First Post and some questions

classic Classic list List threaded Threaded
25 messages Options
12
Reply | Threaded
Open this post in threaded view
|

First Post and some questions

Jan Kandziora
Dear All,

I'm currently developing an automation project - a semiautomatic vending
machine with a lot of devices attached remotely - in which I plan to use
1-Wire chips as interface to "the real world".

I'm glad some people already developed something useful for 1-Wire and Linux.
I have already played a little with owfs and the sample chips I have here, but
now first questions arise:

* The host platform will be MIPS (AMD Alchemy), not i386. May there be any
problems with that e.g. Endian-issues?
* The machine will have more than one host. I understood I can use "owserver"
on the "slave" hosts (modified X-Terminals) to connect 1-Wire devices there
and access them via TCP/IP. Correct?
* How is the timing in this configuration? About 0.5 second granularity in
0-1-0 switches would be nice although 1 second is tolerable, too.
* What's up with the Tcl binding? The ow.so file seems to be missing - at
least, I could not make sense of the SWIG guided build process - know enough
of SWIG to avoid it, though.

I have knowledge in near-hardware programming in C and a little on the Tcl/C
interface to address problems myself if there should arise any. Mind if I
join the development team? However, I need someone to guide me - a project
one didn't start self is totally confusing first.

Kind regards

        Jan


-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r=20
_______________________________________________
Owfs-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/owfs-developers
Reply | Threaded
Open this post in threaded view
|

RE: First Post and some questions

Alfille, Paul H.,M.D.
Jan,

You are absolutely welcome to participate.

To answer your questions:

1. MIPS works. Christian Magnusson is the wizard at porting.

2. Slaves using owserver is the intended design.
We don't currently implement adding/removing owservers sources while the owfs
(or whatever) process is running, but there is nothing in the underlying design
that would prevent it. We'd just have to ponder what the communication channel
to the master should be for these events.

3. Timing is really 1-wire bus limited. OWFS is straight C code, and TCP/IP is
small single packets. I'll test the timing if you wish -- what devices where you
considering?

4. TCL doesn't use SWIG, I believe.
Serg Oskin is the TCL implementer, and has been very active is fixing any
problems.

Paul Alfille

-----Original Message-----
From: [hidden email]
[mailto:[hidden email]]On Behalf Of Jan
Kandziora
Sent: Wednesday, June 08, 2005 4:54 PM
To: [hidden email]
Subject: [Owfs-developers] First Post and some questions


Dear All,

I'm currently developing an automation project - a semiautomatic vending
machine with a lot of devices attached remotely - in which I plan to use
1-Wire chips as interface to "the real world".

I'm glad some people already developed something useful for 1-Wire and Linux.
I have already played a little with owfs and the sample chips I have here, but
now first questions arise:

* The host platform will be MIPS (AMD Alchemy), not i386. May there be any
problems with that e.g. Endian-issues?
* The machine will have more than one host. I understood I can use "owserver"
on the "slave" hosts (modified X-Terminals) to connect 1-Wire devices there
and access them via TCP/IP. Correct?
* How is the timing in this configuration? About 0.5 second granularity in
0-1-0 switches would be nice although 1 second is tolerable, too.
* What's up with the Tcl binding? The ow.so file seems to be missing - at
least, I could not make sense of the SWIG guided build process - know enough
of SWIG to avoid it, though.

I have knowledge in near-hardware programming in C and a little on the Tcl/C
interface to address problems myself if there should arise any. Mind if I
join the development team? However, I need someone to guide me - a project
one didn't start self is totally confusing first.

Kind regards

        Jan


-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r=20
_______________________________________________
Owfs-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/owfs-developers


-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r 
_______________________________________________
Owfs-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/owfs-developers
Reply | Threaded
Open this post in threaded view
|

Re: First Post and some questions

Serg Oskin
In reply to this post by Jan Kandziora
"JK" == Jan Kandziora wrote:

 JK> * What's up with the Tcl binding? The ow.so file seems to be missing - at
 JK> least, I could not make sense of the SWIG guided build process - know enough
 JK> of SWIG to avoid it, though.

owtcl does not compiled by default, run configure or bootsrap with --enable-owtcl

--
        Serg (http://oskin.ru/)
~
~
:q!


