gumstix

Hi Dustin,

On Wed, Aug 19, 2009 at 11:49 AM, Dustin Webb<dustinjwebb@...> wrote:
> I have been able to power my Robostix from a battery and communicate with it
> via serial. Now I would like to attach my Verdex Pro. However the WiFi on my
> Verdex Pro doesn't work if I am powering the stack with the battery, it only
> works if I power the stack with a wall wart. Does anyone know what the
> problem is and how to get around it?

You need to either power both the robostix and the wifistix, or make
the mod described in this document:
<http://pubs.gumstix.com/documents/robostix_power.pdf>

-- 
Dave Hylands
Shuswap, BC, Canada
http://www.DaveHylands.com/

Dave:

Aha! That's the other modification I was talking about last week. That wasn't the problem though. On a whim I tried using two 9V batteries in parallel and it worked. Now I just need something more robust than these little 9V batteries. I'm going to try using two 6 cell NiCD batteries in parallel as soon as I make a second connector. Otherwise I might have to try to find some RC batteries closer to 9V.

Thanks,
Dustin Webb




________________________________
From: Dave Hylands <dhylands@...>
To: General mailing list for gumstix users. <gumstix-users@...>
Sent: Wednesday, August 19, 2009 1:43:38 PM
Subject: Re: [Gumstix-users] WiFi problems

Hi Dustin,

On Wed, Aug 19, 2009 at 11:49 AM, Dustin Webb<dustinjwebb@...> wrote:
> I have been able to power my Robostix from a battery and communicate with it
> via serial. Now I would like to attach my Verdex Pro. However the WiFi on my
> Verdex Pro doesn't work if I am powering the stack with the battery, it only
> works if I power the stack with a wall wart. Does anyone know what the
> problem is and how to get around it?

You need to either power both the robostix and the wifistix, or make
the mod described in this document:
<http://pubs.gumstix.com/documents/robostix_power.pdf>

-- 
Dave Hylands
Shuswap, BC, Canada
http://www.DaveHylands.com/

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with 
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
gumstix-users mailing list
gumstix-users@...
https://lists.sourceforge.net/lists/listinfo/gumstix-users

Hi Dustin,

On Wed, Aug 19, 2009 at 1:23 PM, Dustin Webb<dustinjwebb@...> wrote:
> Dave:
>
> Aha! That's the other modification I was talking about last week. That
> wasn't the problem though. On a whim I tried using two 9V batteries in
> parallel and it worked. Now I just need something more robust than these
> little 9V batteries. I'm going to try using two 6 cell NiCD batteries in
> parallel as soon as I make a second connector. Otherwise I might have to try
> to find some RC batteries closer to 9V.

Ahhh - Yeah 9V batteries are typically only rated to a few hundred
milliamps. With Wifi, you definitely want an amp.

You'd actually be better off to pick batteries closer to 5V.
Everything above 5V is wasted as heat in the voltage regulators. A
5-cell NiCad or NiMH will run from around 7V fully charged down to
about 5V when discharged.

A 4-cell pack will go from about 5.6V down to 4V.

Here's a site that shows a typical discharge curve for NiMH and NiCAD.

If you get a DC-DC converter, like this one:
http://www.sparkfun.com/commerce/product_info.php?products_id=9275
then you can use 9V or 12V and the excess won't be wasted as heat, so
you'll get longer battery life.

-- 
Dave Hylands
Shuswap, BC, Canada
http://www.DaveHylands.com/

WIth linear regulators that would be true, re heat dissipation, but these
are (I thought) using switching regulators which are pretty efficient.
Also, with linear regulators, unless they are low drop out regulators, you
tend to need a bit higher voltage than the regulator's rated voltage or they
go out of regulation and start to sag (voltage wise) which can cause all
sorts of issues with reliable system operation.

Rick


On Wed, Aug 19, 2009 at 4:57 PM, Dave Hylands <dhylands@...> wrote:

> Hi Dustin,
>
> On Wed, Aug 19, 2009 at 1:23 PM, Dustin Webb<dustinjwebb@...> wrote:
> > Dave:
> >
> > Aha! That's the other modification I was talking about last week. That
> > wasn't the problem though. On a whim I tried using two 9V batteries in
> > parallel and it worked. Now I just need something more robust than these
> > little 9V batteries. I'm going to try using two 6 cell NiCD batteries in
> > parallel as soon as I make a second connector. Otherwise I might have to
> try
> > to find some RC batteries closer to 9V.
>
> Ahhh - Yeah 9V batteries are typically only rated to a few hundred
> milliamps. With Wifi, you definitely want an amp.
>
> You'd actually be better off to pick batteries closer to 5V.
> Everything above 5V is wasted as heat in the voltage regulators. A
> 5-cell NiCad or NiMH will run from around 7V fully charged down to
> about 5V when discharged.
>
> A 4-cell pack will go from about 5.6V down to 4V.
>
> Here's a site that shows a typical discharge curve for NiMH and NiCAD.
>
> If you get a DC-DC converter, like this one:
> http://www.sparkfun.com/commerce/product_info.php?products_id=9275
> then you can use 9V or 12V and the excess won't be wasted as heat, so
> you'll get longer battery life.
>
> --
> Dave Hylands
> Shuswap, BC, Canada
> http://www.DaveHylands.com/
>
>
> ------------------------------------------------------------------------------
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
> trial. Simplify your report design, integration and deployment - and focus
> on
> what you do best, core application coding. Discover what's new with
> Crystal Reports now.  http://p.sf.net/sfu/bobj-july
> _______________________________________________
> gumstix-users mailing list
> gumstix-users@...
> https://lists.sourceforge.net/lists/listinfo/gumstix-users
>



-- 
Say you can or say you can't, either way you will be right.
Computers are like old testament gods: Lots of rules and no mercy.

