 at 7:10 PM Tomas K  >
> wrote:
> 
> > If I'd to make a bet - I'd be betting on some /lib versus /lib64
> > snafu
> > while running from live cd/dvd/usb
> > 
> > If that is so perhaps binding the missing lib or fixing broken link
> > could fix it.
> > 
> > Tomas
> > 
> 
> Accepted. If you are wrong then you have to be the next to install
> Slackware.
> ___
> PLUG mailing list
> PLUG@pdxlinux.org
> http://lists.pdxlinux.org/mailman/listinfo/plug
___
PLUG mailing list
PLUG@pdxlinux.org
http://lists.pdxlinux.org/mailman/listinfo/plug

Re: [PLUG] Slackware -- programs missing

[Tomas K]

chroot command simply sets different root directory for a command -
thus that command:
  a) sees different root directory
  b) cannot access anything outside that directory (aka chroot jail)

Example:
  1. Imagine you just mounted:
 mount /dev/sda2 /mnt/hd
  2. Executing:
 sudo chroot /mnt/hd /bin/bash
 will execute bash from inside /mnt/hd (running /mnt/hd/bin/bash)
 as if you would have booted from there

What Ben, most likely, wants to guide you through is running lilo as if
booted from the disk mounted on /mnt/hd - thus updating the Lilo
config.

I will not guess further because I forgot most things about Lilo long
time ago. The Linuxes I use boot using Grub for long time and/or boot
using UEFI directly these days. 

Hope it makes sense.
Tomas

On Sat, 2018-11-24 at 17:26 -0800, Dick Steffens wrote:
> On 11/25/18 5:06 PM, Ben Koenig wrote:
> > Ah ok. I think I know what is happening.
> > 
> > I'm assuming here that your filesystem on /dev/sda2 is functional
> > and has a
> > working install of Slackware. A quick description of the LILO
> > setup.
> > 
> > lilo.conf sets up the configuration for the bootloader. However,
> > after
> > creating your config you have to run the lilo command to upload
> > your
> > kernel, initrd, and the bootloader itself to your MBR.
> > 
> > Your file looks correct, however it still fails to find your root
> > filesystem. I'm hoping this is because lilo was run before you
> > finished
> > your config, which means all we need to do is run it again :-)
> > 
> > To do this, you want to boot the live media again, and manually
> > mount sda2
> > with the following command (as root):
> > mount /dev//sda2 /mnt/hd
> > 
> > Now you can list the contents of /mnt/hd/
> > /mnt/hd/ is a default folder in slackware that you can use for
> > recovery
> > jobs such as this. Temporary mounts here will not interrupt the
> > system. At
> > this point you should verify the existence of the init program with
> > the
> > following command:
> > 
> > $ ls /mnt/hd/sbin/init
> > /sbin/init*
> > 
> > The error you are getting is referring to that file. When the
> > kernel boots,
> > it mounts the rootfs as defined by lilo and then runs /sbin/init.
> > If init
> > does not exist, no init found.
> > I'm pretty sure you have /sbin/init, but it doesn't hurt to double
> > check.
> > Assuming it is still there then we need to do a quick chroot and
> > re-run
> > lilo.
> > 
> > Go ahead and boot live, mount sda2 as described above and verify
> > that
> > /sbin/init exists.
> 
> bash-4.4# mount /dev//sda2 /mnt/hd
> bash-4.4# ls /mnt/hd/sbin/init
> /mnt/hd/sbin/init
> 
> Also confirmed that I see it with ls /mnt/hd/sbin.
> > The next step after that is a chroot and I need to go
> > look those commands up because I can never remember them off the
> > top of my
> > head.
> > We know why it doesn't want to boot, so all we need to do now is
> > fix it. :)
> 
> I appreciate your help. I've heard of chroot, but never had it
> explained 
> in a form I actually understood.
> 
___
PLUG mailing list
PLUG@pdxlinux.org
http://lists.pdxlinux.org/mailman/listinfo/plug

Re: [PLUG] Slackware Install, 2nd try

[Tomas K]

Beware - this is intended to be funny, but factual post, I cannot help
it! This post is based on my speculation, not factual observation.

