Discussion:
Centos7 : some ( official )help wanted
(too old to reply)
j***@telenet.be
2015-06-30 08:00:09 UTC
Permalink
Hello,

I have good results on our first site where we run K12Linux on Centos7 and I am about to ugrade a second, larger
site as well.

I just had a look at the K12Linux website again and just wanted to say I don't agree with this:

Modern users probably do not want to use this type of thin client because video (like Youtube) requires too much bandwidth over remote X desktops, and scalability of X over ssh encrypted tunnels is rather poor. K12Linux is considered a legacy solution for existing deployments of LTSP-type networks and is currently supported only on the legacy EPEL6 platform. Porting to more modern systemd-based Fedora and EPEL7 is technically possible but is not considered a priority given the drawbacks of the legacy LTSP solution.


I think a lot of organsisations have employees who 's first priority is not watching youtube, all though my users have no problems with that on K12Linux.
What hurts me is that because people can not get this to work on Centos7/Rhel7, they are moving to Ubuntu.

I have K12Linux/Centos6 running for about 3 years, and it's incredibly robust. And I think it's the same on Centos7.

greetings, Johan
Radek Bursztynowski
2015-07-01 11:51:47 UTC
Permalink
p { margin-bottom: 0.25cm; line-height: 120%; }


Hello,


I agree with Johan.
I have more than 3 years experience with K12Linux and I am very
satisfied with K12Linux. I use CentOS 6.x and tests CentOS 7.
Regarding video I see limitations LTSP, but in my opinion it isn't
essential prinfold. Let me express my observations:


On LTSP web site I
found information that video could take up to 70 Mb/s. My test showed
that video takes up to 130 Mb/s. LTSP terminal doesn't cache video
because LTSP terminal presents the video proces running on the
server. So, we can say that about 10 users watchnig video on LTSP
terminals fullill gigabit Ethernet. It is real limitation, but this
limitation concerns not only LTSP, but all similar solutions, where
users use thin cliet architecture. What can be done to solve it? I
see two areas of activity:


1) The extension of
the band trasmission between the server and LTSP terminals. But in
this case we should remember, that all users (not only LTSP users)
are limited by the gateway. This solution is not very difficult and
is not expensive (I don't think about 10 Gb/s Ethernet now). It is
real.


Let me add, that we
can add to LTSP thin client image codecs and video player and use
video player locally on LTSP terminal. In this case mentioned above
limitation are fetched to the same situation like with standard PC.


2) To understand
that LTSP solution is not addressed to all using. I estimate that
LTSP solution could be sufficient fot 70% users, in particular for
office users. If the user should to perform some special tasks equip
him with PC or workstation. If the user is standard office user LTSP
solution will be enough.


Let say that most of
us move from one flat to another one including our furniture few
times in our live. It isn't the reason to buy the truck instead of
automobile – most of us buys automobile, not trucks. In my country,
particular in public sector, people practice in IT area strange
philosophy. All of them need the trucks! I try to explain them – if
you are buying the computers using own money – it is your choice,
your responsibility, your expens. But if you are money disposer only
you must to match the tool to the tasks. But they still buy the
trucks!


I see, that quote
appended by Johan aussmes that all people, all workers watch videos
all working day!




Am i right?


Best regards,

Radek




----- Original Message -----
From: ***@telenet.be
Sent: Tue, 6/30/2015 10:00am
To: "Support list for open source software in schools." <***@redhat.com>
Subject: [K12OSN] Centos7 : some ( official )help wanted

Hello,

I have good results on our first site where we run K12Linux on Centos7 and I am about to ugrade a second, larger
site as well.

I just had a look at the K12Linux website again and just wanted to say I don't agree with this:

Modern users probably do not want to use this type of thin client because video (like Youtube) requires too much bandwidth over remote X desktops, and scalability of X over ssh encrypted tunnels is rather poor. K12Linux is considered a legacy solution for existing deployments of LTSP-type networks and is currently supported only on the legacy EPEL6 platform. Porting to more modern systemd-based Fedora and EPEL7 is technically possible but is not considered a priority given the drawbacks of the legacy LTSP solution.

I think a lot of organsisations have employees who 's first priority is not watching youtube, all though my users have no problems with that on K12Linux.
What hurts me is that because people can not get this to work on Centos7/Rhel7, they are moving to Ubuntu.

I have K12Linux/Centos6 running for about 3 years, and it's incredibly robust. And I think it's the same on Centos7.

greetings, Johan
_______________________________________________
K12OSN mailing list
***@redhat.com
https://www.redhat.com/mailman/listinfo/k12osn
For more info see <http://www.k12os.org>;
j***@telenet.be
2015-07-01 15:43:55 UTC
Permalink
Hello,

I agree especially with the analogy with the truck. And even more if it's the tax payers money.

Let me just add:
Centos7 is much faster then Centos6. Let's face it, that is important to desktop users.
This is not the time to drive people to other distro's.
With that I'm not saying that any distro is better than another distro.

Greetings, Johan


----- Oorspronkelijk bericht -----
Van: "Radek Bursztynowski" <***@bursztynowski.waw.pl>
Aan: "Support list for open source software in schools." <***@redhat.com>, "Support list for open source software in schools." <***@redhat.com>
Verzonden: Woensdag 1 juli 2015 13:51:47
Onderwerp: Re: [K12OSN] Centos7 : some ( official )help wanted







p { margin-bottom: 0.25cm; line-height: 120%; }


Hello,


I agree with Johan.
I have more than 3 years experience with K12Linux and I am very
satisfied with K12Linux. I use CentOS 6.x and tests CentOS 7.
Regarding video I see limitations LTSP, but in my opinion it isn't
essential prinfold. Let me express my observations:


On LTSP web site I
found information that video could take up to 70 Mb/s. My test showed
that video takes up to 130 Mb/s. LTSP terminal doesn't cache video
because LTSP terminal presents the video proces running on the
server. So, we can say that about 10 users watchnig video on LTSP
terminals fullill gigabit Ethernet. It is real limitation, but this
limitation concerns not only LTSP, but all similar solutions, where
users use thin cliet architecture. What can be done to solve it? I
see two areas of activity:


1) The extension of
the band trasmission between the server and LTSP terminals. But in
this case we should remember, that all users (not only LTSP users)
are limited by the gateway. This solution is not very difficult and
is not expensive (I don't think about 10 Gb/s Ethernet now). It is
real.