-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r=20
_______________________________________________
Owfs-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/owfs-developers
Reply | Threaded
Open this post in threaded view
|

Re: First Post and some questions

Jan Kandziora
Am Donnerstag, 9. Juni 2005 07:55 schrieb Serg Oskin:
> "JK" == Jan Kandziora wrote:
>
>  JK> * What's up with the Tcl binding? The ow.so file seems to be missing -
> at JK> least, I could not make sense of the SWIG guided build process -
> know enough JK> of SWIG to avoid it, though.
>
> owtcl does not compiled by default, run configure or bootsrap with
> --enable-owtcl
>
Yes, I already enabled that switch (tested on the nightly build). However, all
files I get are /usr/lib/owtcl-0.1/
drwxr-xr-x    2 root root    200 2005-06-09 09:28 ./
drwxr-xr-x  166 root root  55056 2005-06-09 09:28 ../
lrwxrwxrwx    1 root root      6 2005-06-09 09:28 ow -> ow-0.1*
-rwxr-xr-x    1 root root 828597 2005-06-09 09:28 ow-0.1*
-rw-r--r--    1 root root  27830 2005-06-09 09:28 ow.a
-rwxr-xr-x    1 root root    904 2005-06-09 09:28 ow.la*
-rw-r--r--    1 root root    273 2005-06-09 09:28 ow.tcl
-rw-r--r--    1 root root    108 2005-06-09 09:28 pkgIndex.tcl


With the owfs-2.1.0-1.i386.rpm package, I got
drwxr-xr-x    2 root root    200 2005-06-09 09:11 ./
drwxr-xr-x  166 root root  55056 2005-06-09 09:11 ../
lrwxrwxrwx    1 root root      6 2005-06-09 09:11 ow -> ow-0.1*
-rwxr-xr-x    1 root root 248584 2005-04-07 04:25 ow-0.1*
-rw-r--r--    1 root root  27822 2005-04-07 04:25 ow.a
-rwxr-xr-x    1 root root    857 2005-04-07 04:25 ow.la*
-rw-r--r--    1 root root    273 2005-04-07 04:25 ow.tcl
-rw-r--r--    1 root root    108 2005-04-07 04:25 pkgIndex.tcl

$ tclsh
% package require ow
couldn't load file "/usr/lib/owtcl-0.1/ow.so": /usr/lib/owtcl-0.1/ow.so:
cannot open shared object file: No such file or directory

Hm. What to try next?

        Jan
--
43rd Law of Computing:
        Anything that can go wr
fortune: Segmentation violation -- Core dumped



-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r=20
_______________________________________________
Owfs-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/owfs-developers
Reply | Threaded
Open this post in threaded view
|

Re: First Post and some questions

Jan Kandziora
In reply to this post by Alfille, Paul H.,M.D.
Am Mittwoch, 8. Juni 2005 23:12 schrieb Alfille, Paul H.,M.D.:
>
> You are absolutely welcome to participate.
>
Then I'm happy to go on the bug-hunt.

Anyway, for my project I need an automation framework (not a GUI, just a
library) for 1-wire devices and Linux, with some things owfs itself doesn't
address, e.g. timers or grouping of 1-wire chips to "devices" - the goal here
is to uniquely identify "devices" instead of single chips and thus, allow
plug'n'work. I want to develop this in Tcl (that's why I'm after the
binding).

I don't know if you are interested in such a framework - it's a little off the
scope.


>
> 1. MIPS works. Christian Magnusson is the wizard at porting.
>
Great. I'll try out as soon the hardware arrives.


> 2. Slaves using owserver is the intended design.
> We don't currently implement adding/removing owservers sources while the
> owfs (or whatever) process is running, but there is nothing in the
> underlying design that would prevent it. We'd just have to ponder what the
> communication channel to the master should be for these events.
>
I'd expect the owtcl client to throw an error if the server is not reachable.
Am I correct? In that case, the automation framework could try to reconnect
every 10s or the like.



> 3. Timing is really 1-wire bus limited. OWFS is straight C code, and TCP/IP
> is small single packets. I'll test the timing if you wish -- what devices
> where you considering?
>
^^; I thought you already have that data, as this is the first question I'd
expect about any automation system. But Ok, I can measure it myself, too, no
hassle.


> 4. TCL doesn't use SWIG, I believe.
> Serg Oskin is the TCL implementer, and has been very active is fixing any
> problems.
>
Ok, I will ask him on Tcl issues. The first thing on my nails is:


BUGS
       Simultaneous work of several connections is not supported.


I really need multiple simultaneous connections for my application - who
doesn't? Is this a limitation of owlib (global data structures) or simply a
limitation in the owtcl wrapper? I would start with this bug if someone
guides me.