Difficulties with creating bootable disk, especially when using GUIs,
are often caused by a defective mouse operator. When USB is plugged in,
 GUI tends to show some nice colorful, and often movin image. That
proves irresistible to the mouse operator and they must to click on
that shiny colorful thing.

The act of mousing that USB GUI thing typically makes GUI to mount the
thing and displaying something to make the operator happy. Like a
candy.

When that USB thingy is mounted, naturally, attempts to dd stuff to it
typically fails - because it is mounted. Using other methods to make
said bootable USB also fail for the same reason - operator mounted it
with his mouse.

To conclude - either do not mess about by clicking on the inserted USB
GUI things - before trying to write a bootable image to that USB - OR
use some mouse fu tricks to unmout every bit of said USB thingy before
attempting to write bootable image to it.

Hope it hits the spot, based on actual live experience,
Tomas

On Sun, 2018-11-25 at 17:33 -0800, Dick Steffens wrote:
> > On 11/25/18 4:52 PM, Ben Koenig wrote:
> > > 
> > > Make sure that /dev/sdc exists, reinsert the USB stick to get it
> > > to 
> > > show up.
> 
> In the event there's something wrong with the USB stick, I decided
> to 
> try a different one. This one currently has a Linux Mint 19 live
> image 
> on it. When I plug it in, I get the same screwy thing of two icons 
> appearing on the desktop. One is named SlackDVD, and the other is
> named 
> LinuxMint 19 MATE 64 bit. While it's not the Unix way to fix things,
> I'm 
> going to reboot this machine, since something screwy is going in
> with 
> USB mounting.
> 
> I'm saving this as a draft, and will report back after the reboot.
> 
> -
> 
> After the reboot, only the Linux Mint 19 icon appears on the
> desktop. 
> mount reports:
> 
> /dev/sdc1 on /media/dick/Linux Mint 19 MATE 64-bit type iso9660 
> (ro,nosuid,nodev,relatime,nojoliet,check=s,map=n,blocksize=2048,uid=1
> 000,gid=1000,dmode=500,fmode=400,uhelper=udisks2)
> 
> I ran
> 
> sudo umount /dev/sdc1
> 
> and then ran
> 
> sudo dd if=~/Desktop/slackware64-14.2-install-dvd.iso of=/dev/sdc
> bs=1M
> 
> It's still running, so I'm guessing it's doing its thing. I'll
> report 
> back after it finishes.
> 
___
PLUG mailing list
PLUG@pdxlinux.org
http://lists.pdxlinux.org/mailman/listinfo/plug

Re: [PLUG] how to distribute binary

[Tomas K]

Unless you feel the need to resolve your issues:
 - You might need to distribute in appimage or inside a container for
docker.

Tomas

On Thu, 2018-11-22 at 09:18 -0800, VY wrote:
> Dear All
> 
> I am facing an issue and want to get some ideas how to move forward.
> 
> I have developed a Python script, and there are some PyQt within as I
> built
> a very simple UI.   I used Cython to convert it to C and compile it
> to
> binary
> (dynamic linking).  It works if and only if the right version of all
> libraries are installed
> on the deployed machine.  That's obvious since I am doing dynamic
> linking.
> I then tried do link statically but just could not get past the issue
> of
> libc.   It seems
> one of the things I linked needed some special version of libc,
> beyond the
> libc I have
> already linked statically.
> 
> Given the issues that I face, I like to ask the group what's the best
> approach
> to compile and distribute a binary program across different flavors
> of
> Linux?
> 
> thanks
> 
> -v
> ___
> PLUG mailing list
> PLUG@pdxlinux.org
> http://lists.pdxlinux.org/mailman/listinfo/plug
___
PLUG mailing list
PLUG@pdxlinux.org
http://lists.pdxlinux.org/mailman/listinfo/plug

Re: [PLUG] Android upgrade notices (was: Wireless access point issues)

[Tomas K]

Some times these access points have a rule configured whether you can
connect to their web interface via WLAN or LAN only?

