Re: Logjam vulnerability and epiphany.

[James Cameron]

Thanks for the question.

We have no plans to declare end-of-life or end-of-support for the older XO's.

As far as I can tell, in making such a declaration there would be no
net benefit, and some additional cost.

There are very few of the older XO's in use.

On Mon, Aug 31, 2020 at 05:48:12PM -0400, Samuel Greenfeld wrote:
> While it is possible to partially fix the LogJam vulnerability, are there 
> plans
> to declare end-of-life or end-of-support for the older XO's?
> 
> A few years ago, we talked about ending support in 2020 when other Linux
> distributions would stop distributing x86 processor builds.
> 
> And OLPC is not like the OpenWRT project, where developers have kept things
> up-to-date even for legacy hardware.
> 
> ---
> SJG
> 
> On Sun, Aug 30, 2020 at 5:21 PM James Cameron <[1]qu...@laptop.org> wrote:
> 
> Yes, wikimedia servers will be unreachable.  That is their decision.
> 
> Many other servers already do not permit connection, and this varies by
> geography of IP address, based on local laws or a server owner's 
> assessment
> of market value.
> 
> We have no plans to upgrade for XO-1.5.
> 
> Is not a problem on NL3 or ED20 models with OLPC OS 20.04.
> 
> On Sun, Aug 30, 2020 at 06:57:09AM -0700, [2]pe...@easthope.ca wrote:
> > Hi,
> >
> > System 13.2.8 on a XO 1.5 here.
> >
> > Open epiphany and open
> > [3]https://clienttest.ssllabs.com:8443/ssltest/viewMyClient.html .
> >
> > Logjam vulnerability is reported. 
> > [4]https://en.wikipedia.org/wiki/Logjam_(computer_security)
> >
> > A specific consequence is that a wikimedia server will not permit a
> > connection from this epiphany after September 25, 2020. Ie. Wikipedia
> > & etc. will become inaccessible with this browser.  =8~(
> >
> > I guess you fellas are aware of all that.  Is there any plan to
> > introduce an appropriate upgrade?
> >
> > Thanks,                              ... Peter E.
> >
> > --
> > [5]https://en.wikibooks.org/wiki/Medical_Machines
> > [6]https://en.wikibooks.org/wiki/Oberon
> > Tel: +1 604 670 0140            Bcc: peter at easthope. ca
> >
> > ___
> > Devel mailing list
> > [7]Devel@lists.laptop.org
> > [8]http://lists.laptop.org/listinfo/devel
> 
> --
> James Cameron
> [9]http://quozl.netrek.org/
> ___
> Devel mailing list
> [10]Devel@lists.laptop.org
> [11]http://lists.laptop.org/listinfo/devel
> 
> References:
> 
> [1] mailto:qu...@laptop.org
> [2] mailto:pe...@easthope.ca
> [3] https://clienttest.ssllabs.com:8443/ssltest/viewMyClient.html
> [4] https://en.wikipedia.org/wiki/Logjam_(computer_security)
> [5] https://en.wikibooks.org/wiki/Medical_Machines
> [6] https://en.wikibooks.org/wiki/Oberon
> [7] mailto:Devel@lists.laptop.org
> [8] http://lists.laptop.org/listinfo/devel
> [9] http://quozl.netrek.org/
> [10] mailto:Devel@lists.laptop.org
> [11] http://lists.laptop.org/listinfo/devel

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Logjam vulnerability and epiphany.

[James Cameron]

Yes, wikimedia servers will be unreachable.  That is their decision.

Many other servers already do not permit connection, and this varies by 
geography of IP address, based on local laws or a server owner's assessment of 
market value.

We have no plans to upgrade for XO-1.5.

Is not a problem on NL3 or ED20 models with OLPC OS 20.04.

On Sun, Aug 30, 2020 at 06:57:09AM -0700, pe...@easthope.ca wrote:
> Hi,
> 
> System 13.2.8 on a XO 1.5 here.
> 
> Open epiphany and open 
> https://clienttest.ssllabs.com:8443/ssltest/viewMyClient.html .
> 
> Logjam vulnerability is reported.  
> https://en.wikipedia.org/wiki/Logjam_(computer_security)
> 
> A specific consequence is that a wikimedia server will not permit a 
> connection from this epiphany after September 25, 2020. Ie. Wikipedia 
> & etc. will become inaccessible with this browser.  =8~(
> 
> I guess you fellas are aware of all that.  Is there any plan to 
> introduce an appropriate upgrade?
> 
> Thanks,  ... Peter E.
> 
> -- 
> https://en.wikibooks.org/wiki/Medical_Machines
> https://en.wikibooks.org/wiki/Oberon
> Tel: +1 604 670 0140Bcc: peter at easthope. ca
> 
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Hardware test in 1.5 reports temperature rise of 9 or 10 C.

[James Cameron]

Thanks, that's interesting.  This serial number was produced in late
January 2011, about 9.5 years ago.  I would have expected it to fail
much sooner.  It has lasted well.

The temperature rise test was only characterised for manufacturing and
some accelerated aging tests.  The test has no statistical meaning now.
The test should be done at an ambient temperature of 21°C to avoid
false positives.

In my experience the most likely causes of the test failure are;

- high ambient temperature due to season,

- high internal temperature due to repeated testing,

- missing or loose screws holding down the heat spreader,

- dents in heat spreader due to impact,

- age hardening of the silicone plastic thermal pad between the heat
  spreader and the CPU.

I suggest ignoring the problem for now, as servicing may cause damage
beyond economic repair.  In particular for the CPU interposer board
solder balls, which are under significant stress.

On Mon, Aug 10, 2020 at 06:07:02AM -0700, pe...@easthope.ca wrote:
> Hi,
> 
> P.s. This is SKU 133, D5.  
> Present in http://wiki.laptop.org/go/Manufacturing_data .
> 
> The 1.5 D5 here is S/N SHC10400772  OpenFirmware Q3C17  EC Firmware 
> Ver:2.2.10.
> 
> In the hardware test, a temperature rise of 9-10 C is reported.
> 
> I found a thread in the mailing list at 2011-2012.  Appears hardware 
> revision might have continued then.  I haven't found a description of 
> a repair.
> 
> I have no trouble dismantling the machine and performing a small 
> modification or repair.  Before diving in, I am interested in advice 
> and lore.
> 
> Aside from this thermal problem, the machine appears to run as well as 
> when it left the factory.
> 
> Thanks!  ... Peter E.
> 
> -- 
> https://en.wikibooks.org/wiki/Medical_Machines
> https://en.wikibooks.org/wiki/Oberon
> Tel: +1 604 670 0140Bcc: peter at easthope. ca
> 
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: XO1.75

[James Cameron]

Based on memory, 00 means fault in RAM.  Replace RAM.  Could also be 
electrostatic damage to CPU or RAM.  Could also be age degradation of silicon 
chips.

To find the real meaning of 00 takes reading the C code of the CForth startup 
payload, which is loaded from SPI Flash.