Kind regards

        Jan


-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r=20
_______________________________________________
Owfs-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/owfs-developers
Reply | Threaded
Open this post in threaded view
|

Re: First Post and some questions

Paul Alfille
On Thursday 09 June 2005 04:33 am, Jan Kandziora wrote:

> Am Mittwoch, 8. Juni 2005 23:12 schrieb Alfille, Paul H.,M.D.:
> > You are absolutely welcome to participate.
>
> Then I'm happy to go on the bug-hunt.
>
> Anyway, for my project I need an automation framework (not a GUI, just a
> library) for 1-wire devices and Linux, with some things owfs itself doesn't
> address, e.g. timers or grouping of 1-wire chips to "devices" - the goal
> here is to uniquely identify "devices" instead of single chips and thus,
> allow plug'n'work. I want to develop this in Tcl (that's why I'm after the
> binding).
>
> I don't know if you are interested in such a framework - it's a little off
> the scope.
>

You are addressing a more general problem, aggregates of 1-wire devices.
There are a few known sensors, like the pressure sensor, that use more than
one device to make a larger circuit.

The problem is that unless there is a standard way (perhaps in device ROM) for
the devices to autoorganize, you need outside information.

This is nothing new, you need outside information to know which device is the
rooftop temperature vs the server temp.

Most approaches have been to add a layer on top of OWFS -- perhaps by
filesystem links to better filenames, or a database.

If you have a more general method or design, I'd be interested.

> > 1. MIPS works. Christian Magnusson is the wizard at porting.
>
> Great. I'll try out as soon the hardware arrives.
>
> > 2. Slaves using owserver is the intended design.
> > We don't currently implement adding/removing owservers sources while the
> > owfs (or whatever) process is running, but there is nothing in the
> > underlying design that would prevent it. We'd just have to ponder what
> > the communication channel to the master should be for these events.
>
> I'd expect the owtcl client to throw an error if the server is not
> reachable. Am I correct? In that case, the automation framework could try
> to reconnect every 10s or the like.
>
> > 3. Timing is really 1-wire bus limited. OWFS is straight C code, and
> > TCP/IP is small single packets. I'll test the timing if you wish -- what
> > devices where you considering?
>
> ^^; I thought you already have that data, as this is the first question I'd
> expect about any automation system. But Ok, I can measure it myself, too,
> no hassle.
>
> > 4. TCL doesn't use SWIG, I believe.
> > Serg Oskin is the TCL implementer, and has been very active is fixing any
> > problems.
>
> Ok, I will ask him on Tcl issues. The first thing on my nails is:
>
>
> BUGS
>        Simultaneous work of several connections is not supported.
>
You can certainly aggregate several owservers. All the packages use the same
underlying owlib that has multiple data sources built in. The only question
is whether you can pass all their names on the "command line" -- the invoking
call. It is probably better design to "aggregate" to a local owserver, and
then use that owserver for all the TCL calls. That allows more than one
process to talk to the devices, and probably reduces startup times.

>
> I really need multiple simultaneous connections for my application - who
> doesn't? Is this a limitation of owlib (global data structures) or simply a
> limitation in the owtcl wrapper? I would start with this bug if someone
> guides me.
>
> Kind regards
>
> Jan
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you
> shotput a projector? How fast can you ride your desk chair down the office
> luge track? If you want to score the big prize, get to know the little guy.
> Play to win an NEC 61" plasma display: http://www.necitguy.com/?r=20
> _______________________________________________
> Owfs-developers mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/owfs-developers


-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r=20
_______________________________________________
Owfs-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/owfs-developers
Reply | Threaded
Open this post in threaded view
|

Re: First Post and some questions

Jan Kandziora
Am Donnerstag, 9. Juni 2005 11:39 schrieb Paul Alfille:

> >
> > Anyway, for my project I need an automation framework (not a GUI, just a
> > library) for 1-wire devices and Linux, with some things owfs itself
> > doesn't address, e.g. timers or grouping of 1-wire chips to "devices" -
> > the goal here is to uniquely identify "devices" instead of single chips
> > and thus, allow plug'n'work. I want to develop this in Tcl (that's why
> > I'm after the binding).
> >
> > I don't know if you are interested in such a framework - it's a little
> > off the scope.
>
> You are addressing a more general problem, aggregates of 1-wire devices.
> There are a few known sensors, like the pressure sensor, that use more than
> one device to make a larger circuit.
>
> The problem is that unless there is a standard way (perhaps in device ROM)
> for the devices to autoorganize, you need outside information.
>
> This is nothing new, you need outside information to know which device is
> the rooftop temperature vs the server temp.
>
> Most approaches have been to add a layer on top of OWFS -- perhaps by
> filesystem links to better filenames, or a database.
>
> If you have a more general method or design, I'd be interested.
>
I think there are two ways to handle aggregates. The first involves a DS2409
network coupler like described in the Dallas/Maxim application note for that
chip. I find this expensive (both monetary and design) and clumsy, too, as
you can seperate the aggregates from each other, but still don't have a clue
which one is where.