Let me add, that we
can add to LTSP thin client image codecs and video player and use
video player locally on LTSP terminal. In this case mentioned above
limitation are fetched to the same situation like with standard PC.


2) To understand
that LTSP solution is not addressed to all using. I estimate that
LTSP solution could be sufficient fot 70% users, in particular for
office users. If the user should to perform some special tasks equip
him with PC or workstation. If the user is standard office user LTSP
solution will be enough.


Let say that most of
us move from one flat to another one including our furniture few
times in our live. It isn't the reason to buy the truck instead of
automobile – most of us buys automobile, not trucks. In my country,
particular in public sector, people practice in IT area strange
philosophy. All of them need the trucks! I try to explain them – if
you are buying the computers using own money – it is your choice,
your responsibility, your expens. But if you are money disposer only
you must to match the tool to the tasks. But they still buy the
trucks!


I see, that quote
appended by Johan aussmes that all people, all workers watch videos
all working day!




Am i right?


Best regards,

Radek




----- Original Message -----
From: ***@telenet.be
Sent: Tue, 6/30/2015 10:00am
To: "Support list for open source software in schools." <***@redhat.com>
Subject: [K12OSN] Centos7 : some ( official )help wanted

Hello,

I have good results on our first site where we run K12Linux on Centos7 and I am about to ugrade a second, larger
site as well.

I just had a look at the K12Linux website again and just wanted to say I don't agree with this:

Modern users probably do not want to use this type of thin client because video (like Youtube) requires too much bandwidth over remote X desktops, and scalability of X over ssh encrypted tunnels is rather poor. K12Linux is considered a legacy solution for existing deployments of LTSP-type networks and is currently supported only on the legacy EPEL6 platform. Porting to more modern systemd-based Fedora and EPEL7 is technically possible but is not considered a priority given the drawbacks of the legacy LTSP solution.

I think a lot of organsisations have employees who 's first priority is not watching youtube, all though my users have no problems with that on K12Linux.
What hurts me is that because people can not get this to work on Centos7/Rhel7, they are moving to Ubuntu.

I have K12Linux/Centos6 running for about 3 years, and it's incredibly robust. And I think it's the same on Centos7.

greetings, Johan
_______________________________________________
K12OSN mailing list
***@redhat.com
https://www.redhat.com/mailman/listinfo/k12osn
For more info see <http://www.k12os.org>;

_______________________________________________
K12OSN mailing list
***@redhat.com
https://www.redhat.com/mailman/listinfo/k12osn
For more info see <http://www.k12os.org>
Warren Togami Jr.
2015-07-01 15:09:42 UTC
Permalink
Disagree all you want. If no developer is willing to do the work, then it
cannot happen.
Post by j***@telenet.be
Hello,
I have good results on our first site where we run K12Linux on Centos7 and
I am about to ugrade a second, larger
site as well.
I just had a look at the K12Linux website again and just wanted to say I
*Modern users probably do not want to use this type of thin client because
video (like Youtube) requires too much bandwidth over remote X desktops,
and scalability of X over ssh encrypted tunnels is rather poor. K12Linux is
considered a legacy solution for existing deployments of LTSP-type networks
and is currently supported only on the legacy EPEL6 platform. Porting to
more modern systemd-based Fedora and EPEL7 is technically possible but is
not considered a priority given the drawbacks of the legacy LTSP solution.*
I think a lot of organsisations have employees who 's first priority is
not watching youtube, all though my users have no problems with that on
K12Linux.
What hurts me is that because people can not get this to work on
Centos7/Rhel7, they are moving to Ubuntu.
I have K12Linux/Centos6 running for about 3 years, and it's incredibly
robust. And I think it's the same on Centos7.
greetings, Johan
_______________________________________________
K12OSN mailing list
https://www.redhat.com/mailman/listinfo/k12osn
For more info see <http://www.k12os.org>
j***@telenet.be
2015-07-01 15:24:36 UTC
Permalink
----- Oorspronkelijk bericht -----

Van: "Warren Togami Jr." <***@togami.com>
Aan: "Support list for open source software in schools." <***@redhat.com>
Verzonden: Woensdag 1 juli 2015 17:09:42
Onderwerp: Re: [K12OSN] Centos7 : some ( official )help wanted

Disagree all you want. If no developer is willing to do the work, then it cannot happen.

On Tue, Jun 30, 2015 at 1:00 AM, < ***@telenet.be > wrote:



Hello,

I have good results on our first site where we run K12Linux on Centos7 and I am about to ugrade a second, larger
site as well.

I just had a look at the K12Linux website again and just wanted to say I don't agree with this:

Modern users probably do not want to use this type of thin client because video (like Youtube) requires too much bandwidth over remote X desktops, and scalability of X over ssh encrypted tunnels is rather poor. K12Linux is considered a legacy solution for existing deployments of LTSP-type networks and is currently supported only on the legacy EPEL6 platform. Porting to more modern systemd-based Fedora and EPEL7 is technically possible but is not considered a priority given the drawbacks of the legacy LTSP solution.