Hi Rick,

On Wed, Aug 19, 2009 at 2:33 PM, richard dorfner<rdorfner@...> wrote:
> WIth linear regulators that would be true, re heat dissipation, but these
> are (I thought) using switching regulators which are pretty efficient.
> Also, with linear regulators, unless they are low drop out regulators, you
> tend to need a bit higher voltage than the regulator's rated voltage or they
> go out of regulation and start to sag (voltage wise) which can cause all
> sorts of issues with reliable system operation.

All of the regulators on the robostix are LDO (Low Dropout linear),
which includes the one providing 5v to the gumstix. I think that the
gumstix itself uses an onboard switching regulator, but most of the
daughtercards use linear regulators (I haven't seen the Palo).

-- 
Dave Hylands
Shuswap, BC, Canada
http://www.DaveHylands.com/

Ah, Gotcha.  Yeah,the gumstix itself has a switcher, I can see the itty
bitty teeny weeny little inductors. :)
So yup, using a higher voltage source would just generate a lot more heat.
Keeping the voltage of the source lower would help, and keeping the current
higher would help keeping it from sagging and causing problems.

Thanks, Dave.
Rick


On Wed, Aug 19, 2009 at 5:07 PM, Dave Hylands <dhylands@...> wrote:

> Hi Rick,
>
> On Wed, Aug 19, 2009 at 2:33 PM, richard dorfner<rdorfner@...>
> wrote:
> > WIth linear regulators that would be true, re heat dissipation, but these
> > are (I thought) using switching regulators which are pretty efficient.
> > Also, with linear regulators, unless they are low drop out regulators,
> you
> > tend to need a bit higher voltage than the regulator's rated voltage or
> they
> > go out of regulation and start to sag (voltage wise) which can cause all
> > sorts of issues with reliable system operation.
>
> All of the regulators on the robostix are LDO (Low Dropout linear),
> which includes the one providing 5v to the gumstix. I think that the
> gumstix itself uses an onboard switching regulator, but most of the
> daughtercards use linear regulators (I haven't seen the Palo).
>
> --
> Dave Hylands
> Shuswap, BC, Canada
> http://www.DaveHylands.com/
>
>
> ------------------------------------------------------------------------------
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
> trial. Simplify your report design, integration and deployment - and focus
> on
> what you do best, core application coding. Discover what's new with
> Crystal Reports now.  http://p.sf.net/sfu/bobj-july
> _______________________________________________
> gumstix-users mailing list
> gumstix-users@...
> https://lists.sourceforge.net/lists/listinfo/gumstix-users
>



-- 
Say you can or say you can't, either way you will be right.
Computers are like old testament gods: Lots of rules and no mercy.

Thanks again for the info Dave.



> You'd actually be better off to pick batteries closer to 5V.
> Everything above 5V is wasted as heat in the voltage regulators. A
> 5-cell NiCad or NiMH will run from around 7V fully charged down to
> about 5V when discharged.

It's interesting that you say this because this is what I thought based on the documentation. However I find that if I don't have close to 9V then my WiFi doesn't work. In fact i was originally using a 6V battery but had to change batteries because of this problem. Not sure what I am doing wrong.

Dustin

Hi Dustin,

On Wed, Aug 19, 2009 at 6:21 PM, Dustin Webb<dustinjwebb@...> wrote:
> Thanks again for the info Dave.
>
>> You'd actually be better off to pick batteries closer to 5V.
>> Everything above 5V is wasted as heat in the voltage regulators. A
>> 5-cell NiCad or NiMH will run from around 7V fully charged down to
>> about 5V when discharged.
>
> It's interesting that you say this because this is what I thought based on
> the documentation. However I find that if I don't have close to 9V then my
> WiFi doesn't work. In fact i was originally using a 6V battery but had to
> change batteries because of this problem. Not sure what I am doing wrong.

It may depend on the battery. Cheap alkalines, for example, have a
high internal resistance and are unable to deliver big inrushes of
current like the NiCAD and NiMH batteries can.

And the voltage of a battery will droop when you take a big load from
it (Wifi represents a pretty significant load and it's quite bursty).
So it's possible that the 6v battery (the 6v is only a nominal
voltage) may be sagging enough to cause problems.

When you look at batteries they have a couple of different ratings.
One is mAh (milliamp hours) which represents how long it should last
delivering a particular current. The other is N-C where N is a number.

Here's some good background: <http://batteryuniversity.com/print-partone-16.htm>

So having a 6v battery should be fine, provided it has a low internal
reistance (like NiMH or NiCAD) and that you've got enough oomph. So
you'll want a battery with at least 1000 mAh.

-- 
Dave Hylands
Shuswap, BC, Canada
http://www.DaveHylands.com/

In addition to this, if you have an oscilloscope available to you, you can
measure this with the scope and see if this is what's happening. It's
probably a worthwhile experiment for you i fyou have the tools availble to
you.