There is no involvement of eMMC or SD card slots.

On Sat, Jun 06, 2020 at 11:41:33AM +, Yioryos Asprobounitis wrote:
> Hi list
> I dug out my XOs to check their condition and my XO-1.75 (SKU203) refuses to 
> boot as if it was wiped out clean or has no microSD card at all.
> The screen comes up and nothing further.
> Firmware check just show a double zero [00] 
> Trying to install a fresh OS from USB fails (with the same double zero)
> A fast wiki search did not come up with anything relevant.
> If someone can point me to the right direction I would appreciate it.
> Thanks
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Pinebook 1080p with Sugar 0.117 and Ubuntu 18.04 Bionic for arm64

[James Cameron]

I've received a donation of a 11.6" Pinebook Linux Laptop from the
PINE64 Community.

This sub-$100 laptop performs nicely for the price.

It can be charged over USB.

The laptop came with Ubuntu 18.04 and KDE Neon.  A very functional
desktop environment.

Sugar 0.117 can be used.  The Sugar packages for OLPC OS on Ubuntu
20.04 were rebuilt on the Pinebook and then installed.

Some performance timings;

1.  boot time to login prompt, 30 seconds,

2.  login to Sugar, 13 seconds,

3.  wake from sleep to Sugar, 2 seconds,

4.  activity start, 5 seconds.

References:

https://en.wikipedia.org/wiki/Pinebook
https://www.pine64.org/pinebook/

-- 
James Cameron
http://quozl.netrek.org/


signature.asc
Description: PGP signature
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Announcing OLPC OS 13.2.11

[James Cameron]

G'day,

We're pleased to announce the release of OLPC OS 13.2.11 for XO-1,
XO-1.5, XO-1.75 and XO-4.

It is Sugar 0.112 on Fedora 18, with updated activities Browse-157.5,
Calculate-45.1, Chat-85.1, Clock-21, Distance-36, ImageViewer-64.1,
Implode-19.1, Log-40.1, Maze-28.2, Measure-53.1, Memorize-55.1,
Moon-18.1, MusicKeyboard-8.10, Paint-69, Pippy-72.3, Read-118.2,
Record-104, Speak-52.1, StopWatch-20.2, Terminal-46.2, and Write-100.

Details of fixes, known issues, and how to download, install or
upgrade are in the release notes:

http://wiki.laptop.org/go/Release_notes/13.2.11

Our thanks to all contributors, testers, upstreams, and those who
have provided feedback of any kind.

Build file names: 32023o0 32023o1 32023o2 32023o4

-- 
James Cameron
http://quozl.netrek.org/


signature.asc
Description: PGP signature
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: XO as wireless remote terminal for Internet-in-a-box on Raspberry Pi Zero W

[James Cameron]

Best ask Internet-in-a-Box project about minimum browser requirements;
as far as I know they don't participate here on devel@ or maintain
IIAB for use with XOs.

On Tue, Dec 10, 2019 at 05:41:57PM -0500, Carrol Riddle wrote:
> sudo apt install dillo   works. (did first removed files from download of 
> tarball method)
> Website http://wiki.laptop.org rendering is usable. 
> Internet-in-a-Box home page is largely not usable.  Presumably, this is due 
> to use of Javascript pages.
> Will try other lite browsers which claim to handle javascript.
> 
> > On December 10, 2019 at 2:28 PM James Cameron  wrote:
> > 
> > 
> > Have you tried "sudo apt install dillo"?
> > 
> > On Tue, Dec 10, 2019 at 02:19:12PM -0500, Carrol Riddle wrote:
> > > James,
> > > Thanks for intro to field of "lite" browsers -- different from tried 
> > > links2.
> > > Not successful with Dillo yet. Make for fltk, a prerequisite, fails with 
> > > errors (configure errors ?).
> > > Will also look at others.
> > > Carrol Riddle
> > > 
> > > > On December 9, 2019 at 1:07 AM James Cameron  wrote:
> > > > 
> > > > 
> > > > Most web content demands large browser like Chromium or Firefox.
> > > > 
> > > > But where the web content is your own, or under your control, then a
> > > > simplified web browser like Dillo may be helpful.
> > > > 
> > > > On Mon, Dec 09, 2019 at 12:13:36AM -0500, Carrol Riddle wrote:
> > > > > XO as wireless remote terminal for Internet-in-a-box on Raspberry Pi 
> > > > > Zero W using X11 forwarding.
> > > > > 
> > > > > XO usb port can be used to power Pi Zero W.  XO can be used to 
> > > > > prepare >64 GB SD by installing Exfat (per earlier post).
> > > > > 
> > > > > Headless Pi Zero can be set up wirelessly with standard method of 
> > > > > connecting to a wireless router.
> > > > > A prepared Raspbian SD is edited to place empty SSH file in /boot to 
> > > > > enable SSH and place a model wpa_supplicant file in /boot.  The 
> > > > > wpa_supplicant is edit to provide ssid or ip address of router and 
> > > > > its password. The Pi Zero W is functioning as an AP client.
> > > > > 
> > > > > Connection to the Pi Zero W is made using its host name (raspberrypi, 
> > > > > password raspberry) or ip address (192.168.128.4) with a computer 
> > > > > (XO) also connected to the wireless router.  
> > > > > 
> > > > > The wireless router is connected to an internet AP (in my case 
> > > > > broadband AP/wireless router).  Raspbian Desktop is loaded and then 
> > > > > Internet-in-a-Box (IIAB) is loaded.  iiab-hotspot-on command is run 
> > > > > to enable IIAB's AP.  This also disables the connection to the 
> > > > > wireless router and the only access to the IIAB becomes its AP and 
> > > > > there is no internet connection (in spite of enabling internet on 
> > > > > gateway).
> > > > > 
> > > > > If one needs to connect to the internet to obtain more content, then 
> > > > > iiab-hotspot-off . The Pi Zero W can then be access with an XO (same 
> > > > > XO as for power) over its usb port, which provides internet access.  
> > > > > XO setup  and accessing Chromium in the Raspbian Desktop has been 
> > > > > described in an earlier post. IIAB is run in Chromium to access the 
> > > > > Admin Console to manage the content.
> > > > > 
> > > > > Return to normal use of IIAB by running iiab-hotspot-on .   The Pi 
> > > > > Zero W can be moved to any source of usb power.
> > > > > 
> > > > > Connect XO wirelessly to IIAB AP.   SSH / desktop with ssh -X 
> > > > > pi@box.local /etc/X11/xinit/xintrc
> > > > > Run Chrommium with URL of http://box.local.   This can be simplified 
> > > > > by setting startup page in Chromium to http://box.local/home  .
> > > > > 
> > > > > Chromium is very, very slow, but IIAB is at XO speed.
> > > > > 
> > > > > If one wants to otherwise use Chromium, enable Home key in task bar 
> > > > > and assign it to Chrome://
> > > > > 
> > > > > Direct Chromium access with  works 
> > > > > when 

Re: XO as wireless remote terminal for Internet-in-a-box on Raspberry Pi Zero W

[James Cameron]

Have you tried "sudo apt install dillo"?