I think a lot of organsisations have employees who 's first priority is not watching youtube, all though my users have no problems with that on K12Linux.
What hurts me is that because people can not get this to work on Centos7/Rhel7, they are moving to Ubuntu.

I have K12Linux/Centos6 running for about 3 years, and it's incredibly robust. And I think it's the same on Centos7.

greetings, Johan




Warren,

Then it should say on the site " no developer wants to do the work".
Now it says K12Linux is legacy.

And Warren, I am grateful for the work that you put in for K12Linux.

What I'm saying here, and I think what Radek is saying, is it's not legacy, It works.

Greetings, Johan

<blockquote>



_______________________________________________
K12OSN mailing list
***@redhat.com
https://www.redhat.com/mailman/listinfo/k12osn
For more info see < http://www.k12os.org >

</blockquote>



_______________________________________________
K12OSN mailing list
***@redhat.com
https://www.redhat.com/mailman/listinfo/k12osn
For more info see <http://www.k12os.org>
Jim Kinney
2015-07-01 19:10:15 UTC
Permalink
Post by j***@telenet.be
Aan: "Support list for open source software in schools." <
Verzonden: Woensdag 1 juli 2015 17:09:42
Onderwerp: Re: [K12OSN] Centos7 : some ( official )help wanted
Disagree all you want. If no developer is willing to do the work,
then it cannot happen.
Post by j***@telenet.be
Hello,
I have good results on our first site where we run K12Linux on
Centos7 and I am about to ugrade a second, larger
site as well.
I just had a look at the K12Linux website again and just wanted to
Modern users probably do not want to use this type of thin client
because video (like Youtube) requires too much bandwidth over
remote X desktops, and scalability of X over ssh encrypted tunnels
is rather poor. K12Linux is considered a legacy solution for
existing deployments of LTSP-type networks and is currently
supported only on the legacy EPEL6 platform. Porting to more modern
systemd-based Fedora and EPEL7 is technically possible but is not
considered a priority given the drawbacks of the legacy LTSP
solution.
I think a lot of organsisations have employees who 's first
priority is not watching youtube, all though my users have no
problems with that on K12Linux.
What hurts me is that because people can not get this to work on
Centos7/Rhel7, they are moving to Ubuntu.
I have K12Linux/Centos6 running for about 3 years, and it's
incredibly robust. And I think it's the same on Centos7.
greetings, Johan
Warren,
Then it should say on the site " no developer wants to do the work".
Now it says K12Linux is legacy.
And Warren, I am grateful for the work that you put in for K12Linux.
What I'm saying here, and I think what Radek is saying, is it's not legacy, It works.
Legacy doesn't mean it doesn't work. It simply means no further
development will be done for that particular setup. Sometimes a program
becomes legacy because no one is working on it. Sometimes because it
get superseded by newer projects.
K12Linux on CentOS 6 is a great tool. But porting it to be fully CentOS
7 ready will require development from people other than Warren.
K12Linux will be around for many more years since CentOS 6 will be
around as well.
I think it's time to look at alternate methods of achieving a similar
result. The PXE boot aspect will need to stay but the server platform
needs to be revisited. The IT world has embraced virtual machines.
Maybe a thin client should be running a VM client that provides console
to a remote VM running on a server farm. The spice client is remarkably
light and can push screen bits as well if not better than NX and RDP
(and way faster than X over SSH or VNC). Using a management tool like
Ovirt would allow tech support to build and maintain student machines
as pool devices and thus only require updates to a single machine. Some
coding is required to provide a single sign on screen that would
connect to the next available VM console (which then auto-mounts the
students home directory). The VM client (spice) can be 32 bit as well
as 64 so it will run on older client hardware and it supports USB pass
through for thumb drives. Sound also works.
3 years ago I was doing a demo talk of Ovirt and connected through 2
layers of VPNs back to my work lab to show both windows and Linux
desktops. I started a browser and went to YouTube and the flash video
played smoothly and with sound synched to the video.
As KVM, the technology behind Ovirt virtual machines, supports "memory
ballooning", different VMs that need to use the same block of memory
can share the block (read only - best for OS and application memory)
and thus can perform similar types of resource sharing that K12Linux
provided. Best of all, each user is truly isolated from all the others
so an application crash (ahem - flash) will only affect the single user
even with shared memory.
Post by j***@telenet.be
Post by j***@telenet.be
Greetings, Johan> >
_______________________________________________
K12OSN mailing list
https://www.redhat.com/mailman/listinfo/k12osn
For more info see <http://www.k12os.org>
_______________________________________________
K12OSN mailing list
https://www.redhat.com/mailman/listinfo/k12osn
For more info see
_______________________________________________
K12OSN mailing list
***@redhat.com>
https://www.redhat.com/mailman/listinfo/k12osn
Post by j***@telenet.be
For more info see <http://www.k12os.org>
--
James P. Kinney III

Every time you stop a school, you will have to build a jail. What you
gain at one end you lose at the other. It's like feeding a dog on his
own tail. It won't fatten the dog.
- Speech 11/23/1900 Mark Twain

http://heretothereideas.blogspot.com/
Warren Togami Jr.
2015-07-02 16:07:01 UTC
Permalink
Post by j***@telenet.be
Warren,
Then it should say on the site " no developer wants to do the work".
Now it says K12Linux is legacy.
And Warren, I am grateful for the work that you put in for K12Linux.
What I'm saying here, and I think what Radek is saying, is it's not legacy, It works.
Legacy doesn't mean it doesn't work. It simply means no further
development will be done for that particular setup. Sometimes a program
becomes legacy because no one is working on it. Sometimes because it get
superseded by newer projects.
K12Linux on CentOS 6 is a great tool. But porting it to be fully CentOS 7
ready will require development from people other than Warren. K12Linux will
be around for many more years since CentOS 6 will be around as well.
Thank you. You understand the situation exactly.