Embedded software guys and an oscope, don't leave home without it. :)

Let us know what you find!
Rick


On Thu, Aug 20, 2009 at 2:18 AM, Dave Hylands <dhylands@...> wrote:

> Hi Dustin,
>
> On Wed, Aug 19, 2009 at 6:21 PM, Dustin Webb<dustinjwebb@...> wrote:
> > Thanks again for the info Dave.
> >
> >> You'd actually be better off to pick batteries closer to 5V.
> >> Everything above 5V is wasted as heat in the voltage regulators. A
> >> 5-cell NiCad or NiMH will run from around 7V fully charged down to
> >> about 5V when discharged.
> >
> > It's interesting that you say this because this is what I thought based
> on
> > the documentation. However I find that if I don't have close to 9V then
> my
> > WiFi doesn't work. In fact i was originally using a 6V battery but had to
> > change batteries because of this problem. Not sure what I am doing wrong.
>
> It may depend on the battery. Cheap alkalines, for example, have a
> high internal resistance and are unable to deliver big inrushes of
> current like the NiCAD and NiMH batteries can.
>
> And the voltage of a battery will droop when you take a big load from
> it (Wifi represents a pretty significant load and it's quite bursty).
> So it's possible that the 6v battery (the 6v is only a nominal
> voltage) may be sagging enough to cause problems.
>
> When you look at batteries they have a couple of different ratings.
> One is mAh (milliamp hours) which represents how long it should last
> delivering a particular current. The other is N-C where N is a number.
>
> Here's some good background: <
> http://batteryuniversity.com/print-partone-16.htm>
>
> So having a 6v battery should be fine, provided it has a low internal
> reistance (like NiMH or NiCAD) and that you've got enough oomph. So
> you'll want a battery with at least 1000 mAh.
>
> --
> Dave Hylands
> Shuswap, BC, Canada
> http://www.DaveHylands.com/
>
>
> ------------------------------------------------------------------------------
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
> trial. Simplify your report design, integration and deployment - and focus
> on
> what you do best, core application coding. Discover what's new with
> Crystal Reports now.  http://p.sf.net/sfu/bobj-july
> _______________________________________________
> gumstix-users mailing list
> gumstix-users@...
> https://lists.sourceforge.net/lists/listinfo/gumstix-users
>



-- 
Say you can or say you can't, either way you will be right.
Computers are like old testament gods: Lots of rules and no mercy.

Hi Richard:


> In addition to this, if you have an oscilloscope available to you, you can measure this with the scope and see if this is what's happening.

Unfortunately I don't have access to an oscilloscope though I have been seriously considering buying the little one on seeedstudio.com <http://www.seeedstudio.com/depot/digital-storage-oscilloscope-diy-kit-with-panels-p-166.html>. If you or anyone else has an opinion of that device I would be interested to hear it.

Dustin

Hi Dave:


> It may depend on the battery. Cheap alkalines, for example, have a
> high internal resistance and are unable to deliver big inrushes of
> current like the NiCAD and NiMH batteries can.

I have actually tried both. My NiCD are 7.2V and my NiMH are 6V but I think they may have been low on charge so I am recharging them. I'll let you know what I find.

> And the voltage of a battery will droop when you take a big load from
> it (Wifi represents a pretty significant load and it's quite bursty).

I have actually observed this though I didn't realize what I was seeing.

> When you look at batteries they have a couple of different ratings.
> One is mAh (milliamp hours) which represents how long it should last
>delivering a particular current. The other is N-C where N is a number.
>
> Here's some good background: <http://batteryuniversity.com/print-partone-16.htm>

As usual you are a fountain of information. I will definitely read over that doc.

Thanks again,
Dustin

welll.. it's uhm.. *cough* certainly inexpensive. :)  Honeslty though, you
would have to read reviews of folks who have used it. The trick will be
getting it to trigger properly and I have no clue if it has a good trigger
circuit on it or not.. and if it can capture just a single trace or if it
would continuously trigger etc etc.. and the real cost will be the current
proble since that is what will let you calculate the power being used by the
device.

Generally speaking, power measurements are something that takes a fair
amount of not cheap equipment to get. It lets you do things like find inrush
currents and operrating currents curing various useages.

If you are working on at a hobbyist level, then you might be able to simply
use teh data sheets for the wifi device and simply use max rated power
consumption to estimate your overall power needs. If this is for an actual
product development for a client, I would suggest asking the client if they
have that kind of equipment available or know of someone who has it
available so that you can get "proper" power measurements.

I would suspect trying to get that information on the cheap will be a
frustrating excersise at best, and generating misleading results or
impossible at worst.

Rick


On Thu, Aug 20, 2009 at 1:09 PM, Dustin Webb <dustinjwebb@...> wrote:

> Hi Richard:
>
> > In addition to this, if you have an oscilloscope available to you, you
> can measure this with the scope and see if this is what's happening.
>
> Unfortunately I don't have access to an oscilloscope though I have been
> seriously considering buying the little one on seeedstudio.com <
> http://www.seeedstudio.com/depot/digital-storage-oscilloscope-diy-kit-with-panels-p-166.html>.
> If you or anyone else has an opinion of that device I would be interested to
> hear it.
>
> Dustin
>
>
>
>
> ------------------------------------------------------------------------------
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
> trial. Simplify your report design, integration and deployment - and focus
> on
> what you do best, core application coding. Discover what's new with
> Crystal Reports now.  http://p.sf.net/sfu/bobj-july
> _______________________________________________
> gumstix-users mailing list
> gumstix-users@...
> https://lists.sourceforge.net/lists/listinfo/gumstix-users
>
>


-- 
Say you can or say you can't, either way you will be right.
Computers are like old testament gods: Lots of rules and no mercy.

I am trying to get the WiFi working on my Overo Fire follwoing the
instructions posted at http://wiki.gumstix.org/index.php?title=Overo_Wifi
and I am getting problems at boot. Once I remove networkmanager the wlan0
interface does not seem to come up until some time after the boot sequence
has completed and as a result several of the boot steps fail. I can get the
WiFi to work if I restart the network after boot /etc/init.d/networking
restart although it sometimes takes a couple of attempts. I am assuming the
hardware is good, but that there is something flaky with the kernel or
driver, any pointers on debuging or fixing the problem would be much
appreciated.

Below are the boot errors I am seeing.