On Tue, Dec 10, 2019 at 02:19:12PM -0500, Carrol Riddle wrote:
> James,
> Thanks for intro to field of "lite" browsers -- different from tried links2.
> Not successful with Dillo yet. Make for fltk, a prerequisite, fails with 
> errors (configure errors ?).
> Will also look at others.
> Carrol Riddle
> 
> > On December 9, 2019 at 1:07 AM James Cameron  wrote:
> > 
> > 
> > Most web content demands large browser like Chromium or Firefox.
> > 
> > But where the web content is your own, or under your control, then a
> > simplified web browser like Dillo may be helpful.
> > 
> > On Mon, Dec 09, 2019 at 12:13:36AM -0500, Carrol Riddle wrote:
> > > XO as wireless remote terminal for Internet-in-a-box on Raspberry Pi Zero 
> > > W using X11 forwarding.
> > > 
> > > XO usb port can be used to power Pi Zero W.  XO can be used to prepare 
> > > >64 GB SD by installing Exfat (per earlier post).
> > > 
> > > Headless Pi Zero can be set up wirelessly with standard method of 
> > > connecting to a wireless router.
> > > A prepared Raspbian SD is edited to place empty SSH file in /boot to 
> > > enable SSH and place a model wpa_supplicant file in /boot.  The 
> > > wpa_supplicant is edit to provide ssid or ip address of router and its 
> > > password. The Pi Zero W is functioning as an AP client.
> > > 
> > > Connection to the Pi Zero W is made using its host name (raspberrypi, 
> > > password raspberry) or ip address (192.168.128.4) with a computer (XO) 
> > > also connected to the wireless router.  
> > > 
> > > The wireless router is connected to an internet AP (in my case broadband 
> > > AP/wireless router).  Raspbian Desktop is loaded and then 
> > > Internet-in-a-Box (IIAB) is loaded.  iiab-hotspot-on command is run to 
> > > enable IIAB's AP.  This also disables the connection to the wireless 
> > > router and the only access to the IIAB becomes its AP and there is no 
> > > internet connection (in spite of enabling internet on gateway).
> > > 
> > > If one needs to connect to the internet to obtain more content, then 
> > > iiab-hotspot-off . The Pi Zero W can then be access with an XO (same XO 
> > > as for power) over its usb port, which provides internet access.  XO 
> > > setup  and accessing Chromium in the Raspbian Desktop has been described 
> > > in an earlier post. IIAB is run in Chromium to access the Admin Console 
> > > to manage the content.
> > > 
> > > Return to normal use of IIAB by running iiab-hotspot-on .   The Pi Zero W 
> > > can be moved to any source of usb power.
> > > 
> > > Connect XO wirelessly to IIAB AP.   SSH / desktop with ssh -X 
> > > pi@box.local /etc/X11/xinit/xintrc
> > > Run Chrommium with URL of http://box.local.   This can be simplified by 
> > > setting startup page in Chromium to http://box.local/home  .
> > > 
> > > Chromium is very, very slow, but IIAB is at XO speed.
> > > 
> > > If one wants to otherwise use Chromium, enable Home key in task bar and 
> > > assign it to Chrome://
> > > 
> > > Direct Chromium access with  works 
> > > when accessed over the usb, but does not work with IIAB AP (something 
> > > about Xlib).
> > > 
> > > A simple Sugar Activity can run a script to handle all this startup of 
> > > ssh / chromium / iiab .
> > > 
> > > My question for Devel list:   is there a simpler (faster) application 
> > > than desktop Chromium for displaying web pages from Terminal ??
> > > 
> > > links2 is one such application, but I have not been successful in setting 
> > > up options for x display.
> > > 
> > > 
> > > Carrol Riddle
> > > ___
> > > Devel mailing list
> > > Devel@lists.laptop.org
> > > http://lists.laptop.org/listinfo/devel
> > 
> > -- 
> > James Cameron
> > http://quozl.netrek.org/
> > ___
> > Devel mailing list
> > Devel@lists.laptop.org
> > http://lists.laptop.org/listinfo/devel
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: XO as wireless remote terminal for Internet-in-a-box on Raspberry Pi Zero W

[James Cameron]

Most web content demands large browser like Chromium or Firefox.

But where the web content is your own, or under your control, then a
simplified web browser like Dillo may be helpful.

On Mon, Dec 09, 2019 at 12:13:36AM -0500, Carrol Riddle wrote:
> XO as wireless remote terminal for Internet-in-a-box on Raspberry Pi Zero W 
> using X11 forwarding.
> 
> XO usb port can be used to power Pi Zero W.  XO can be used to prepare >64 GB 
> SD by installing Exfat (per earlier post).
> 
> Headless Pi Zero can be set up wirelessly with standard method of connecting 
> to a wireless router.
> A prepared Raspbian SD is edited to place empty SSH file in /boot to enable 
> SSH and place a model wpa_supplicant file in /boot.  The wpa_supplicant is 
> edit to provide ssid or ip address of router and its password. The Pi Zero W 
> is functioning as an AP client.
> 
> Connection to the Pi Zero W is made using its host name (raspberrypi, 
> password raspberry) or ip address (192.168.128.4) with a computer (XO) also 
> connected to the wireless router.  
> 
> The wireless router is connected to an internet AP (in my case broadband 
> AP/wireless router).  Raspbian Desktop is loaded and then Internet-in-a-Box 
> (IIAB) is loaded.  iiab-hotspot-on command is run to enable IIAB's AP.  This 
> also disables the connection to the wireless router and the only access to 
> the IIAB becomes its AP and there is no internet connection (in spite of 
> enabling internet on gateway).
> 
> If one needs to connect to the internet to obtain more content, then 
> iiab-hotspot-off . The Pi Zero W can then be access with an XO (same XO as 
> for power) over its usb port, which provides internet access.  XO setup  and 
> accessing Chromium in the Raspbian Desktop has been described in an earlier 
> post. IIAB is run in Chromium to access the Admin Console to manage the 
> content.
> 
> Return to normal use of IIAB by running iiab-hotspot-on .   The Pi Zero W can 
> be moved to any source of usb power.
> 
> Connect XO wirelessly to IIAB AP.   SSH / desktop with ssh -X pi@box.local 
> /etc/X11/xinit/xintrc
> Run Chrommium with URL of http://box.local.   This can be simplified by 
> setting startup page in Chromium to http://box.local/home  .
> 
> Chromium is very, very slow, but IIAB is at XO speed.
> 
> If one wants to otherwise use Chromium, enable Home key in task bar and 
> assign it to Chrome://
> 
> Direct Chromium access with  works when 
> accessed over the usb, but does not work with IIAB AP (something about Xlib).
> 
> A simple Sugar Activity can run a script to handle all this startup of ssh / 
> chromium / iiab .
> 
> My question for Devel list:   is there a simpler (faster) application than 
> desktop Chromium for displaying web pages from Terminal ??
> 
> links2 is one such application, but I have not been successful in setting up 
> options for x display.
> 
> 
> Carrol Riddle
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: XO's cannot use mirror repo's in YUM update or install

[James Cameron]

Thanks for the problem report and workaround.