CentOS 7 actually contains all the same components as CentOS 6 but in a
different form. It would be theoretically possible to adapt those pieces
to work in the same manner as CentOS 6. There would be drawbacks though
... like CentOS 7 is x86_64 only, so the client OS would no longer be able
to be built from the same OS as the server.
Post by j***@telenet.be
I think it's time to look at alternate methods of achieving a similar
result. The PXE boot aspect will need to stay but the server platform needs
to be revisited. The IT world has embraced virtual machines. Maybe a thin
client should be running a VM client that provides console to a remote VM
running on a server farm. The spice client is remarkably light and can push
screen bits as well if not better than NX and RDP (and way faster than X
over SSH or VNC). Using a management tool like Ovirt would allow tech
support to build and maintain student machines as pool devices and thus
only require updates to a single machine. Some coding is required to
provide a single sign on screen that would connect to the next available VM
console (which then auto-mounts the students home directory). The VM client
(spice) can be 32 bit as well as 64 so it will run on older client hardware
and it supports USB pass through for thumb drives. Sound also works.
3 years ago I was doing a demo talk of Ovirt and connected through 2
layers of VPNs back to my work lab to show both windows and Linux desktops.
I started a browser and went to YouTube and the flash video played smoothly
and with sound synched to the video.
As KVM, the technology behind Ovirt virtual machines, supports "memory
ballooning", different VMs that need to use the same block of memory can
share the block (read only - best for OS and application memory) and thus
can perform similar types of resource sharing that K12Linux provided. Best
of all, each user is truly isolated from all the others so an application
crash (ahem - flash) will only affect the single user even with shared
memory.
KVM would be quite nice as the SPICE remote desktop protocol is
significantly better than X. SPICE would handle remote USB, full screen
video and sound in a better way than X/ltspfs/pulseaudio had done in the
past. It however has major drawbacks in memory overhead compared to the
old LTSP model, existing servers would be able to handle far fewer clients.

Warren Togami
Gianluca Cecchi
2015-07-02 17:04:28 UTC
Permalink
Post by Warren Togami Jr.
Thank you. You understand the situation exactly.
CentOS 7 actually contains all the same components as CentOS 6 but in a
different form. It would be theoretically possible to adapt those pieces
to work in the same manner as CentOS 6. There would be drawbacks though
... like CentOS 7 is x86_64 only, so the client OS would no longer be able
to be built from the same OS as the server.
Just to share information, there has been some work in place to have a
CentOS 7 i686 architecture version too.
See for example this thread for the beta release:
http://lists.centos.org/pipermail/centos-devel/2015-June/013426.html

I installed it (configured as a desktop system) and promises good.

Gianluca
j***@telenet.be
2015-07-02 18:26:02 UTC
Permalink
----- Oorspronkelijk bericht -----

Van: "Warren Togami Jr." <***@togami.com>
Aan: "Support list for open source software in schools." <***@redhat.com>
Verzonden: Donderdag 2 juli 2015 18:07:01
Onderwerp: Re: [K12OSN] Centos7 : some ( official )help wanted

On Wed, Jul 1, 2015 at 12:10 PM, Jim Kinney < ***@gmail.com > wrote:




<blockquote>

Warren,

Then it should say on the site " no developer wants to do the work".
Now it says K12Linux is legacy.

And Warren, I am grateful for the work that you put in for K12Linux.

What I'm saying here, and I think what Radek is saying, is it's not legacy, It works.




Legacy doesn't mean it doesn't work. It simply means no further development will be done for that particular setup. Sometimes a program becomes legacy because no one is working on it. Sometimes because it get superseded by newer projects.

K12Linux on CentOS 6 is a great tool. But porting it to be fully CentOS 7 ready will require development from people other than Warren. K12Linux will be around for many more years since CentOS 6 will be around as well.

</blockquote>


Thank you. You understand the situation exactly.

CentOS 7 actually contains all the same components as CentOS 6 but in a different form. It would be theoretically possible to adapt those pieces to work in the same manner as CentOS 6. There would be drawbacks though ... like CentOS 7 is x86_64 only, so the client OS would no longer be able to be built from the same OS as the server.

<blockquote>


I think it's time to look at alternate methods of achieving a similar result. The PXE boot aspect will need to stay but the server platform needs to be revisited. The IT world has embraced virtual machines. Maybe a thin client should be running a VM client that provides console to a remote VM running on a server farm. The spice client is remarkably light and can push screen bits as well if not better than NX and RDP (and way faster than X over SSH or VNC). Using a management tool like Ovirt would allow tech support to build and maintain student machines as pool devices and thus only require updates to a single machine. Some coding is required to provide a single sign on screen that would connect to the next available VM console (which then auto-mounts the students home directory). The VM client (spice) can be 32 bit as well as 64 so it will run on older client hardware and it supports USB pass through for thumb drives. Sound also works.

3 years ago I was doing a demo talk of Ovirt and connected through 2 layers of VPNs back to my work lab to show both windows and Linux desktops. I started a browser and went to YouTube and the flash video played smoothly and with sound synched to the video.

As KVM, the technology behind Ovirt virtual machines, supports "memory ballooning", different VMs that need to use the same block of memory can share the block (read only - best for OS and application memory) and thus can perform similar types of resource sharing that K12Linux provided. Best of all, each user is truly isolated from all the others so an application crash (ahem - flash) will only affect the single user even with shared memory.