The only way to get rid of this problem is the database approach you
mentioned. After the aggregates are put together, connect them to a 1-wire
bus solely and maintain a database of the chip's IDs. Put the aggregate's ID
into the database and on an adhesive label on the PCB.

Later, the personnel which does the installation can look up the aggregate's
ID from the labels and make the correct connections in the software.

There is one problem left which cannot be adressed this way: Multiple chips of
the same type in one aggregate - usually, you have no control which single
chip is mounted at which position on a PCB. The most simple solution is to
sacrifice a port pin which is connected to 5V on one chip, 0V on the other.
Another possible solution is again the DS2409 to get at least 3 identifiable
branches per device.


> >
> > > 4. TCL doesn't use SWIG, I believe.
> > > Serg Oskin is the TCL implementer, and has been very active is fixing
> > > any problems.
> >
> > Ok, I will ask him on Tcl issues. The first thing on my nails is:
> >
> >
> > BUGS
> >        Simultaneous work of several connections is not supported.
>
> You can certainly aggregate several owservers. All the packages use the
> same underlying owlib that has multiple data sources built in. The only
> question is whether you can pass all their names on the "command line" --
> the invoking call. It is probably better design to "aggregate" to a local
> owserver, and then use that owserver for all the TCL calls. That allows
> more than one process to talk to the devices, and probably reduces startup
> times.
>
If I use it that way, I would have have no control about the data passed via
TCP. What happens if a local owserver cannot connect to a remote one? Does it
try again? This is really a knockout criteria, as the device should be
"self-healing" to the greatest extent possible.

        Jan



-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r=20
_______________________________________________
Owfs-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/owfs-developers
Reply | Threaded
Open this post in threaded view
|

RE: First Post and some questions

Alfille, Paul H.,M.D.
In reply to this post by Jan Kandziora
What would you like it to do?
We have several choices:
1. Report (after a timeout) as no devices present on that owserver, and keep
reporting that until reconnected.
2. Have a separate means of testing status and response. Adding a "present" flag
would have nice homology with the 1-wire device structure.
3. Error and die.

I'll have to test, but I think we do number 1.

Paul Alfille

------------------
>
> You can certainly aggregate several owservers. All the packages use the
> same underlying owlib that has multiple data sources built in. The only
> question is whether you can pass all their names on the "command line" --
> the invoking call. It is probably better design to "aggregate" to a local
> owserver, and then use that owserver for all the TCL calls. That allows
> more than one process to talk to the devices, and probably reduces startup
> times.
>
If I use it that way, I would have have no control about the data passed via
TCP. What happens if a local owserver cannot connect to a remote one? Does it
try again? This is really a knockout criteria, as the device should be
"self-healing" to the greatest extent possible.

        Jan



-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r 
_______________________________________________
Owfs-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/owfs-developers
Reply | Threaded
Open this post in threaded view
|

Re: First Post and some questions

Serg Oskin
In reply to this post by Jan Kandziora
"JK" == Jan Kandziora wrote:

 JK> Am Donnerstag, 9. Juni 2005 07:55 schrieb Serg Oskin:
 >> "JK" == Jan Kandziora wrote:
 >>
 JK> * What's up with the Tcl binding? The ow.so file seems to be missing -
 >> at JK> least, I could not make sense of the SWIG guided build process -
 >> know enough JK> of SWIG to avoid it, though.
 >>
 >> owtcl does not compiled by default, run configure or bootsrap with
 >> --enable-owtcl
 >>
 JK> Yes, I already enabled that switch (tested on the nightly build). However, all
 JK> files I get are /usr/lib/owtcl-0.1/
 JK> drwxr-xr-x    2 root root    200 2005-06-09 09:28 ./
 JK> drwxr-xr-x  166 root root  55056 2005-06-09 09:28 ../
 JK> lrwxrwxrwx    1 root root      6 2005-06-09 09:28 ow -> ow-0.1*
 JK> -rwxr-xr-x    1 root root 828597 2005-06-09 09:28 ow-0.1*
 JK> -rw-r--r--    1 root root  27830 2005-06-09 09:28 ow.a
 JK> -rwxr-xr-x    1 root root    904 2005-06-09 09:28 ow.la*
 JK> -rw-r--r--    1 root root    273 2005-06-09 09:28 ow.tcl
 JK> -rw-r--r--    1 root root    108 2005-06-09 09:28 pkgIndex.tcl