The cause is an SSLv3 Handshake Failure, apparently a result of
tightened security configuration at fedoraproject.org which is no
longer compatible with Fedora 18.

Unfortunately yum does not report the actual problem.

Here's how to catch proof;

1.  use tcpdump to capture network packets and then wireshark to
analyse,

2.  look for the "Alert (Level: Fatal, Description: Handshake
Failure)",

3.  look for the immediately preceeding SSLv3 Client Hello message,

4.  note the Cipher Suites list contains some that are no longer
acceptable.

Your workaround is fine.  It is similar to the one I used for XO-1.75
and XO-4 in 13.2.8;
https://github.com/quozl/olpc-os-builder/commit/f2cb3908aff0cc7bc3ba7937a93b0337140dd81e

Another workaround is to change from https to http in the mirrorlist
entries.

sudo sed -i 's/mirrorlist=https/mirrorlist=http/g' /etc/yum.repos.d/*.repo

However, while this is faster, it also lowers the overall security
because it makes a man in the middle attack easier.

Best way to image a set of laptops with rpmfusion packages is to build
an image using olpc-os-builder.  I've got notes on how to do that.

On Sun, Nov 10, 2019 at 03:15:55PM -0500, Carrol Riddle wrote:
> Have been able to Yum install exfat files on my XO-1, but everywhere had to 
> block mirrorline and use baseurl. 
> 
> Still do not know why mirrors do not work.
> 
> The baseurl for fedora.repo is 
> http://dl.fedoraproject.org/pub/archive/fedora/linux/$releasever/Everything/$basearch/os/
> 
> The baseurl for rpmfusion is 
> http://archive.rpmfusion.org/free-archive/fedora/releases/$releasever/Everything/$basearch/os/
> 
> dl, download and archive all seem to work as first term in fedora path.
> 
> Modified the *-update.repo files similarly (but not same).
> 
> I had been using http://wiki.laptop.org/go/Gstreamer  method of installing 
> rpmfusion,
> but simpler and newer is:
> wget -c 
> download1.rpmfusion.org/free/fedora/rpmfusion-free-release-18.noarch.rpm
> and rpm  -i rpmfusion-free-release-18.noarch.rpm
> 
> Removed extraneous rpmfusion  repos from /etc/yum.repos.d/
> 
> This effort was to allow installing Internet-in-a-Box on a larger SD for 
> Raspberry Pi Zero W using only XO and the Zero.
> Two external ports are needed and had previously used Pi 4 to prepare SD.
> The single USB port on Zero is used for the connection to an XO using 
> X11Forwarding for display, keyboard and shared WiFi (secondary to Zero W 
> on-board WiFi or as primary for simple Zero).
> 
> Still looking for cause of YUM Mirrors failure.
> 
> Carrol Riddle
> 
> 
> > On November 10, 2019 at 9:57 AM Peter Robinson  wrote:
> > 
> > 
> > On Sun, Nov 10, 2019 at 5:29 AM Carrol Riddle  wrote:
> > >
> > > XO's attempting to run YUM update or install are unable to use fedora 
> > > mirror sites (https://) but able to use primary fedora site (http://).
> > >
> > > Is this a matter of https vs http / ca-certificates or changes in mirror 
> > > structures ?  Ca-certificates update have not been done, but could be 
> > > done.
> > >
> > >  Running OLPC 13.2.10 with current date / time and hwclock -w to sync.
> > >
> > > Primaries used by editing /etc/yum.repos.d/fedora.repo and commenting out 
> > > mirrorlist line and uncommenting baseurl line (and adding "archive" to 
> > > url path after /pub/).
> > >
> > > There are no entries in yum.log and error message is:
> > > "Cannot retrieve metalink for repository: fedora/18/i386.  Please verify 
> > > its path and try again."
> > >
> > > My specific case is trying to install rpmfusion in preparing to install 
> > > exfat-utils and fuse-exfat ,  but occurs with other installs that have 
> > > been done in the past.
> > 
> > I'm guessing you might need to update for content that has been
> > archived, I thought the mirror manager dealt with redirects
> > automatically there but I don't know exactly.
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: OpenFirmware and Linux v5.0 on XO-1.75

[James Cameron]

On Sun, Apr 21, 2019 at 12:02:56PM +0200, Lubomir Rintel wrote:
> On Sat, 2019-04-20 at 18:08 +1000, James Cameron wrote:
> > On Fri, Apr 19, 2019 at 08:58:15AM +0200, Lubomir Rintel wrote:
> > > On Mon, 2019-02-25 at 20:00 +1100, James Cameron wrote:
> > > > On Sat, Feb 23, 2019 at 06:06:56PM +0100, Lubomir Rintel wrote:
> > > > > On Sat, 2019-02-23 at 10:52 +1100, James Cameron wrote:
> > > > > > On Fri, Feb 22, 2019 at 12:14:00PM +0100, Lubomir Rintel wrote:
> > > > > > > Also, which config? Mine is basically this:
> > > > > > > https://raw.githubusercontent.com/hackerspace/olpc-xo175-linux/lr/olpc-xo175/arch/arm/configs/olpc_xo175_defconfig
> > > > > > 
> > > > > > The config used by Fedora.
> > > > > > 
> > > > > > Can we work toward some kind of reproducible build?
> > > > > 
> > > > > Not sure what you mean here.
> > > > 
> > > > Sorry.  Configuring and building kernels is for me a rare thing to do,
> > > > and whenever I try I'm usually interrupted by something more urgent,
> > > > as I've quite a few other things I've got to do.  I'm still a newbie
> > > > at it because I can't dedicate the time.  [...]
> > > 
> > > So, over the last week or so, I spent some effort making this work with
> > > the OLPC RPM build tooling. Here's what I came up with:
> > > 
> > > http://v3.sk/~lkundrak/olpc/green_ears.jpeg
> > > [...]
> > > Firmware:
> > > 
> > >   git pull https://github.com/lkundrak/openfirmware/ lr/olpc-xo175-3
> > > 
> > > A couple more small fixups here and there since lr/olpc-xo175-2. The
> > > most notable fix is for a regression that caused the RTC to be cleared
> > > on each boot.
> > 
> > Pushed as
> > https://github.com/quozl/openfirmware/commits/lr/olpc-xo175-3
> > 
> > Released q4e00ja.rom from this as is;
> > http://dev.laptop.org/~quozl/q4e00ja.rom
> > 
> > Removed the dtcompat.fth fload and released q4e01ja.rom;
> > http://dev.laptop.org/~quozl/q4e01ja.rom
> 
> This second image removes the ablity to boot the legacy OLPC OS kernel,
> doesn't it?