[snip]
usbcore: registered new interface driver option
option: v0.7.2:USB Driver for GSM modems
lib80211: common routines for IEEE802.11 drivers
cfg80211: Calling CRDA to update world regulatory domain
libertas_sdio: Libertas SDIO driver
libertas_sdio: Copyright Pierre Ossman
Remounting root file system...

udevadm settle - timeout of 8 seconds reached, the event queue contains:
 
/sys/devices/platform/omap/omap_hsmmc.0/mmc_host/mmc0/mmc0:b368/block/mmcblk0
(639)
 
/sys/devices/platform/omap/omap_hsmmc.0/mmc_host/mmc0/mmc0:b368/block/mmcblk0/mmcblk0p1
(640)
 
/sys/devices/platform/omap/omap_hsmmc.0/mmc_host/mmc0/mmc0:b368/block/mmcblk0/mmcblk0p2
(641)
 
/sys/devices/platform/omap/omap_hsmmc.1/mmc_host/mmc1/mmc1:0001/mmc1:0001:1
(644)
Caching udev devnodes
Populating dev cache
NET: Registered protocol family 10
ALSA: Restoring mixer settings...
Cleaning: /etc/network/run/ifstate.
Configuring network interfaces... Error for wireless request "Set Encode"
(8B2A) :
    SET failed on device wlan0 ; No such device.
Failed to bring up wlan0.
done.
Starting portmap daemon: portmap.
[snip]
...
[snip]
Starting GPE display manager: gpe-dm

.---O---.
|       |                  .-.           o o
|   |   |-----.-----.-----.| |   .----..-----.-----.
|       |     | __  |  ---'| '--.|  .-'|     |     |
|   |   |  |  |     |---  ||  --'|  |  |  '  | | | |
'---'---'--'--'--.  |-----''----''--'  '-----'-'-'-'
                -'  |
                '---'

The Angstrom Distribution overo ttyO2

Angstrom 2011.03 overo ttyO2

overo login: libertas: 00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303
libertas_sdio mmc1:0001:1: wlan0: Features changed: 0x00004800 -> 0x00004000
libertas: wlan0: Marvell WLAN 802.11 adapter
libertas: assoc: bss   (null) not in scan results
ADDRCONF(NETDEV_UP): wlan0: link is not ready

-- 
View this message in context: http://old.nabble.com/WiFi-problems-tp32473984p32473984.html
Sent from the Gumstix mailing list archive at Nabble.com.

I had some similar issues and was able to work through them.  What does your
/etc/network/interfaces file look like (any uncommented wlan0 entries)?

Brett
On Sep 19, 2011 5:58 PM, "joncoops" <joncoops@...> wrote:
>
>
> I am trying to get the WiFi working on my Overo Fire follwoing the
> instructions posted at http://wiki.gumstix.org/index.php?title=Overo_Wifi
> and I am getting problems at boot. Once I remove networkmanager the wlan0
> interface does not seem to come up until some time after the boot sequence
> has completed and as a result several of the boot steps fail. I can get
the
> WiFi to work if I restart the network after boot /etc/init.d/networking
> restart although it sometimes takes a couple of attempts. I am assuming
the
> hardware is good, but that there is something flaky with the kernel or
> driver, any pointers on debuging or fixing the problem would be much
> appreciated.
>
> Below are the boot errors I am seeing.
>
>
> [snip]
> usbcore: registered new interface driver option
> option: v0.7.2:USB Driver for GSM modems
> lib80211: common routines for IEEE802.11 drivers
> cfg80211: Calling CRDA to update world regulatory domain
> libertas_sdio: Libertas SDIO driver
> libertas_sdio: Copyright Pierre Ossman
> Remounting root file system...
>
> udevadm settle - timeout of 8 seconds reached, the event queue contains:
>
>
/sys/devices/platform/omap/omap_hsmmc.0/mmc_host/mmc0/mmc0:b368/block/mmcblk0
> (639)
>
>
/sys/devices/platform/omap/omap_hsmmc.0/mmc_host/mmc0/mmc0:b368/block/mmcblk0/mmcblk0p1
> (640)
>
>
/sys/devices/platform/omap/omap_hsmmc.0/mmc_host/mmc0/mmc0:b368/block/mmcblk0/mmcblk0p2
> (641)
>
>
/sys/devices/platform/omap/omap_hsmmc.1/mmc_host/mmc1/mmc1:0001/mmc1:0001:1
> (644)
> Caching udev devnodes
> Populating dev cache
> NET: Registered protocol family 10
> ALSA: Restoring mixer settings...
> Cleaning: /etc/network/run/ifstate.
> Configuring network interfaces... Error for wireless request "Set Encode"
> (8B2A) :
>    SET failed on device wlan0 ; No such device.
> Failed to bring up wlan0.
> done.
> Starting portmap daemon: portmap.
> [snip]
> ...
> [snip]
> Starting GPE display manager: gpe-dm
>
> .---O---.
> |       |                  .-.           o o
> |   |   |-----.-----.-----.| |   .----..-----.-----.
> |       |     | __  |  ---'| '--.|  .-'|     |     |
> |   |   |  |  |     |---  ||  --'|  |  |  '  | | | |
> '---'---'--'--'--.  |-----''----''--'  '-----'-'-'-'
>                -'  |
>                '---'
>
> The Angstrom Distribution overo ttyO2
>
> Angstrom 2011.03 overo ttyO2
>
> overo login: libertas: 00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303
> libertas_sdio mmc1:0001:1: wlan0: Features changed: 0x00004800 ->
0x00004000
> libertas: wlan0: Marvell WLAN 802.11 adapter
> libertas: assoc: bss   (null) not in scan results
> ADDRCONF(NETDEV_UP): wlan0: link is not ready
>
> --
> View this message in context:
http://old.nabble.com/WiFi-problems-tp32473984p32473984.html
> Sent from the Gumstix mailing list archive at Nabble.com.
>
>
>
------------------------------------------------------------------------------
> All the data continuously generated in your IT infrastructure contains a
> definitive record of customers, application performance, security
> threats, fraudulent activity and more. Splunk takes this data and makes
> sense of it. Business sense. IT sense. Common sense.
> http://p.sf.net/sfu/splunk-d2dcopy1
> _______________________________________________
> gumstix-users mailing list
> gumstix-users@...
> https://lists.sourceforge.net/lists/listinfo/gumstix-users