 JK> With the owfs-2.1.0-1.i386.rpm package, I got
 JK> drwxr-xr-x    2 root root    200 2005-06-09 09:11 ./
 JK> drwxr-xr-x  166 root root  55056 2005-06-09 09:11 ../
 JK> lrwxrwxrwx    1 root root      6 2005-06-09 09:11 ow -> ow-0.1*
 JK> -rwxr-xr-x    1 root root 248584 2005-04-07 04:25 ow-0.1*
 JK> -rw-r--r--    1 root root  27822 2005-04-07 04:25 ow.a
 JK> -rwxr-xr-x    1 root root    857 2005-04-07 04:25 ow.la*
 JK> -rw-r--r--    1 root root    273 2005-04-07 04:25 ow.tcl
 JK> -rw-r--r--    1 root root    108 2005-04-07 04:25 pkgIndex.tcl

 JK> $ tclsh
 JK> % package require ow
 JK> couldn't load file "/usr/lib/owtcl-0.1/ow.so": /usr/lib/owtcl-0.1/ow.so:
 JK> cannot open shared object file: No such file or directory

 JK> Hm. What to try next?

Don't run bootstrap, try
aclocal && autoheader && autoconf && libtoolize -f -c && automake -a -c && ./configure --enable-tcl ....

To Paul: https://sourceforge.net/mailarchive/message.php?msg_id=10764821 https://sourceforge.net/mailarchive/forum.php?thread_id=6320202&forum_id=34644

--
        Serg (http://oskin.ru/)
~
~
:q!


-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r=20
_______________________________________________
Owfs-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/owfs-developers
Reply | Threaded
Open this post in threaded view
|

Re: First Post and some questions

Serg Oskin
In reply to this post by Jan Kandziora
"JK" == Jan Kandziora wrote:

 JK> Am Mittwoch, 8. Juni 2005 23:12 schrieb Alfille, Paul H.,M.D.:
 >>
 >> You are absolutely welcome to participate.
 >>
 JK> Then I'm happy to go on the bug-hunt.

 JK> Anyway, for my project I need an automation framework (not a GUI, just a
 JK> library) for 1-wire devices and Linux, with some things owfs itself doesn't
 JK> address, e.g. timers or grouping of 1-wire chips to "devices" - the goal here
 JK> is to uniquely identify "devices" instead of single chips and thus, allow
 JK> plug'n'work. I want to develop this in Tcl (that's why I'm after the
 JK> binding).

 JK> I don't know if you are interested in such a framework - it's a little off the
 JK> scope.

It should support your application, but not OWFS.
Look here it as an example: ftp://oskin.ru/pub/linux/1wire/
It was written before creation owtcl.

 >>
 >> 1. MIPS works. Christian Magnusson is the wizard at porting.
 >>
 JK> Great. I'll try out as soon the hardware arrives.


 >> 2. Slaves using owserver is the intended design.
 >> We don't currently implement adding/removing owservers sources while the
 >> owfs (or whatever) process is running, but there is nothing in the
 >> underlying design that would prevent it. We'd just have to ponder what the
 >> communication channel to the master should be for these events.
 >>
 JK> I'd expect the owtcl client to throw an error if the server is not reachable.
 JK> Am I correct? In that case, the automation framework could try to reconnect
 JK> every 10s or the like.

owtcl transfers almost all mistakes OWFS in to your application. For
reliability I recommend to check presence of any 1wire chip's on the
bus (see my example).

 >> 3. Timing is really 1-wire bus limited. OWFS is straight C code, and TCP/IP
 >> is small single packets. I'll test the timing if you wish -- what devices
 >> where you considering?
 >>
 JK> ^^; I thought you already have that data, as this is the first question I'd
 JK> expect about any automation system. But Ok, I can measure it myself, too, no
 JK> hassle.


 >> 4. TCL doesn't use SWIG, I believe.
 >> Serg Oskin is the TCL implementer, and has been very active is fixing any
 >> problems.
 >>
 JK> Ok, I will ask him on Tcl issues. The first thing on my nails is:


 JK> BUGS
 JK> Simultaneous work of several connections is not supported.


 JK> I really need multiple simultaneous connections for my application - who
 JK> doesn't? Is this a limitation of owlib (global data structures) or simply a
 JK> limitation in the owtcl wrapper? I would start with this bug if someone
 JK> guides me.

Sorry, this information has become outdated. Fixed.

--
        Serg (http://oskin.ru/)
~
~
:q!