For secure boot, yes.  But you could still use root=.

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: OpenFirmware and Linux v5.0 on XO-1.75

[James Cameron]

On Fri, Apr 19, 2019 at 08:58:15AM +0200, Lubomir Rintel wrote:
> On Mon, 2019-02-25 at 20:00 +1100, James Cameron wrote:
> > On Sat, Feb 23, 2019 at 06:06:56PM +0100, Lubomir Rintel wrote:
> > > On Sat, 2019-02-23 at 10:52 +1100, James Cameron wrote:
> > > > On Fri, Feb 22, 2019 at 12:14:00PM +0100, Lubomir Rintel wrote:
> > > > > Also, which config? Mine is basically this:
> > > > > https://raw.githubusercontent.com/hackerspace/olpc-xo175-linux/lr/olpc-xo175/arch/arm/configs/olpc_xo175_defconfig
> > > > 
> > > > The config used by Fedora.
> > > > 
> > > > Can we work toward some kind of reproducible build?
> > > 
> > > Not sure what you mean here.
> > 
> > Sorry.  Configuring and building kernels is for me a rare thing to do,
> > and whenever I try I'm usually interrupted by something more urgent,
> > as I've quite a few other things I've got to do.  I'm still a newbie
> > at it because I can't dedicate the time.  [...]
> 
> So, over the last week or so, I spent some effort making this work with
> the OLPC RPM build tooling. Here's what I came up with:
> 
> http://v3.sk/~lkundrak/olpc/green_ears.jpeg

Thanks, that's fantastic.  I've mostly reproduced your work, and have
my build of your 5.0 kernel running on an XO-1.75 with an adjusted
Fedora 18 user space and a fixed root= argument.

> The kernel:
> 
>   git pull https://github.com/hackerspace/olpc-xo175-linux/ olpc-5.0
> 
> Based on vanilla v5.0.8, a few parts taken from olpc-4.8,
> olpc-3.0-arm, along with my defconfig and a couple of changes to
> support cross-build on Fedora 30. I've not tested native builds, I
> didn't dare to run it on the XO.

Pushed as http://dev.laptop.org/git/olpc-kernel/log/?h=olpc-5.0

My dmesg;
http://dev.laptop.org/~quozl/z/1hHiN5.txt

I used an Ubuntu 18.04 cross-build to make the kernel I'm using at the
moment, as our production builder is Fedora 18.  I've started to
iterate on a Fedora 30 builder, but I'm not sure I've got everything I
need.  Here's the packages I'm adding;

gcc-arm-linux-gnu binutils-arm-linux-gnu
rpm-build bison flex m4 make openssl-devel perl

> Firmware:
> 
>   git pull https://github.com/lkundrak/openfirmware/ lr/olpc-xo175-3
> 
> A couple more small fixups here and there since lr/olpc-xo175-2. The
> most notable fix is for a regression that caused the RTC to be cleared
> on each boot.

Pushed as
https://github.com/quozl/openfirmware/commits/lr/olpc-xo175-3

Released q4e00ja.rom from this as is;
http://dev.laptop.org/~quozl/q4e00ja.rom

Removed the dtcompat.fth fload and released q4e01ja.rom;
http://dev.laptop.org/~quozl/q4e01ja.rom

As the firmware will update on a standard system before the kernel
will boot, this seems an okay way to do it.  What do you think?

>   git pull https://github.com/lkundrak/dracut-modules-olpc/ master
> 
> There's a fix for assumptions about the mmc controller
> numbering. Also, to boot a FDT-based kernel the initramfs needs to
> avoid triggering the compat boot path (that lies about bootpath and
> disables the DT flattening).

Thanks.  I'm yet to use this, but plan to.

>   git pull https://github.com/lkundrak/olpc-utils/ v5.0
>   git pull https://github.com/lkundrak/olpc-utils/ master
> 
> Fixes the X11 video.

Thanks, yes, it does work, though I had to recreate the xorg.conf.d
symlink, not sure why.

Pushed as http://dev.laptop.org/git/projects/olpc-utils/log/?h=v5.1

Packaged as
http://dev.laptop.org/~quozl/olpc-utils-5.1.0-0.olpc.armv7hl.rpm
http://dev.laptop.org/~quozl/olpc-utils-5.1.0-0.olpc.src.rpm

> The patched packages are here: http://v3.sk/~lkundrak/olpc/
> The dracut-modules-olpc package needs to be installed prior to the
> kernel.

Oops, I should have read all this way before acting.  I got caught up
in code review.  Sorry.  I'll do another test using your binaries.

By the way, there's an interesting symptom on WiFi, a variable latency
on inbound ssh, also shows up as a latency staircase effect in
outbound "ping -n -i 0.200".

Also, power is not turned off on system halt.  I remember fixing that
once, so no biggie.

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: XO 1.75 mainlining status

[James Cameron]

On Fri, Apr 19, 2019 at 04:23:28PM +0200, Lubomir Rintel wrote:
> I'm wondering how the clocks are determined for the GC860.
> That should be the APMU + 0xcc register (0xd42828cc)

Got it.

> Sources [1] and [2] suggest that there is a multiplexer that chooses
> between PLL1, PLL2 and USB PLL along with a divisor and that there are
> separate bus and GPU core clocks. It is not clear to me which bits
> control which clock.

Bit 12 is a clock select,

bits 7:6 are clock source multiplexor,

when bit 12 is 0, bits 6:7 have meaning; 0x0 selects PLL1 divided by
two, 0x1 is PLL1 divided by three, 0x2 is PLL2, 0x3 is PLL2 divided by
three.

when bit 12 is 1, bits 6:7 have meaning; 0x0 selects PLL2 divided by
four, 0x1 USB PLL, remainder reserved.

> The Etnaviv driver also expects another "shader" clock for the core
> that executes the 3D command buffers. Apart from the meaning of the
> bits set in gc800_clk_enable(), I wonder what is the purpose of the
> udelay()s. I've noticed that if I just set the register as a whole
> the board sometimes hangs and I'm wondering why could that be.

Other bits in the register let me put a story to this;

1.  the module is powered up (3<<9), and the peripheral (1<<3) and AXI
bus clocks (1<<2) are turned on, then there is a udelay(150),

2.  isolation is disabled (1<<8), and then there is a udelay(1),

3.  reset is released on the AXI2MC interface (1<<15), reset is
released on the GC controller (1<<0), and then there is a udelay(100),

4   reset is released on the GPU AXI (1<<1), and then there is a
udelay(100).

Don't know why the delays are needed, sorry.