Hi
I am pasting the whole process I documented that worked for me:

*For testing:*

root@...:~# iwconfig wlan0 essid any

root@...:~# ifconfig wlan0 up

root@...:~# iwlist wlan0 scan



*Now configuration:*

root@...:~# cat >> /etc/network/interfaces

allow-hotplug wlan0

auto wlan0

iface wlan0 inet dhcp

        pre-up /sbin/iwconfig wlan0 essid any

        wireless_mode managed

(pressed ctrl C to exit this mode)

*Remove networkmanager: (I had to remove it for wifi to work)*

root@...:~# opkg remove networkmanager --force-depends

root@...:~# ifdown wlan0

root@...:~# ifup wlan0



*I entered*: ping google.com to check if you have a connection.

root@...:~# ping google.com

If there is no connection you can try bringing network down and up again.

This will give you information regarding your connections:

root@...:~# ifconfig –a

For more information:

http://wiki.gumstix.org/index.php?title=Overo_Wifi

The problem for me is that this way wifi does not start by itself on boot,
so I tried the following:

1)

Creating the file : "/etc/modprobe.conf" with the content:

options libertas_sdio helper_name="sd8686_helper.bin"

options libertas_sdio fw_name="sd8686.bin"



This only works if I connect through serial as usual, and then type "boot",
and then it works automatically.



2)

Add a script into /etc/init.d/, but first change directory:

cd /etc/init.d/



I added a file using "nano myInit.sh"

Then added the following text to the file:

#!/bin/sh

Ifdown wlan0

Ifup wlan0



This script should be marked as executable:

chmod +x myInit.sh



Your next step is to add a symlink to this script from /etc/rc5.d/

First change directory "cd /etc/rc5.d/"

Then type: "ln -s ../init.d/myInit.sh S96myInit"



S: basically says to run a script on startup,

96: is for the ordering of when the scripts are run.

myInit: is a name for our own reference.



In order to check that you don't use the same ordering number twice, check
using this command:

"ls –la" while in the /etc/rc5.d/ directory

The highest number script which is generally pre-installed on a basic
Gumstix Buildroot or OE system is S90, but better check!



To check that it works beforehand, type:

/etc/rc5.d/S96myInit



For more information:

http://www.iualdii.net/articles/gumstix-boot-scripts


On 20 September 2011 03:08, Brett Hemes <brett.hemes@...> wrote:

> I had some similar issues and was able to work through them.  What does
> your /etc/network/interfaces file look like (any uncommented wlan0 entries)?
>
> Brett
>
> On Sep 19, 2011 5:58 PM, "joncoops" <joncoops@...> wrote:
> >
> >
> > I am trying to get the WiFi working on my Overo Fire follwoing the
> > instructions posted at
> http://wiki.gumstix.org/index.php?title=Overo_Wifi
> > and I am getting problems at boot. Once I remove networkmanager the wlan0
> > interface does not seem to come up until some time after the boot
> sequence
> > has completed and as a result several of the boot steps fail. I can get
> the
> > WiFi to work if I restart the network after boot /etc/init.d/networking
> > restart although it sometimes takes a couple of attempts. I am assuming
> the
> > hardware is good, but that there is something flaky with the kernel or
> > driver, any pointers on debuging or fixing the problem would be much
> > appreciated.
> >
> > Below are the boot errors I am seeing.
> >
> >
> > [snip]
> > usbcore: registered new interface driver option
> > option: v0.7.2:USB Driver for GSM modems
> > lib80211: common routines for IEEE802.11 drivers
> > cfg80211: Calling CRDA to update world regulatory domain
> > libertas_sdio: Libertas SDIO driver
> > libertas_sdio: Copyright Pierre Ossman
> > Remounting root file system...
> >
> > udevadm settle - timeout of 8 seconds reached, the event queue contains:
> >
> >
> /sys/devices/platform/omap/omap_hsmmc.0/mmc_host/mmc0/mmc0:b368/block/mmcblk0
> > (639)
> >
> >
> /sys/devices/platform/omap/omap_hsmmc.0/mmc_host/mmc0/mmc0:b368/block/mmcblk0/mmcblk0p1
> > (640)
> >
> >
> /sys/devices/platform/omap/omap_hsmmc.0/mmc_host/mmc0/mmc0:b368/block/mmcblk0/mmcblk0p2
> > (641)
> >
> >
> /sys/devices/platform/omap/omap_hsmmc.1/mmc_host/mmc1/mmc1:0001/mmc1:0001:1
> > (644)
> > Caching udev devnodes
> > Populating dev cache
> > NET: Registered protocol family 10
> > ALSA: Restoring mixer settings...
> > Cleaning: /etc/network/run/ifstate.
> > Configuring network interfaces... Error for wireless request "Set Encode"
> > (8B2A) :
> >    SET failed on device wlan0 ; No such device.
> > Failed to bring up wlan0.
> > done.
> > Starting portmap daemon: portmap.
> > [snip]
> > ...
> > [snip]
> > Starting GPE display manager: gpe-dm
> >
> > .---O---.
> > |       |                  .-.           o o
> > |   |   |-----.-----.-----.| |   .----..-----.-----.
> > |       |     | __  |  ---'| '--.|  .-'|     |     |
> > |   |   |  |  |     |---  ||  --'|  |  |  '  | | | |
> > '---'---'--'--'--.  |-----''----''--'  '-----'-'-'-'
> >                -'  |
> >                '---'
> >
> > The Angstrom Distribution overo ttyO2
> >
> > Angstrom 2011.03 overo ttyO2
> >
> > overo login: libertas: 00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303
> > libertas_sdio mmc1:0001:1: wlan0: Features changed: 0x00004800 ->
> 0x00004000
> > libertas: wlan0: Marvell WLAN 802.11 adapter
> > libertas: assoc: bss   (null) not in scan results
> > ADDRCONF(NETDEV_UP): wlan0: link is not ready
> >
> > --
> > View this message in context:
> http://old.nabble.com/WiFi-problems-tp32473984p32473984.html
> > Sent from the Gumstix mailing list archive at Nabble.com.
> >
> >
> >
> ------------------------------------------------------------------------------
> > All the data continuously generated in your IT infrastructure contains a
> > definitive record of customers, application performance, security
> > threats, fraudulent activity and more. Splunk takes this data and makes
> > sense of it. Business sense. IT sense. Common sense.
> > http://p.sf.net/sfu/splunk-d2dcopy1
> > _______________________________________________
> > gumstix-users mailing list
> > gumstix-users@...
> > https://lists.sourceforge.net/lists/listinfo/gumstix-users
>
>
> ------------------------------------------------------------------------------
> All the data continuously generated in your IT infrastructure contains a
> definitive record of customers, application performance, security
> threats, fraudulent activity and more. Splunk takes this data and makes
> sense of it. Business sense. IT sense. Common sense.
> http://p.sf.net/sfu/splunk-d2dcopy1
> _______________________________________________
> gumstix-users mailing list
> gumstix-users@...
> https://lists.sourceforge.net/lists/listinfo/gumstix-users
>
>