Did you try to access the web interface from both LAN and WLAN sides?

Tomas

On Wed, 2019-02-27 at 07:45 -0800, Rich Shepard wrote:
> On Wed, 27 Feb 2019, Rich Shepard wrote:
> 
> > No. When I selected the WiFi icon Android found all the access
> > points in
> > the neighborhood with mine at the top of the list. I selected mine
> > and
> > entered my user password for network hosts as I never set up a user
> > password for me on the WAP.
> 
> I've moved this to PLUG-talk as the focus has shifted to Android, not
> linux.
> 
> Backing up and puting this WAP issue in context will help us all.
> 
> Here in the office I've no need for the WAP; it serves only to allow
> the
> Sony Vaio to access the router and FiOS connection to the 'Net. Every
> host
> in the office has an ethernet connection to the router; when I'm not
> in the
> office I've no need for a computer and 'Net access. So, why did I
> start the
> thread?
> 
> I read that when Android updates the OS and installed applications on
> the
> Nokia they can take a long time so it's better to use the phone's
> WiFi
> capabilities rather than the mobile carrier's.
> 
> The Nokia has Android 9 installed so I imagine that a long running OS
> upgrade is not in the immediate future and I do not need to connect
> it to
> the WAP here.
> 
> But I do have a relevant question because I've not had an Android-
> powered
> mobile phone before. How do I know when there are upgrades available
> that
> should be applied?
> 
> If I'm notified somehow, perhaps by a text message sent by Google, I
> can
> walk to the nearby library branch, connect to their WiFi for the
> upgrade,
> and read while the upgrade(s) are processed.
> 
> I'll use the phone as a phone here in the office, but not to access
> the Web
> or other 'Net functions. I hope this clarifies the situation. I'll
> re-work
> the WAP some time in the future as it's not immediately needed.
> 
> Thanks all,
> 
> Rich
> 
> ___
> PLUG mailing list
> PLUG@pdxlinux.org
> http://lists.pdxlinux.org/mailman/listinfo/plug
___
PLUG mailing list
PLUG@pdxlinux.org
http://lists.pdxlinux.org/mailman/listinfo/plug

Re: [PLUG] Open read-only partition read-write

[Tomas K]

Here is my advice:
  1. go outside, away from your computer - panic
  2. when you calm down do back and send output of: mount
  3. If you notice that something is mounted more than once - and
without remount option - wait for other people to respond with ideas
  4. If 3. is the case rub your lucky charm, do your lucky dance, pray
- whatever works for you to perform lucky miracles
  5. do not panic - if 3. is the case and you somehow mounted root
partition multiple times (mount should fight you) and your computer is
still running and you write these emails on it - it is probably not too
bad.

Tomas


On Fri, 2019-02-22 at 22:47 -0800, John Jason Jordan wrote:
> On Fri, 22 Feb 2019 11:56:02 -0800
> John Jason Jordan  dijo:
> 
> > I should know how to do this, but I can't figure it out.
> 
> OK, I just discovered significant evidence, although I don't know
> what
> it means.
> 
> 1) / (root) is now mounted in /media/jjj
> 
> 2) Looking at it I noticed that mnt is not there. Looking further
> down
> I found it as a file. Right-clicking on it in PCFileMan gives the
> popup: "mnt" inode/x-corrupted type.
> 
> Throughout today's saga I could not help but think that there was
> something seriously wrong with the filesystem on /. The other
> filesystems may also be corrupted, although their bizarre behavior
> may
> be do to their being mounted incorrectly.
> 
> The first step is to fix /, which I know there is a command for. But
> it's too late to remember it. Tomorrow morning I will tackle this
> first
> thing. Also I will try to figure out how this happened.
> ___
> PLUG mailing list
> PLUG@pdxlinux.org
> http://lists.pdxlinux.org/mailman/listinfo/plug
___
PLUG mailing list
PLUG@pdxlinux.org
http://lists.pdxlinux.org/mailman/listinfo/plug

Re: [PLUG] FYI - ROSEBURG , Oregon HAS FIBER IN ~50% OF COUNTY PER Douglas Fast Net (DFN)

[Tomas K]

There has been many replies - I've been diligently removing them
because they were all hopeless and/or desperate, despite finding some
truth in all them.