-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r=20
_______________________________________________
Owfs-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/owfs-developers
Reply | Threaded
Open this post in threaded view
|

Re: First Post and some questions

Scott Walde
In reply to this post by Jan Kandziora
Jan Kandziora wrote:

> Am Donnerstag, 9. Juni 2005 11:39 schrieb Paul Alfille:

>>You are addressing a more general problem, aggregates of 1-wire devices.
>>There are a few known sensors, like the pressure sensor, that use more than
>>one device to make a larger circuit.

> I think there are two ways to handle aggregates. The first involves a DS2409

Hi.  I've been pretty quiet on this list so far, so I feel a bit awkward
jumping in right here.  Anyways, I have another suggestion for
identifying aggregates.  You can put an EPROM device such as the DS2505
into each aggregate device and place information on that EPROM
describing the components on that device.  The device could contain a
magic code to indicate it belongs to your system, then a device
identifier to indicate which type of aggregate device it is (assuming
you have more than one type) and finally a list of addresses of the
individual devices that make up the aggregate device.  The address of
the DS2505 could be used to uniquely identify each aggregate device.

In fact, if this became a frequent request, it might be possible to add
a pluggable architecture into OWFS to handle these aggregates
automagically when a DS2505 is detected with the magic word -- but I'm
probably getting way ahead on this now.

ttyl
srw


-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r=20
_______________________________________________
Owfs-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/owfs-developers
Reply | Threaded
Open this post in threaded view
|

RE: First Post and some questions

Alfille, Paul H.,M.D.
In reply to this post by Jan Kandziora
Absolutely right idea. I thnk the DS2438 is the prefered configuration chip (It
has many functions).

Paul

-----Original Message-----
From: [hidden email]
[mailto:[hidden email]]On Behalf Of Scott
Walde
Sent: Thursday, June 09, 2005 10:08 AM
To: [hidden email]
Subject: Re: [Owfs-developers] First Post and some questions


Jan Kandziora wrote:

> Am Donnerstag, 9. Juni 2005 11:39 schrieb Paul Alfille:

>>You are addressing a more general problem, aggregates of 1-wire devices.
>>There are a few known sensors, like the pressure sensor, that use more than
>>one device to make a larger circuit.

> I think there are two ways to handle aggregates. The first involves a DS2409

Hi.  I've been pretty quiet on this list so far, so I feel a bit awkward
jumping in right here.  Anyways, I have another suggestion for
identifying aggregates.  You can put an EPROM device such as the DS2505
into each aggregate device and place information on that EPROM
describing the components on that device.  The device could contain a
magic code to indicate it belongs to your system, then a device
identifier to indicate which type of aggregate device it is (assuming
you have more than one type) and finally a list of addresses of the
individual devices that make up the aggregate device.  The address of
the DS2505 could be used to uniquely identify each aggregate device.

In fact, if this became a frequent request, it might be possible to add
a pluggable architecture into OWFS to handle these aggregates
automagically when a DS2505 is detected with the magic word -- but I'm
probably getting way ahead on this now.

ttyl
srw


-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r=20
_______________________________________________
Owfs-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/owfs-developers


-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r 
_______________________________________________
Owfs-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/owfs-developers
Reply | Threaded
Open this post in threaded view
|

Re: First Post and some questions

Jan Kandziora
In reply to this post by Scott Walde
Am Donnerstag, 9. Juni 2005 16:08 schrieb Scott Walde:

> Jan Kandziora wrote:
> > Am Donnerstag, 9. Juni 2005 11:39 schrieb Paul Alfille:
> >>You are addressing a more general problem, aggregates of 1-wire devices.
> >>There are a few known sensors, like the pressure sensor, that use more
> >> than one device to make a larger circuit.
> >
> > I think there are two ways to handle aggregates. The first involves a
> > DS2409
>
> Hi.  I've been pretty quiet on this list so far, so I feel a bit awkward
> jumping in right here.  Anyways, I have another suggestion for
> identifying aggregates.  You can put an EPROM device such as the DS2505
> into each aggregate device and place information on that EPROM
> describing the components on that device.
>
Hm. I don't see a big benefit over the unique IDs all devices have. You still
have to maintain a database of serial numbers. The crucial part is to get the
same label *into* the hardware and *onto* the hardware. I does not matter
which data is placed on the label, it just has to be unique.

Kind regards

        Jan



-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r=20
_______________________________________________
Owfs-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/owfs-developers
Reply | Threaded
Open this post in threaded view
|

RE: First Post and some questions

Alfille, Paul H.,M.D.
In reply to this post by Jan Kandziora
Tested it out:
If an owserver process fails, The system reports no content on that bus entry,
until it is restored. Probably exactly what you want.