Hello,

I guess I have the same problem on the latest console Image from
http://cumulus.gumstix.org/images/angstrom/developer/2011-08-30-1058/

Sometimes during startup it lasts around 2 Minutes till the line
[snip]
libertas: 00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303
[snip]

is printed to the console and 
sometimes the startup runs through quite fluently and after some time I get
the output as Jon was pasting it. After that it is possible to start the
wlan using iwconfig, but not before (see snip from Jon below).

joncoops wrote:
> 
> [snip]
> libertas_sdio: Copyright Pierre Ossman
> ....
> overo login: libertas: 00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303
> 

Jon, do you have also sometimes the situation, that "libertas:
00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303" is already printed during
startup after around 2 Minutes?

@Nick and Eric:
Either "Ifdown wlan0..." nor adjustments in "/etc/.../Interfaces" will work
as long as this "libertas: 00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303"
output was not printed to the console

Gerhard
-- 
View this message in context: http://old.nabble.com/WiFi-problems-tp32473984p32503864.html
Sent from the Gumstix mailing list archive at Nabble.com.

Gerhard,

Nice information.

I must correct my first post. Adding "iwlist..." to /etc/network/interfaces
alone does not solve the problem I experienced wherein DHCP is not assigned.
I used Nick Starzicki's method, in his post in this thread, and now I see
100% success on power-up boot. That is, just add a boot process (mine in
/etc/rc5.d) that is second to last just behind the final login process and
have it perform "ifdown -a" followed by "ifup -a". I still find the need for
"iwlist..." in interfaces, however.

Eric


Gerhard H. wrote:
> 
> Hello,
> 
> I guess I have the same problem on the latest console Image from
> http://cumulus.gumstix.org/images/angstrom/developer/2011-08-30-1058/
> 
> Sometimes during startup it lasts around 2 Minutes till the line
> [snip]
> libertas: 00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303
> [snip]
> 
> is printed to the console and 
> sometimes the startup runs through quite fluently and after some time I
> get the output as Jon was pasting it. After that it is possible to start
> the wlan using iwconfig, but not before (see snip from Jon below).
> 
> joncoops wrote:
>> 
>> [snip]
>> libertas_sdio: Copyright Pierre Ossman
>> ....
>> overo login: libertas: 00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303
>> 
> 
> Jon, do you have also sometimes the situation, that "libertas:
> 00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303" is already printed during
> startup after around 2 Minutes?
> 
> @Nick and Eric:
> Either "Ifdown wlan0..." nor adjustments in "/etc/.../Interfaces" will
> work as long as this "libertas: 00:19:88:3e:7e:d2, fw 9.70.7p0, cap
> 0x00000303" output was not printed to the console
> 
> Gerhard
> 

-- 
View this message in context: http://old.nabble.com/WiFi-problems-tp32473984p32503899.html
Sent from the Gumstix mailing list archive at Nabble.com.

Hello,

I guess I have the same problem on the latest console Image from
http://cumulus.gumstix.org/images/angstrom/developer/2011-08-30-1058/

Sometimes during startup it lasts around 2 Minutes till the line
[snip]
libertas: 00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303
[snip]

is printed to the console and 
sometimes the startup runs through quite fluently and after some time I get
the output as Jon was pasting it. After that it is possible to start the
wlan using iwconfig, but not before (see snip from Jon below).

joncoops wrote:
> 
> [snip]
> libertas_sdio: Copyright Pierre Ossman
> ....
> overo login: libertas: 00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303
> 

Jon, do you have also sometimes the situation, that "libertas:
00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303" is already printed during
startup after around 2 Minutes?

@Nick and Eric:
Either "Ifdown wlan0..." nor adjustments in "/etc/.../Interfaces" will work
as long as this "libertas: 00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303"
output was not printed to the console

Gerhard

FYI: sent again because I had problems that the post was accepted by the
mailing list
-- 
View this message in context: http://old.nabble.com/WiFi-problems-tp32473984p32503929.html
Sent from the Gumstix mailing list archive at Nabble.com.