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: OpenFirmware and Linux v5.0 on XO-1.75

[James Cameron]

On Sat, Feb 23, 2019 at 06:06:56PM +0100, Lubomir Rintel wrote:
> On Sat, 2019-02-23 at 10:52 +1100, James Cameron wrote:
> > On Fri, Feb 22, 2019 at 12:14:00PM +0100, Lubomir Rintel wrote:
> > > Also, which config? Mine is basically this:
> > > https://raw.githubusercontent.com/hackerspace/olpc-xo175-linux/lr/olpc-xo175/arch/arm/configs/olpc_xo175_defconfig
> > 
> > The config used by Fedora.
> > 
> > Can we work toward some kind of reproducible build?
> 
> Not sure what you mean here.

Sorry.  Configuring and building kernels is for me a rare thing to do,
and whenever I try I'm usually interrupted by something more urgent,
as I've quite a few other things I've got to do.  I'm still a newbie
at it because I can't dedicate the time.

http://dev.laptop.org/git/olpc-kernel/commit/?h=olpc-4.8=4731695ff517ccb145e60d68acd2f7f15eb4ab6b
is an example patch which, in addition to a few irrelevant changes,
adds our OLPC RPM build process;

- our defconfig,

- a spec file for rpmbuild,

- a build script,

- an openfirmware boot script.

Some of this may have bit-rotted.

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: OpenFirmware and Linux v5.0 on XO-1.75

[James Cameron]

On Fri, Feb 22, 2019 at 12:14:00PM +0100, Lubomir Rintel wrote:
> On Fri, 2019-02-22 at 17:23 +1100, James Cameron wrote:
> > Thanks, very good progress.  Here's what I've done;
> > 
> > - reviewed the aggregate change from master branch, and each commit,
> 
> Does it look, eh, reasonable? Got any comments/suggestions?

Yes, it looks reasonable.  Given that the firmware would not need to
run in factory and would only be used with a new kernel, the rest of
the firmware functions won't need to be considered.  I'm not worried
if it breaks the self-test features, for example.

> > - built the firmware on my xo-4 build server, flashed an xo-1.75 c2
> >   sku200x2; it boots fine the old kernel from arm-3.0-wip branch, with
> >   some unimportant problems like keymapping,
> 
> I intend to look into the key mapping at some point, because I've
> noticed the keyboard sometimes sends scancodes the kernel doesn't
> recognize.
> 
> By the way, my unit has has the "olpcm" non-membrane keyboard. I'm
> wondering if the scan codes it sends are the same as the membrane
> one?

I can't remember, sorry.

> Will the key mapping in hwdb need to distinguish between the two? (I
> also have a membrane keyboard, so I could actually just check that
> myself...)
> 
> > - on the fedora 18 root filesystem, installed the 5.0.0
> >   kernel{-core,-modules,} with --nodeps and --force,
> > 
> > - adjusted boot/ so that olpc.fth runs the 5.0.0 kernel,
> > 
> > - booted it a few times trying to fix the missing root filesystem;
> >   more work needed, the device name may have changed and i've not
> >   found a way to find what it is, or it isn't being detected; serial
> >   console doesn't work even with console=ttyS2,115200
> 
> Yeah, the device names are not stable for some reason. I don't know how
> are they determined, I'll need to take a look. Perhaps it's just a
> matter of adding the right aliases to the device tree.
> 
> Somewhat wierdly, my stripped down monolithic kernel calls the UART2
> ttyS2, while the Fedora kernel ends up with ttyS0.

Thanks, switching to ttyS0 worked.

> Similar issue with the MMC; the SD card ends up mmcblk1 with one
> kernel, mmcblk0 with another.

No MMC detects.

> The actual boot parameters I am testing with are in the lower half of
> my boot script (it's somewhat messy, copied it directly from my /boot
> without an attempt to tidy it up):
> https://people.freedesktop.org/~lkundrak/lr-olpc-boot/boot/menu.fth

Thanks.

> > and the
> >   keyboard is unresponsive in the dracut shell.
> 
> Which exact kernel are you using? Keyboard is not expected to work
> before rc6.

[0.00] Linux version 5.0.0-0.rc7.git2.1.fc31.armv7hl 
(mockbu...@buildvm-armv7-08.arm.fedoraproject.org) (gcc version 9.0.1 20190209 
(Red Hat 9.0.1-0.4) (GCC)) #1 SMP Wed Feb 20 21:06:49 UTC 2019

You pointed to it.

> Also, which config? Mine is basically this:
> https://raw.githubusercontent.com/hackerspace/olpc-xo175-linux/lr/olpc-xo175/arch/arm/configs/olpc_xo175_defconfig

The config used by Fedora.

Can we work toward some kind of reproducible build?

> > My mind has bitrotted.
> > 
> > On your interest in building on x86_64, suggestions;
> > 
> > - there are six 0.1" pitch pads on the back of the PCB which expose
> >   the SPI Flash chip pins, so you can hook a programmer to them, but
> >   check the voltage levels; some units used 1.8V chips, most used
> >   3.3V.
> 
> Ah, cool. Good to know there's a reasonable recovery option. Hope my
> chip is 3.3V, because I dropped the programmer that could do 1.8V on
> the floor and it seems it needs repairs :) 3.3V one could be programmed
> with a Rasbperry Pi, and I even have some spare 3.3V chips if I fuck up
> majorly.
> 
> But for now I just stay off overwriting cforth because I don't even
> feel like opening the machine again.
> 
> > - build a composite image by hand using the cforth you know already
> >   works, and the openfirmware built on x86_64,
> > 
> > - use binary comparison of the .rom file to make sure the cforth
> >   section hasn't changed much; if it hasn't, probably good to go, but
> >   if it has, no idea.
> > [...]

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: OpenFirmware and Linux v5.0 on XO-1.75

[James Cameron]

Thanks, very good progress.  Here's what I've done;

- reviewed the aggregate change from master branch, and each commit,

- built the firmware on my xo-4 build server, flashed an xo-1.75 c2
  sku200x2; it boots fine the old kernel from arm-3.0-wip branch, with
  some unimportant problems like keymapping,

- on the fedora 18 root filesystem, installed the 5.0.0
  kernel{-core,-modules,} with --nodeps and --force,

- adjusted boot/ so that olpc.fth runs the 5.0.0 kernel,

- booted it a few times trying to fix the missing root filesystem;
  more work needed, the device name may have changed and i've not
  found a way to find what it is, or it isn't being detected; serial
  console doesn't work even with console=ttyS2,115200, and the
  keyboard is unresponsive in the dracut shell.

My mind has bitrotted.

On your interest in building on x86_64, suggestions;

- there are six 0.1" pitch pads on the back of the PCB which expose
  the SPI Flash chip pins, so you can hook a programmer to them, but
  check the voltage levels; some units used 1.8V chips, most used
  3.3V.

- build a composite image by hand using the cforth you know already
  works, and the openfirmware built on x86_64,

- use binary comparison of the .rom file to make sure the cforth
  section hasn't changed much; if it hasn't, probably good to go, but
  if it has, no idea.