Will you know all the owservers at initiation time, or will you need to add more
as your system functions (hotplugging)?

Paul

-----Original Message-----
From: [hidden email]
[mailto:[hidden email]]On Behalf Of
Alfille, Paul H.,M.D.
Sent: Thursday, June 09, 2005 6:47 AM
To: [hidden email]
Subject: RE: [Owfs-developers] First Post and some questions


What would you like it to do?
We have several choices:
1. Report (after a timeout) as no devices present on that owserver, and keep
reporting that until reconnected.
2. Have a separate means of testing status and response. Adding a "present" flag
would have nice homology with the 1-wire device structure.
3. Error and die.

I'll have to test, but I think we do number 1.

Paul Alfille

------------------
>
> You can certainly aggregate several owservers. All the packages use the
> same underlying owlib that has multiple data sources built in. The only
> question is whether you can pass all their names on the "command line" --
> the invoking call. It is probably better design to "aggregate" to a local
> owserver, and then use that owserver for all the TCL calls. That allows
> more than one process to talk to the devices, and probably reduces startup
> times.
>
If I use it that way, I would have have no control about the data passed via
TCP. What happens if a local owserver cannot connect to a remote one? Does it
try again? This is really a knockout criteria, as the device should be
"self-healing" to the greatest extent possible.

        Jan



-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r 
_______________________________________________
Owfs-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/owfs-developers
Reply | Threaded
Open this post in threaded view
|

Re: First Post and some questions

Jan Kandziora
In reply to this post by Alfille, Paul H.,M.D.
Am Donnerstag, 9. Juni 2005 12:47 schrieb Alfille, Paul H.,M.D.:

> What would you like it to do?
> We have several choices:
> 1. Report (after a timeout) as no devices present on that owserver, and
> keep reporting that until reconnected.
> 2. Have a separate means of testing status and response. Adding a "present"
> flag would have nice homology with the 1-wire device structure.
> 3. Error and die.
>
> I'll have to test, but I think we do number 1.
>
I think so, too.

How does owserver reacts if the serial/usb etc. 1-wire adapter is pulled and
reconnected later? As far as I checked, owserver does not recognize the
reconnect. How to solve this?

Kind regards

        Jan



-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r=20
_______________________________________________
Owfs-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/owfs-developers
Reply | Threaded
Open this post in threaded view
|

Re: First Post and some questions

Jan Kandziora
In reply to this post by Serg Oskin
Am Donnerstag, 9. Juni 2005 14:06 schrieb Serg Oskin:

> "JK" == Jan Kandziora wrote:
>
>  JK> Am Mittwoch, 8. Juni 2005 23:12 schrieb Alfille, Paul H.,M.D.:
>  >> You are absolutely welcome to participate.
>
>  JK> Then I'm happy to go on the bug-hunt.
>
>  JK> Anyway, for my project I need an automation framework (not a GUI, just
> a JK> library) for 1-wire devices and Linux, with some things owfs itself
> doesn't JK> address, e.g. timers or grouping of 1-wire chips to "devices" -
> the goal here JK> is to uniquely identify "devices" instead of single chips
> and thus, allow JK> plug'n'work. I want to develop this in Tcl (that's why
> I'm after the JK> binding).
>
>  JK> I don't know if you are interested in such a framework - it's a little
> off the JK> scope.
>
> It should support your application, but not OWFS.
> Look here it as an example: ftp://oskin.ru/pub/linux/1wire/
> It was written before creation owtcl.
>
I will take a look.


>
>  JK> I'd expect the owtcl client to throw an error if the server is not
> reachable. JK> Am I correct? In that case, the automation framework could
> try to reconnect JK> every 10s or the like.
>
> owtcl transfers almost all mistakes OWFS in to your application. For
> reliability I recommend to check presence of any 1wire chip's on the
> bus (see my example).
>
Yes, that's a must anyway.


>
>
>  JK> BUGS
>  JK> Simultaneous work of several connections is not supported.
>
>
>  JK> I really need multiple simultaneous connections for my application -
> who JK> doesn't? Is this a limitation of owlib (global data structures) or
> simply a JK> limitation in the owtcl wrapper? I would start with this bug
> if someone JK> guides me.
>
> Sorry, this information has become outdated. Fixed.
>
Wow, that was a fast fix. ^_^
Or do you mean, by using owserver?

Kind regards

        Jan



-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r=20
_______________________________________________
Owfs-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/owfs-developers
Reply | Threaded
Open this post in threaded view
|

Re: First Post and some questions

Scott Walde
In reply to this post by Alfille, Paul H.,M.D.
Alfille, Paul H.,M.D. wrote:
> Absolutely right idea. I thnk the DS2438 is the prefered configuration chip (It
> has many functions).