Gerhard,

As you observe it seems to take a while and then I get the 
[snip]
libertas: 00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303
[snip]
message after that I can start the interface. Before this when typing
ifconfig the interface for the WiFi does not show up.

I have added a deamon process to check periodicallty for connectivity and
restart the network interface if it cannot reach the internet. This seems
like using a sledge hammer to crack a walnut, but it works. I am really
curious why the WiFi hardware does not seem to come up on boot.

Jon



Gerhard H. wrote:
> 
> 
> Hello,
> 
> I guess I have the same problem on the latest console Image from
> http://cumulus.gumstix.org/images/angstrom/developer/2011-08-30-1058/
> 
> Sometimes during startup it lasts around 2 Minutes till the line
> [snip]
> libertas: 00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303
> [snip]
> 
> is printed to the console and 
> sometimes the startup runs through quite fluently and after some time I
> get
> the output as Jon was pasting it. After that it is possible to start the
> wlan using iwconfig, but not before (see snip from Jon below).
> 
> joncoops wrote:
>> 
>> [snip]
>> libertas_sdio: Copyright Pierre Ossman
>> ....
>> overo login: libertas: 00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303
>> 
> 
> Jon, do you have also sometimes the situation, that "libertas:
> 00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303" is already printed during
> startup after around 2 Minutes?
> 
> @Nick and Eric:
> Either "Ifdown wlan0..." nor adjustments in "/etc/.../Interfaces" will
> work
> as long as this "libertas: 00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303"
> output was not printed to the console
> 
> Gerhard
> -- 
> View this message in context:
> http://old.nabble.com/WiFi-problems-tp32473984p32503864.html
> Sent from the Gumstix mailing list archive at Nabble.com.
> 
> 
> ------------------------------------------------------------------------------
> All of the data generated in your IT infrastructure is seriously valuable.
> Why? It contains a definitive record of application performance, security
> threats, fraudulent activity, and more. Splunk takes this data and makes
> sense of it. IT sense. And common sense.
> http://p.sf.net/sfu/splunk-d2dcopy2
> _______________________________________________
> gumstix-users mailing list
> gumstix-users@...
> https://lists.sourceforge.net/lists/listinfo/gumstix-users
> 
> 

-- 
View this message in context: http://old.nabble.com/WiFi-problems-tp32473984p32537024.html
Sent from the Gumstix mailing list archive at Nabble.com.

I just wanted to note that I am experiencing a delay in wlan0
appearing. It takes several minutes, 2-5, for the messages like

libertas: 00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303

to show up. After it does everything works great. I don't know if
Nick's method was attempting to solve this but it did not work for me.

I am compiled from openembedded using the over-2011.03 branch, current
as of today.

On 27 September 2011 10:56, joncoops <joncoops@...> wrote:
>
> Gerhard,
>
> As you observe it seems to take a while and then I get the
> [snip]
> libertas: 00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303
> [snip]
> message after that I can start the interface. Before this when typing
> ifconfig the interface for the WiFi does not show up.
>
> I have added a deamon process to check periodicallty for connectivity and
> restart the network interface if it cannot reach the internet. This seems
> like using a sledge hammer to crack a walnut, but it works. I am really
> curious why the WiFi hardware does not seem to come up on boot.
>
> Jon
>
>
>
> Gerhard H. wrote:
>>
>>
>> Hello,
>>
>> I guess I have the same problem on the latest console Image from
>> http://cumulus.gumstix.org/images/angstrom/developer/2011-08-30-1058/
>>
>> Sometimes during startup it lasts around 2 Minutes till the line
>> [snip]
>> libertas: 00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303
>> [snip]
>>
>> is printed to the console and
>> sometimes the startup runs through quite fluently and after some time I
>> get
>> the output as Jon was pasting it. After that it is possible to start the
>> wlan using iwconfig, but not before (see snip from Jon below).
>>
>> joncoops wrote:
>>>
>>> [snip]
>>> libertas_sdio: Copyright Pierre Ossman
>>> ....
>>> overo login: libertas: 00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303
>>>
>>
>> Jon, do you have also sometimes the situation, that "libertas:
>> 00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303" is already printed during
>> startup after around 2 Minutes?
>>
>> @Nick and Eric:
>> Either "Ifdown wlan0..." nor adjustments in "/etc/.../Interfaces" will
>> work
>> as long as this "libertas: 00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303"
>> output was not printed to the console
>>
>> Gerhard
>> --
>> View this message in context:
>> http://old.nabble.com/WiFi-problems-tp32473984p32503864.html
>> Sent from the Gumstix mailing list archive at Nabble.com.
>>
>>
>> ------------------------------------------------------------------------------
>> All of the data generated in your IT infrastructure is seriously valuable.
>> Why? It contains a definitive record of application performance, security
>> threats, fraudulent activity, and more. Splunk takes this data and makes
>> sense of it. IT sense. And common sense.
>> http://p.sf.net/sfu/splunk-d2dcopy2
>> _______________________________________________
>> gumstix-users mailing list
>> gumstix-users@...
>> https://lists.sourceforge.net/lists/listinfo/gumstix-users
>>
>>
>
> --
> View this message in context: http://old.nabble.com/WiFi-problems-tp32473984p32537024.html
> Sent from the Gumstix mailing list archive at Nabble.com.
>
>
> ------------------------------------------------------------------------------
> All the data continuously generated in your IT infrastructure contains a
> definitive record of customers, application performance, security
> threats, fraudulent activity and more. Splunk takes this data and makes
> sense of it. Business sense. IT sense. Common sense.
> http://p.sf.net/sfu/splunk-d2dcopy1
> _______________________________________________
> gumstix-users mailing list
> gumstix-users@...
> https://lists.sourceforge.net/lists/listinfo/gumstix-users
>