On Thu, Feb 21, 2019 at 11:54:15AM +0100, Lubomir Rintel wrote:
> Hi,
> 
> for the past few days I've been looking into updating the XO-1.75
> OpenFirmware so that it's good enough to boot mainline Linux.
> 
> It now looks usable enough: the essentials such as simple framebuffer,
> keyboard, Wi-Fi or USB all seem to work.
> 
> The branch's pretty large; counting 60 commits at the moment. Get it
> from:
> 
>   git pull https://github.com/lkundrak/openfirmware lr/olpc-xo175-1
> 
> It's not done or finished (see the TODOs in many commits). Some
> bindings are not settled in Linux tree. Howerver I still think it may
> be a good idea to share it early to get some feedback and identify bits
> that obviously stink.
> 
> I've tested it with the latest Fedora kernel [1] build (yay!) and also
> booted the latest OLPC OS release. When booting the latter, there were
> no differencies in "find /sys/devices -type d |sort" output, so I
> assume the drivers that would use the device tree (there probably
> aren't many) bind just fine.
> 
> [1] https://koji.fedoraproject.org/koji/buildinfo?buildID=1214041
> 
> I tried not to break other boards. olpc/4.0 still builds fine, but is
> likely to end up with three clock nodes (/pmua, /apbc and /clocks).
> olpc/3.0 was bitrotten before and I did not try doing x86 build, for
> the most part I've been building natively on the XO-1.75.
> 
> For a x86_64 hosted build I needed to patch cforth. See [2]. The 
> MitchBradley/cforth [1] master branch actually takes a similar
> approach, but there the 1.75 support there seems severely bitrotten.
> 
> [2] https://github.com/lkundrak/cforth/commit/c88790fd32.patch
> [3] https://github.com/MitchBradley/cforth
> 
> I didn't have the guts to actually flash and run the image built on
> x86_64. I don't not seem to be able to program the spi flash by
> attaching a soic8 clip to it, without unsoldering the chip and I don't
> feel like doing that if I fuck things up.
> 
> At some point I'll hopefully follow up with something that could be
> actually merged into the OpenFirmware, perhaps in a month or so. Until
> then some more bindings may settle.
> 
> In particular, my hopes are that some of Armada DRM or EC may make it
> into 5.1. Camera works, but needs some more love, perhaps 5.2.
> 
> Take care
> Lubo
> 

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Into 2019: XOs Laptops & software upgrades

[James Cameron]

On Wed, Jan 02, 2019 at 10:26:30PM -0500, tkkang wrote:
> Happy New Year.
> 
> Good to see the release of 13.2.10 OS for XO1 to XO4. Great work from James 
> and
> new developers who just keep making it better. With that I hope 2019 can be
> more productive for deployments still using XOs.
> 
> I have a few things I hope to see if he XO can function properly.
> 
> 1. Can Python 3.6 and above be installed. If Yes how?
>

On an XO-4 running OLPC OS 13.2.10, the latest version of Python,
which is 3.7.2, does build like this;

yum install -y gcc make zlib-devel
./configure
make
make install

Here's the interactive interpreter running;

Python 3.7.2 (default, Jan  3 2019, 08:52:48) 
[GCC 4.7.2 20121109 (Red Hat 4.7.2-8)] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>>

However, there are other optional dependencies that should be
installed first, so that the build has all the features possible.  The
Fedora 18 source package for python3 has a good starting point for
dependencies.

> Like to see if this could work [1]https://codewith.mu/en/tutorials/1.0/
> microbitas   new   Python needed

mu may need other dependencies.  You may need to experiment for
several iterations.

> 2. What is the best Firefox version that will load and and run on XO 1.5 to
> XO4? Need the good HTML 5 support.

Firefox installed by yum does have "good" HTML 5 support, but the
sites you refer to may require more than that.

> Currently I could  not run firefox for [2]
> https://www.edblocksapp.com/toprogram Edison Robot or [3]https://
> makecode.microbit.org/for Microbits
> 
> 3. Ways to load restricted so
> 
> References:
> 
> [1] https://codewith.mu/en/tutorials/1.0/microbit
> [2] https://www.edblocksapp.com/to
> [3] https://makecode.microbit.org/

> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel


-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Announcing OLPC OS 13.2.10

[James Cameron]

G'day,

We're pleased to announce the release of OLPC OS 13.2.10 for XO-1,
XO-1.5, XO-1.75 and XO-4.

It is Sugar 0.112 on Fedora 18, with updated activities Clock-20,
GetBooks-18.1, ImageViewer-64, Implode-19, Jukebox-34, Log-39,
Maze-28, Memorize-55, Paint-68, Physics-34, Pippy-72, Record-103,
StopWatch-20.1, Terminal-45.4, TurtleBlocks-218 and Write-99.1.

Details of fixes, known issues, and how to download, install or
upgrade are in the release notes:

http://wiki.laptop.org/go/Release_notes/13.2.10

Our thanks to all contributors, testers, upstreams, and those who
have provided feedback of any kind.

Build file names: 32022o0 32022o1 32022o2 32022o4

-- 
James Cameron
http://quozl.netrek.org/


signature.asc
Description: PGP signature
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Linux next-20181204 on a XO-1.75

[James Cameron]

On Tue, Dec 18, 2018 at 11:39:19AM +0100, Lubomir Rintel wrote:
> On Sat, 2018-12-08 at 17:26 +1100, James Cameron wrote:
> > Thanks.  On my test unit, this change was needed;
> > 
> > --- dt.fth.orig 2018-12-04 18:23:57.0 +1100
> > +++ dt.fth  2018-12-08 17:18:42.143073750 +1100
> > @@ -362,7 +362,7 @@
> >  " /clocks" encode-phandle  MMP2_CLK_TWSI5 encode-int encode+  "
> > resets" property
> >  device-end
> >  
> > -" dev /i2c@d4034000/accelerometer@1d" evaluate
> > +" dev /i2c@d4034000/accelerometer@19" evaluate
> >  " st,lis3lv02d" +compatible
> >  " st,lis331dlh" +compatible
> >  device-end
> 
> Thanks. I guess there's no need for the bus address there, "
> /i2c@d4034000/accelerometer" resolves to the correct node, regardless
> of what model/address the actual accelerometer is:

Yes.  Chip is LIS33DE in older builds, and LIS3DHTR in newer builds.

> https://people.freedesktop.org/~lkundrak/olpc/dt.fth
> ^ here's the current version I have.

Had a look, saw nothing obviously wrong.  Am a bit worried about the
amount of dictionary space remaining; you could reduce that worry by
removing those constants and adding the names as comments later.

Also you might check for data or control stack excursions; check
balance of the stacks across the fload.

> Includes camera and display. I hope I'll manage to follow up with the
> DRM patches later today. I'll very much appreciate if you take a look
> then as there are some bits I couldn't figure out without the panel and
> SoC manuals (and might be even wrong in OFW).

Ok.