</blockquote>


KVM would be quite nice as the SPICE remote desktop protocol is significantly better than X. SPICE would handle remote USB, full screen video and sound in a better way than X/ltspfs/pulseaudio had done in the past. It however has major drawbacks in memory overhead compared to the old LTSP model, existing servers would be able to handle far fewer clients.

Warren Togami

Hello,

thanks for taking the time for the detailed and very helpful answers.
I now understand the situation much better.
No matter how this continues, with K12Linux or Spice, I have nothing but respect for the developers.

Greetings, Johan

_______________________________________________
K12OSN mailing list
***@redhat.com
https://www.redhat.com/mailman/listinfo/k12osn
For more info see <http://www.k12os.org>

Radek Bursztynowski
2015-07-02 07:16:43 UTC
Permalink
p { margin-bottom: 0.25cm; line-height: 120%; }


Jim,


Thanks you for your
clear explanation. It means, that there are two separated cases. The
first one is the K12Linux's usefulness and the second one closing
down the project. In spite of K12Linux's usefulness this project is
shutting down. Is it appropriate understanding?


If yes, I regret. I
don't evaluate why, but I regret.


Are there any
chances to reanimate K12Linux project?


Best regards,

Radek




----- Original Message -----
From: "Jim Kinney" <***@gmail.com>
Sent: Wed, 7/1/2015 9:10pm
To: "Support list for open source software in schools." <***@redhat.com>
Subject: Re: [K12OSN] Centos7 : some ( official )help wanted

On Wed, 2015-07-01 at 17:24 +0200, ***@telenet.be wrote:

Van: "Warren Togami Jr." <***@togami.com>
Aan: "Support list for open source software in schools." <***@redhat.com>
Verzonden: Woensdag 1 juli 2015 17:09:42
Onderwerp: Re: [K12OSN] Centos7 : some ( official )help wanted

Disagree all you want. If no developer is willing to do the work, then it cannot happen.
On Tue, Jun 30, 2015 at 1:00 AM, <***@telenet.be> wrote:
Hello,

I have good results on our first site where we run K12Linux on Centos7 and I am about to ugrade a second, larger
site as well.

I just had a look at the K12Linux website again and just wanted to say I don't agree with this:

Modern users probably do not want to use this type of thin client because video (like Youtube) requires too much bandwidth over remote X desktops, and scalability of X over ssh encrypted tunnels is rather poor. K12Linux is considered a legacy solution for existing deployments of LTSP-type networks and is currently supported only on the legacy EPEL6 platform. Porting to more modern systemd-based Fedora and EPEL7 is technically possible but is not considered a priority given the drawbacks of the legacy LTSP solution.

I think a lot of organsisations have employees who 's first priority is not watching youtube, all though my users have no problems with that on K12Linux.
What hurts me is that because people can not get this to work on Centos7/Rhel7, they are moving to Ubuntu.

I have K12Linux/Centos6 running for about 3 years, and it's incredibly robust. And I think it's the same on Centos7.

greetings, Johan

Warren,

Then it should say on the site " no developer wants to do the work".
Now it says K12Linux is legacy.

And Warren, I am grateful for the work that you put in for K12Linux.

What I'm saying here, and I think what Radek is saying, is it's not legacy, It works.

Legacy doesn't mean it doesn't work. It simply means no further development will be done for that particular setup. Sometimes a program becomes legacy because no one is working on it. Sometimes because it get superseded by newer projects.
K12Linux on CentOS 6 is a great tool. But porting it to be fully CentOS 7 ready will require development from people other than Warren. K12Linux will be around for many more years since CentOS 6 will be around as well.
I think it's time to look at alternate methods of achieving a similar result. The PXE boot aspect will need to stay but the server platform needs to be revisited. The IT world has embraced virtual machines. Maybe a thin client should be running a VM client that provides console to a remote VM running on a server farm. The spice client is remarkably light and can push screen bits as well if not better than NX and RDP (and way faster than X over SSH or VNC). Using a management tool like Ovirt would allow tech support to build and maintain student machines as pool devices and thus only require updates to a single machine. Some coding is required to provide a single sign on screen that would connect to the next available VM console (which then auto-mounts the students home directory). The VM client (spice) can be 32 bit as well as 64 so it will run on older client hardware and it supports USB pass through for thumb drives. Sound also works.
3 years ago I was doing a demo talk of Ovirt and connected through 2 layers of VPNs back to my work lab to show both windows and Linux desktops. I started a browser and went to YouTube and the flash video played smoothly and with sound synched to the video.
As KVM, the technology behind Ovirt virtual machines, supports "memory ballooning", different VMs that need to use the same block of memory can share the block (read only - best for OS and application memory) and thus can perform similar types of resource sharing that K12Linux provided. Best of all, each user is truly isolated from all the others so an application crash (ahem - flash) will only affect the single user even with shared memory.

Greetings, Johan


_______________________________________________
K12OSN mailing list
***@redhat.com
https://www.redhat.com/mailman/listinfo/k12osn
For more info see <http://www.k12os.org>


_______________________________________________
K12OSN mailing list
***@redhat.com
https://www.redhat.com/mailman/listinfo/k12osn
For more info see <http://www.k12os.org>
_______________________________________________
K12OSN mailing list
***@redhat.com
https://www.redhat.com/mailman/listinfo/k12osn
For more info see <http://www.k12os.org>--
James P. Kinney III

Every time you stop a school, you will have to build a jail. What you
gain at one end you lose at the other. It's like feeding a dog on his
own tail. It won't fatten the dog.
- Speech 11/23/1900 Mark Twain

http://heretothereideas.blogspot.com/
_______________________________________________
K12OSN mailing list
***@redhat.com
https://www.redhat.com/mailman/listinfo/k12osn
For more info see <http://www.k12os.org>;
Jim Kinney
2015-07-02 11:03:19 UTC
Permalink
In order to develop K12Linux further, it will require developers. Warren
brought it from Centos 5 to Centos 6. He did this working alone. To migrate
it to Centos 7 will require new developers to step forward and take over. I
don't think Warren is interested. It was hoped that RedHat would pick up
K12linux. That doesn't seem to be an option any more as their focus is on
VM technology.