Hello,
any solution for this? I have a similar situation. Sometimes the Overo COM
starts but it doesn't detect the wifi until some minutes, and other times it
hangs the booting some minutes until it detects the wireless interface.
When the booting is stoping I get these messages:

....
Starting udev
udevd (74): /proc/74/oom_adj is deprecated, please use
/proc/74/oom_score_adj instead.
lib80211: common routines for IEEE802.11 drivers
cfg80211: Calling CRDA to update world regulatory domain
libertas_sdio: Libertas SDIO driver
libertas_sdio: Copyright Pierre Ossman

(... wait some minutes ..)

libertas_sdio mmc1:0001:1: (unregistered net_device): 00:19:88:3a:af:63, fw
9.70.7p0, cap 0x00000303
libertas_sdio mmc1:0001:1: wlan0: Marvell WLAN 802.11 adapter
...

I am compiled from openembedded using the over-2011.03 branch, too.

root@...:~# uname -a
Linux overo 3.0.0 #1 Tue Nov 8 17:52:03 CET 2011 armv7l GNU/Linux

Thanks.


Replic8tor wrote:
> 
> I just wanted to note that I am experiencing a delay in wlan0
> appearing. It takes several minutes, 2-5, for the messages like
> 
> libertas: 00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303
> 
> to show up. After it does everything works great. I don't know if
> Nick's method was attempting to solve this but it did not work for me.
> 
> I am compiled from openembedded using the over-2011.03 branch, current
> as of today.
> 
> On 27 September 2011 10:56, joncoops <joncoops@...> wrote:
>>
>> Gerhard,
>>
>> As you observe it seems to take a while and then I get the
>> [snip]
>> libertas: 00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303
>> [snip]
>> message after that I can start the interface. Before this when typing
>> ifconfig the interface for the WiFi does not show up.
>>
>> I have added a deamon process to check periodicallty for connectivity and
>> restart the network interface if it cannot reach the internet. This seems
>> like using a sledge hammer to crack a walnut, but it works. I am really
>> curious why the WiFi hardware does not seem to come up on boot.
>>
>> Jon
>>
>>
>>
> 

-- 
View this message in context: http://old.nabble.com/WiFi-problems-tp32473984p32810337.html
Sent from the Gumstix mailing list archive at Nabble.com.

Hello everybody,

The same for the last prebuild image from the Gumstix repository (for 
omap3-console-image), that use the kernel 3.0 too.
How can I stabilize this situation?
I boot and I don't have wifi until 1-5 minutes later.

Thanks a lot.


----- Original Message ----- 
From: "Víctor Andrés" <victor@...>
To: <gumstix-users@...>
Sent: Wednesday, November 09, 2011 12:45 PM
Subject: Re: [Gumstix-users] WiFi problems


>
> Hello,
> any solution for this? I have a similar situation. Sometimes the Overo COM
> starts but it doesn't detect the wifi until some minutes, and other times 
> it
> hangs the booting some minutes until it detects the wireless interface.
> When the booting is stoping I get these messages:
>
> ....
> Starting udev
> udevd (74): /proc/74/oom_adj is deprecated, please use
> /proc/74/oom_score_adj instead.
> lib80211: common routines for IEEE802.11 drivers
> cfg80211: Calling CRDA to update world regulatory domain
> libertas_sdio: Libertas SDIO driver
> libertas_sdio: Copyright Pierre Ossman
>
> (... wait some minutes ..)
>
> libertas_sdio mmc1:0001:1: (unregistered net_device): 00:19:88:3a:af:63, 
> fw
> 9.70.7p0, cap 0x00000303
> libertas_sdio mmc1:0001:1: wlan0: Marvell WLAN 802.11 adapter
> ...
>
> I am compiled from openembedded using the over-2011.03 branch, too.
>
> root@...:~# uname -a
> Linux overo 3.0.0 #1 Tue Nov 8 17:52:03 CET 2011 armv7l GNU/Linux
>
> Thanks.
>
>
> Replic8tor wrote:
>>
>> I just wanted to note that I am experiencing a delay in wlan0
>> appearing. It takes several minutes, 2-5, for the messages like
>>
>> libertas: 00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303
>>
>> to show up. After it does everything works great. I don't know if
>> Nick's method was attempting to solve this but it did not work for me.
>>
>> I am compiled from openembedded using the over-2011.03 branch, current
>> as of today.
>>
>> On 27 September 2011 10:56, joncoops <joncoops@...> wrote:
>>>
>>> Gerhard,
>>>
>>> As you observe it seems to take a while and then I get the
>>> [snip]
>>> libertas: 00:19:88:3e:7e:d2, fw 9.70.7p0, cap 0x00000303
>>> [snip]
>>> message after that I can start the interface. Before this when typing
>>> ifconfig the interface for the WiFi does not show up.
>>>
>>> I have added a deamon process to check periodicallty for connectivity 
>>> and
>>> restart the network interface if it cannot reach the internet. This 
>>> seems
>>> like using a sledge hammer to crack a walnut, but it works. I am really
>>> curious why the WiFi hardware does not seem to come up on boot.
>>>
>>> Jon
>>>
>>>
>>>
>>
>
> -- 
> View this message in context: 
> http://old.nabble.com/WiFi-problems-tp32473984p32810337.html
> Sent from the Gumstix mailing list archive at Nabble.com.
>
>
> ------------------------------------------------------------------------------
> RSA(R) Conference 2012
> Save $700 by Nov 18
> Register now
> http://p.sf.net/sfu/rsa-sfdev2dev1
> _______________________________________________
> gumstix-users mailing list
> gumstix-users@...
> https://lists.sourceforge.net/lists/listinfo/gumstix-users