As you can see, I could not let this go - it is symptomatic to how the
society organizes and clusters into incoherent/clan based groups with
diverse, but local incentive structure. Naturally, this is trivial to
exploit not only in telecommunications, but also in general - economy,
labor markets and politics.

In my opinion, PDX is unlikely to build common data distribution
infrastructure, because:
  * It is relatively cheap and easy to wire up metropolitan area with
good return on investment. So, the market encumbents can effectively
respond to protect their interest.
  * decent and fair data distribution networks are low margin, low tech
race to the bottom - this does not attract competing investment -
unless you remove a lot of friction from the market place.
  * Market place friction reduction needed:
* dividing distribution and service markets - (networks from ISP,
power cables from transmission and generation, etc.)
* mandating inter-connectivity
* mandating open access for customers and service suppliers
* mandating non discriminatory and transparent pricing - this alone
could improve many bad things - imagine the same price for everyone
with the same service and charged as published
* ban on service bundling other than - sum of available individual
parts
  * local PDX government have strong incentives favoring status quo:
* the less you stand out - the higher the chance to be elected
* any change invites controversy - which sadly is poison chance to
re-election
* this is multi term project for more than just one person
* funding is no problem as there is good ROI
* any local improvement is likely to further stress housing market
* in general decline and local price inflation works for local
government - less work, more taxes, hard to be held responsible
  * Small places like Sandy and Roseburg are successful because:
* "easy to wire up metropolitan are with quick return on
investment" does not apply
* they have no other choice than community effort
* they can minimize the planning friction
* they often have access to rural infrastructure funding outside
the local encumbent political influence
* lack of other opportunities and can do today approach

So, the only way this can be done in PDX would be through:
  a) local government measures easily labeled and quashed as
"socialist" or unfair to the encumbents.
  b) By the community, without formal local government help - willing
to put 10-20 year of hard labor to it.

Scale:
PDX metro is roughly 40x25km on 80m grid = 12,500km fiber +
trenches/poleAttachments + 10k switches/routers + 60k residence
attachments. This is pretty small at today's scales - still, wiring up
1000 miles + 6k residences per year, every year is not DIY project.

I, like many others, would love decent and fair data infrastructure,
but for the reasons above - I doubt that there is group of 30-50 people
with unlimited perseverance, wisdom, patience and the right skills
willing to come together and spend life doing this for others despite
all the odds.

Linux + free software, on the other hand is land of free an plentiful
opportunities, even in Roseburg <-- while we wait eternity for decent
government for the people. :-)

-T

On Sat, 2019-02-02 at 07:57 -0800, mitch Stanley wrote:
> We are thinking of relocating to Roseburg , OR  & surprise they have
> Fiber
> per DFN  . I was pleasantly surprised ! I
> feel
> the prices are reasonable - 100Mbps @39.99 / 250Mbps @$54.99 & Giga@
> $89.99
> 
> 
> If Roseburg can do it , Portland should have accomplished this
> too!The
> only thing affecting this lack of Fiber in PDX is  entrenched
> interests -
>  Why offer
> fiber when the Big Guys are offering  Broad band @ $65/mo  ( My
> Xfinity
> bill for 100Mbps )  & you don't have to invest $$$ anymore plus you
> can Buy
> - SKY, Unviversal Pictures , & maybe a TV Network , like NBC.
> 
> Sorry for preaching to the Choir , I do not know whom to write ,
> call,ect
> my frustration with the entrenched system.  We would like to live in
> Semi
> Rural area
> but lack of Broadband drastically limits one choices but to larger
> cities /
> urban areas.
> 
> Have a Great Weekend,
> 
> Sincerely ,
> 
> Mitch S.
> ___
> PLUG mailing list
> PLUG@pdxlinux.org
> http://lists.pdxlinux.org/mailman/listinfo/plug
___
PLUG mailing list
PLUG@pdxlinux.org
http://lists.pdxlinux.org/mailman/listinfo/plug