I can't take on this myself for many reasons. The main one is I see the
spice/VM process as a better successor. My day job has me swamped so my
time is zero.
Post by Radek Bursztynowski
p { margin-bottom: 0.25cm; line-height: 120%; }
Jim,
Thanks you for your
clear explanation. It means, that there are two separated cases. The
first one is the K12Linux's usefulness and the second one closing
down the project. In spite of K12Linux's usefulness this project is
shutting down. Is it appropriate understanding?
If yes, I regret. I
don't evaluate why, but I regret.
Are there any
chances to reanimate K12Linux project?
Best regards,
Radek
----- Original Message -----
Sent: Wed, 7/1/2015 9:10pm
Subject: Re: [K12OSN] Centos7 : some ( official )help wanted
Aan: "Support list for open source software in schools." <
Verzonden: Woensdag 1 juli 2015 17:09:42
Onderwerp: Re: [K12OSN] Centos7 : some ( official )help wanted
Disagree all you want. If no developer is willing to do the work, then it cannot happen.
Hello,
I have good results on our first site where we run K12Linux on Centos7 and
I am about to ugrade a second, larger
site as well.
I just had a look at the K12Linux website again and just wanted to say I
Modern users probably do not want to use this type of thin client because
video (like Youtube) requires too much bandwidth over remote X desktops,
and scalability of X over ssh encrypted tunnels is rather poor. K12Linux is
considered a legacy solution for existing deployments of LTSP-type networks
and is currently supported only on the legacy EPEL6 platform. Porting to
more modern systemd-based Fedora and EPEL7 is technically possible but is
not considered a priority given the drawbacks of the legacy LTSP solution.
I think a lot of organsisations have employees who 's first priority is
not watching youtube, all though my users have no problems with that on
K12Linux.
What hurts me is that because people can not get this to work on
Centos7/Rhel7, they are moving to Ubuntu.
I have K12Linux/Centos6 running for about 3 years, and it's incredibly
robust. And I think it's the same on Centos7.
greetings, Johan
Warren,
Then it should say on the site " no developer wants to do the work".
Now it says K12Linux is legacy.
And Warren, I am grateful for the work that you put in for K12Linux.
What I'm saying here, and I think what Radek is saying, is it's not legacy, It works.
Legacy doesn't mean it doesn't work. It simply means no further
development will be done for that particular setup. Sometimes a program
becomes legacy because no one is working on it. Sometimes because it get
superseded by newer projects.
K12Linux on CentOS 6 is a great tool. But porting it to be fully CentOS 7
ready will require development from people other than Warren. K12Linux will
be around for many more years since CentOS 6 will be around as well.
I think it's time to look at alternate methods of achieving a similar
result. The PXE boot aspect will need to stay but the server platform needs
to be revisited. The IT world has embraced virtual machines. Maybe a thin
client should be running a VM client that provides console to a remote VM
running on a server farm. The spice client is remarkably light and can push
screen bits as well if not better than NX and RDP (and way faster than X
over SSH or VNC). Using a management tool like Ovirt would allow tech
support to build and maintain student machines as pool devices and thus
only require updates to a single machine. Some coding is required to
provide a single sign on screen that would connect to the next available VM
console (which then auto-mounts the students home directory). The VM client
(spice) can be 32 bit as well as 64 so it will run on older client hardware
and it supports USB pass through for thumb drives. Sound also works.
3 years ago I was doing a demo talk of Ovirt and connected through 2
layers of VPNs back to my work lab to show both windows and Linux desktops.
I started a browser and went to YouTube and the flash video played smoothly
and with sound synched to the video.
As KVM, the technology behind Ovirt virtual machines, supports "memory
ballooning", different VMs that need to use the same block of memory can
share the block (read only - best for OS and application memory) and thus
can perform similar types of resource sharing that K12Linux provided. Best
of all, each user is truly isolated from all the others so an application
crash (ahem - flash) will only affect the single user even with shared
memory.
Greetings, Johan
_______________________________________________
K12OSN mailing list
https://www.redhat.com/mailman/listinfo/k12osn
For more info see <http://www.k12os.org>
_______________________________________________
K12OSN mailing list
https://www.redhat.com/mailman/listinfo/k12osn
For more info see <http://www.k12os.org>
_______________________________________________
K12OSN mailing list
https://www.redhat.com/mailman/listinfo/k12osn
For more info see <http://www.k12os.org>--
James P. Kinney III
Every time you stop a school, you will have to build a jail. What you
gain at one end you lose at the other. It's like feeding a dog on his
own tail. It won't fatten the dog.
- Speech 11/23/1900 Mark Twain
http://heretothereideas.blogspot.com/
_______________________________________________
K12OSN mailing list
https://www.redhat.com/mailman/listinfo/k12osn
For more info see <http://www.k12os.org>;
_______________________________________________
K12OSN mailing list
https://www.redhat.com/mailman/listinfo/k12osn
For more info see <http://www.k12os.org>
Radek Bursztynowski
2015-07-02 12:49:52 UTC
Permalink
Many thanks for information.

Does Ubuntu stilll develop LTSP? And what about SUSE-KIWI?

Best regards,
Radek

------

In order to develop K12Linux further, it will require developers. Warren brought it from Centos 5 to Centos 6. He did this working alone. To migrate it to Centos 7 will require new developers to step forward and take over. I don't think Warren is interested. It was hoped that RedHat would pick up K12linux. That doesn't seem to be an option any more as their focus is on VM technology.