Hmmm.  Yes, it's a useful chip, but with only 40 bytes I'm not sure if
it would have enough storage to do what I was thinking.  Of course, it
may be plenty for what Jan wants to do.

ttyl
srw


-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r=20
_______________________________________________
Owfs-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/owfs-developers
Reply | Threaded
Open this post in threaded view
|

Re: First Post and some questions

Scott Walde
In reply to this post by Jan Kandziora

>>Hi.  I've been pretty quiet on this list so far, so I feel a bit awkward
>>jumping in right here.  Anyways, I have another suggestion for
>>identifying aggregates.  You can put an EPROM device such as the DS2505
>>into each aggregate device and place information on that EPROM
>>describing the components on that device.
>>
>
> Hm. I don't see a big benefit over the unique IDs all devices have. You still
> have to maintain a database of serial numbers. The crucial part is to get the
> same label *into* the hardware and *onto* the hardware. I does not matter
> which data is placed on the label, it just has to be unique.

The advantage is with this method, you only need to recognize the
DS2505.  Once you read the data off the DS2505, you can use that
information to logically cluster the rest of the devices.

Perhaps I wasn't completely clear on my idea...  When you build the
aggregate device, the DS2505 would be programmed with the addresses
(serial numbers) of each individual component in the aggregate device.
Then, when you later discover a 2505 on the bus, you read the data out
of it to obtain the list of addresses (serial numbers) of the individual
components in that aggregate device.  Now, the only serial number you
need to know to identify the aggregate device is the serial number of
the DS2505.

ttyl
srw


-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r=20
_______________________________________________
Owfs-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/owfs-developers
Reply | Threaded
Open this post in threaded view
|

Re: First Post and some questions

Jan Kandziora
Am Donnerstag, 9. Juni 2005 23:08 schrieb Scott Walde:

> >>Hi.  I've been pretty quiet on this list so far, so I feel a bit awkward
> >>jumping in right here.  Anyways, I have another suggestion for
> >>identifying aggregates.  You can put an EPROM device such as the DS2505
> >>into each aggregate device and place information on that EPROM
> >>describing the components on that device.
> >
> > Hm. I don't see a big benefit over the unique IDs all devices have. You
> > still have to maintain a database of serial numbers. The crucial part is
> > to get the same label *into* the hardware and *onto* the hardware. I does
> > not matter which data is placed on the label, it just has to be unique.
>
> The advantage is with this method, you only need to recognize the
> DS2505.  Once you read the data off the DS2505, you can use that
> information to logically cluster the rest of the devices.
>
> Perhaps I wasn't completely clear on my idea...  When you build the
> aggregate device, the DS2505 would be programmed with the addresses
> (serial numbers) of each individual component in the aggregate device.
> Then, when you later discover a 2505 on the bus, you read the data out
> of it to obtain the list of addresses (serial numbers) of the individual
> components in that aggregate device.  Now, the only serial number you
> need to know to identify the aggregate device is the serial number of
> the DS2505.
>
My idea is to recognize one single chip of the aggregate, too. Then look it up
in my database and still knowing which other chips are on the device in
question. That information was stored previously, when the module was hooked
to a 1-Wire-Bus solely. You have to do the same, as you don't know the chip
IDs beforehand.

Our ideas only differ in one thing: You want to put additional data about the
module into an additional chip on the module, I will put this information
onto the host's harddisk/flash.

Kind regards

        Jan



-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r=20
_______________________________________________
Owfs-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/owfs-developers
Reply | Threaded
Open this post in threaded view
|

Re: First Post and some questions

Jan Kandziora
In reply to this post by Alfille, Paul H.,M.D.
Am Donnerstag, 9. Juni 2005 21:03 schrieb Alfille, Paul H.,M.D.:
> Tested it out:
> If an owserver process fails, The system reports no content on that bus
> entry, until it is restored. Probably exactly what you want.
>
Yes, already found this out, too. Excellent!


> Will you know all the owservers at initiation time, or will you need to add
> more as your system functions (hotplugging)?
>
For the configuration program, I need this - but this is no critical software,
so I can work out restarting owserver with some scripts.

After the configuration of the machine in question is completed and the
machine works as expected, "hotplugging" will only occur when some component
failed and is reconnected again - I think of lose cables, broken ethernet
switches (most of their AC adaptors are pure crap) or simply brownouts or
blackouts (mains cable pulled).

Kind regards

        Jan



-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r=20
_______________________________________________
Owfs-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/owfs-developers
12