I can't take on this myself for many reasons. The main one is I see the spice/VM process as a better successor. My day job has me swamped so my time is zero.

On Jul 2, 2015 3:20 AM, "Radek Bursztynowski" <***@bursztynowski.waw.pl> wrote:











p { margin-bottom: 0.25cm; line-height: 120%; }





Jim,





Thanks you for your

clear explanation. It means, that there are two separated cases. The

first one is the K12Linux's usefulness and the second one closing

down the project. In spite of K12Linux's usefulness this project is

shutting down. Is it appropriate understanding?





If yes, I regret. I

don't evaluate why, but I regret.





Are there any

chances to reanimate K12Linux project?





Best regards,



Radek









----- Original Message -----

From: "Jim Kinney" <***@gmail.com>

Sent: Wed, 7/1/2015 9:10pm

To: "Support list for open source software in schools." <***@redhat.com>

Subject: Re: [K12OSN] Centos7 : some ( official )help wanted



On Wed, 2015-07-01 at 17:24 +0200, ***@telenet.be wrote:



Van: "Warren Togami Jr." <***@togami.com>

Aan: "Support list for open source software in schools." <***@redhat.com>

Verzonden: Woensdag 1 juli 2015 17:09:42

Onderwerp: Re: [K12OSN] Centos7 : some ( official )help wanted



Disagree all you want. If no developer is willing to do the work, then it cannot happen.

On Tue, Jun 30, 2015 at 1:00 AM, <***@telenet.be> wrote:

Hello,



I have good results on our first site where we run K12Linux on Centos7 and I am about to ugrade a second, larger

site as well.



I just had a look at the K12Linux website again and just wanted to say I don't agree with this:



Modern users probably do not want to use this type of thin client because video (like Youtube) requires too much bandwidth over remote X desktops, and scalability of X over ssh encrypted tunnels is rather poor. K12Linux is considered a legacy solution for existing deployments of LTSP-type networks and is currently supported only on the legacy EPEL6 platform. Porting to more modern systemd-based Fedora and EPEL7 is technically possible but is not considered a priority given the drawbacks of the legacy LTSP solution.



I think a lot of organsisations have employees who 's first priority is not watching youtube, all though my users have no problems with that on K12Linux.

What hurts me is that because people can not get this to work on Centos7/Rhel7, they are moving to Ubuntu.



I have K12Linux/Centos6 running for about 3 years, and it's incredibly robust. And I think it's the same on Centos7.



greetings, Johan



Warren,



Then it should say on the site " no developer wants to do the work".

Now it says K12Linux is legacy.



And Warren, I am grateful for the work that you put in for K12Linux.



What I'm saying here, and I think what Radek is saying, is it's not legacy, It works.



Legacy doesn't mean it doesn't work. It simply means no further development will be done for that particular setup. Sometimes a program becomes legacy because no one is working on it. Sometimes because it get superseded by newer projects.

K12Linux on CentOS 6 is a great tool. But porting it to be fully CentOS 7 ready will require development from people other than Warren. K12Linux will be around for many more years since CentOS 6 will be around as well.

I think it's time to look at alternate methods of achieving a similar result. The PXE boot aspect will need to stay but the server platform needs to be revisited. The IT world has embraced virtual machines. Maybe a thin client should be running a VM client that provides console to a remote VM running on a server farm. The spice client is remarkably light and can push screen bits as well if not better than NX and RDP (and way faster than X over SSH or VNC). Using a management tool like Ovirt would allow tech support to build and maintain student machines as pool devices and thus only require updates to a single machine. Some coding is required to provide a single sign on screen that would connect to the next available VM console (which then auto-mounts the students home directory). The VM client (spice) can be 32 bit as well as 64 so it will run on older client hardware and it supports USB pass through for thumb drives. Sound also works.

3 years ago I was doing a demo talk of Ovirt and connected through 2 layers of VPNs back to my work lab to show both windows and Linux desktops. I started a browser and went to YouTube and the flash video played smoothly and with sound synched to the video.

As KVM, the technology behind Ovirt virtual machines, supports "memory ballooning", different VMs that need to use the same block of memory can share the block (read only - best for OS and application memory) and thus can perform similar types of resource sharing that K12Linux provided. Best of all, each user is truly isolated from all the others so an application crash (ahem - flash) will only affect the single user even with shared memory.



Greetings, Johan





_______________________________________________

K12OSN mailing list

***@redhat.com

https://www.redhat.com/mailman/listinfo/k12osn

For more info see <http://www.k12os.org>





_______________________________________________

K12OSN mailing list

***@redhat.com

https://www.redhat.com/mailman/listinfo/k12osn

For more info see <http://www.k12os.org>

_______________________________________________

K12OSN mailing list

***@redhat.com

https://www.redhat.com/mailman/listinfo/k12osn

For more info see <http://www.k12os.org>--

James P. Kinney III



Every time you stop a school, you will have to build a jail. What you

gain at one end you lose at the other. It's like feeding a dog on his

own tail. It won't fatten the dog.

- Speech 11/23/1900 Mark Twain



http://heretothereideas.blogspot.com/

_______________________________________________

K12OSN mailing list

***@redhat.com

https://www.redhat.com/mailman/listinfo/k12osn

For more info see <http://www.k12os.org>;



_______________________________________________

K12OSN mailing list

***@redhat.com

https://www.redhat.com/mailman/listinfo/k12osn

For more info see <http://www.k12os.org>


_______________________________________________
K12OSN mailing list
***@redhat.com
https://www.redhat.com/mailman/listinfo/k12osn
For more info see <http://www.k12os.org>;
Jim Kinney
2015-07-02 13:48:05 UTC
Permalink
I don't use either of those so I can't answer that.
Post by Radek Bursztynowski
Many thanks for information.
Does Ubuntu stilll develop LTSP? And what about SUSE-KIWI?
Best regards,
Radek
------
In order to develop K12Linux further, it will require developers. Warren
brought it from Centos 5 to Centos 6. He did this working alone. To migrate
it to Centos 7 will require new developers to step forward and take over. I
don't think Warren is interested. It was hoped that RedHat would pick up
K12linux. That doesn't seem to be an option any more as their focus is on
VM technology.
I can't take on this myself for many reasons. The main one is I see the
spice/VM process as a better successor. My day job has me swamped so my
time is zero.
p { margin-bottom: 0.25cm; line-height: 120%; }
Jim,
Thanks you for your
clear explanation. It means, that there are two separated cases. The
first one is the K12Linux's usefulness and the second one closing
down the project. In spite of K12Linux's usefulness this project is
shutting down. Is it appropriate understanding?
If yes, I regret. I
don't evaluate why, but I regret.
Are there any
chances to reanimate K12Linux project?
Best regards,
Radek
----- Original Message -----
Sent: Wed, 7/1/2015 9:10pm
Subject: Re: [K12OSN] Centos7 : some ( official )help wanted
Aan: "Support list for open source software in schools." <
Verzonden: Woensdag 1 juli 2015 17:09:42
Onderwerp: Re: [K12OSN] Centos7 : some ( official )help wanted
Disagree all you want. If no developer is willing to do the work, then it cannot happen.
Hello,
I have good results on our first site where we run K12Linux on Centos7 and
I am about to ugrade a second, larger
site as well.
I just had a look at the K12Linux website again and just wanted to say I
Modern users probably do not want to use this type of thin client because
video (like Youtube) requires too much bandwidth over remote X desktops,
and scalability of X over ssh encrypted tunnels is rather poor. K12Linux is
considered a legacy solution for existing deployments of LTSP-type networks
and is currently supported only on the legacy EPEL6 platform. Porting to
more modern systemd-based Fedora and EPEL7 is technically possible but is
not considered a priority given the drawbacks of the legacy LTSP solution.
I think a lot of organsisations have employees who 's first priority is
not watching youtube, all though my users have no problems with that on
K12Linux.
What hurts me is that because people can not get this to work on
Centos7/Rhel7, they are moving to Ubuntu.
I have K12Linux/Centos6 running for about 3 years, and it's incredibly
robust. And I think it's the same on Centos7.
greetings, Johan
Warren,
Then it should say on the site " no developer wants to do the work".
Now it says K12Linux is legacy.
And Warren, I am grateful for the work that you put in for K12Linux.
What I'm saying here, and I think what Radek is saying, is it's not legacy, It works.
Legacy doesn't mean it doesn't work. It simply means no further
development will be done for that particular setup. Sometimes a program
becomes legacy because no one is working on it. Sometimes because it get
superseded by newer projects.
K12Linux on CentOS 6 is a great tool. But porting it to be fully CentOS 7
ready will require development from people other than Warren. K12Linux will
be around for many more years since CentOS 6 will be around as well.
I think it's time to look at alternate methods of achieving a similar
result. The PXE boot aspect will need to stay but the server platform needs
to be revisited. The IT world has embraced virtual machines. Maybe a thin
client should be running a VM client that provides console to a remote VM
running on a server farm. The spice client is remarkably light and can push
screen bits as well if not better than NX and RDP (and way faster than X
over SSH or VNC). Using a management tool like Ovirt would allow tech
support to build and maintain student machines as pool devices and thus
only require updates to a single machine. Some coding is required to
provide a single sign on screen that would connect to the next available VM
console (which then auto-mounts the students home directory). The VM client
(spice) can be 32 bit as well as 64 so it will run on older client hardware
and it supports USB pass through for thumb drives. Sound also works.
3 years ago I was doing a demo talk of Ovirt and connected through 2
layers of VPNs back to my work lab to show both windows and Linux desktops.
I started a browser and went to YouTube and the flash video played smoothly
and with sound synched to the video.
As KVM, the technology behind Ovirt virtual machines, supports "memory
ballooning", different VMs that need to use the same block of memory can
share the block (read only - best for OS and application memory) and thus
can perform similar types of resource sharing that K12Linux provided. Best
of all, each user is truly isolated from all the others so an application
crash (ahem - flash) will only affect the single user even with shared
memory.
Greetings, Johan
_______________________________________________
K12OSN mailing list
https://www.redhat.com/mailman/listinfo/k12osn
For more info see <http://www.k12os.org>
_______________________________________________
K12OSN mailing list
https://www.redhat.com/mailman/listinfo/k12osn
For more info see <http://www.k12os.org>
_______________________________________________
K12OSN mailing list
https://www.redhat.com/mailman/listinfo/k12osn
For more info see <http://www.k12os.org>--
James P. Kinney III
Every time you stop a school, you will have to build a jail. What you
gain at one end you lose at the other. It's like feeding a dog on his
own tail. It won't fatten the dog.
- Speech 11/23/1900 Mark Twain
http://heretothereideas.blogspot.com/
_______________________________________________
K12OSN mailing list
https://www.redhat.com/mailman/listinfo/k12osn
For more info see <http://www.k12os.org>;
_______________________________________________
K12OSN mailing list
https://www.redhat.com/mailman/listinfo/k12osn
For more info see <http://www.k12os.org>
_______________________________________________
K12OSN mailing list
https://www.redhat.com/mailman/listinfo/k12osn
For more info see <http://www.k12os.org>;
Loading...