> There are probably bugs. I've observed some problems, such as an
> occasional "Data Abort" when the USB ethernet is not plugged in (?)
> that can be fixed by merely splitting the file into two and floading
> them separately. I guess I'm corrupting something somewhere, but I find
> figuring out precisely what is going on non-trivial.

Yes, figuring out is non-trivial.  We rarely made significant device
tree changes at this stage, we try to bring them into the source.
There is simultaneous activity going on (keyboard, USB), and it's not
as multi-tasking as other environments.  Use ftrace after the "Data
Abort" to see if the saved exception stack can tell you anything
interesting?

> Also, the modifications to the internal sdhci node prevent OFW from
> booting from the internal emmc.  That said, it still serves as a good
> reference for the changes that will need to be done to support mainline
> kernels once the bits settle.

Thanks.

> PS: The previous message didn't make it to the list as it seems to
> require moderation. I guess this one will neither, unless you approve
> it.

I've just now located and approved your subscription.  As in the list
info, we get a lot of bogus subscriptions for some reason.

> Lubo
> 

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Linux next-20181204 on a XO-1.75

[James Cameron]

Thanks.  On my test unit, this change was needed;

--- dt.fth.orig 2018-12-04 18:23:57.0 +1100
+++ dt.fth  2018-12-08 17:18:42.143073750 +1100
@@ -362,7 +362,7 @@
 " /clocks" encode-phandle  MMP2_CLK_TWSI5 encode-int encode+  " resets" 
property
 device-end
 
-" dev /i2c@d4034000/accelerometer@1d" evaluate
+" dev /i2c@d4034000/accelerometer@19" evaluate
 " st,lis3lv02d" +compatible
 " st,lis331dlh" +compatible
 device-end

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: XO Spare Parts

[James Cameron]

Thanks Terry.  When you say large, how large is large?  Rough order of
magnitude?  I'll let my team know.

On Mon, Aug 06, 2018 at 06:17:55PM +1000, T Gillett wrote:
> Just to clarify, the idea is to make available suitable quantities
> of the parts to various groups supporting deployments, rather than
> try to ship the whole lot to one destination.
> 
> On Mon, Aug 6, 2018 at 11:43 AM, T Gillett <[1]tgill...@gmail.com> wrote:
> 
> Hi All
> 
> I have a large quantity of XO spare parts that are surplus from the
> Australian XO-4 deployment, including screens, batteries, keyboards and
> antennas.
> 
> The keyboards are the type with separate keys, not the membrane type, as
> used in some XO-1.75 and XO-4 laptops, and are very easy to change.
> The batteries and screens can be used across any model XO.
> 
> I am happy to make these available to anyone who wants them for just the
> cost of shipping. The equipment is located in Australia so international
> postage/shipping may be non-trivial.
> 
> Please let me know if you are interested in any of this equipment.
> 
> Regards
> Terry
> 

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Disable (?) NM random mac for shared XO wifi - USB0 / Chrome/ Pi Zero

[James Cameron]

That's great.  I've a Raspberry Pi Zero and XO-1 that I can put
together.  Might even be room inside or etch some of the back
plastic.

On Fri, Jul 27, 2018 at 08:22:43PM -0400, Carrol Riddle wrote:
> All:
> With the possibilities narrowed by James Cameron, found something that works 
> repeatedly on my test XO-1.  Must test again on a fresh system to find 
> minimum configuration.  Changes made to correct board id's, udev rules, 
> connection method prepared from scratch, and unmanaged mac entry in 
> /etc/NetworkManager.conf  .
> 
> Will post configuration later to close thread (if I can replicate it :)  ) 
> 
> Thanks all for help.
> Carrol Riddle
> 
> > > Now, setting MACAddressPolicy to none has no effect, which
> > > suggests the random address is coming from the kernel.
> > > 
> > > My guess is that the reason why your udev rpi script fails is that it
> > > is triggered on more than one udev event, or races with other things
> > > for access to the device.  Add more logging and debugging to it.
> > > 
> > > Reference:
> > > 
> > > https://www.freedesktop.org/software/systemd/man/systemd.link.html

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Disable (?) NM random mac for shared XO wifi - USB0 / Chrome/ Pi Zero

[James Cameron]

In my tests just now on OLPC OS 13.2.9 and an XO-4, my USB ethernet
adapters always come back after reconnection or reboot with the same
MAC address.  With or without Network Manager running.  So that
suggests that for adapters with a physical address ROM or
preprogrammed flash, they will appear predictable.

My tests above were not with Raspberry Pi Zero.  So I now turn to
testing with one;

1.  the MAC addresses change on each boot of the Zero, and are
unpredictable,

2.  the MAC address shown on each end is different; comparing usb0 on
the XO vs usb0 on the Zero,

3.  turning off Network Manager has no effect; therefore this is not a
Network Manager behaviour as you surmised,

I've also tested with kernel v4.15 on an OLPC NL3, and the same
unpredictable MAC address behaviour occurs.

Perhaps it is udev?  Using udevadm --monitor --property, shows
several events from the kernel, and the first one to mention the
MAC address eventually used is this one;

UDEV  [861.284453] add  
/devices/pci:00/:00:14.0/usb1/1-4/1-4.1/1-4.1:1.0/net/enp0s20u4u1 (net)
.MM_USBIFNUM=00
ACTION=add
DEVPATH=/devices/pci:00/:00:14.0/usb1/1-4/1-4.1/1-4.1:1.0/net/enp0s20u4u1
ID_BUS=usb
ID_MM_CANDIDATE=1
ID_MODEL=RNDIS_Ethernet_Gadget
ID_MODEL_ENC=RNDIS\x2fEthernet\x20Gadget
ID_MODEL_FROM_DATABASE=Linux-USB Ethernet/RNDIS Gadget
ID_MODEL_ID=a4a2
ID_NET_DRIVER=cdc_ether
ID_NET_LINK_FILE=/lib/systemd/network/99-default.link
ID_NET_NAME=enp0s20u4u1
ID_NET_NAME_MAC=enxca2a7e605b6f <-- encx followed by the address
ID_NET_NAME_PATH=enp0s20u4u1
ID_PATH=pci-:00:14.0-usb-0:4.1:1.0
ID_PATH_TAG=pci-_00_14_0-usb-0_4_1_1_0
ID_REVISION=0409
ID_SERIAL=Linux_4.9.35+_with_2098.usb_RNDIS_Ethernet_Gadget
ID_TYPE=generic
ID_USB_CLASS_FROM_DATABASE=Communications
ID_USB_DRIVER=cdc_ether
ID_USB_INTERFACES=:0202ff:0a:020600:
ID_USB_INTERFACE_NUM=00
ID_VENDOR=Linux_4.9.35+_with_2098.usb
ID_VENDOR_ENC=Linux\x204.9.35+\x20with\x202098.usb
ID_VENDOR_FROM_DATABASE=Netchip Technology, Inc.
ID_VENDOR_ID=0525
IFINDEX=7
INTERFACE=enp0s20u4u1
INTERFACE_OLD=usb0
SEQNUM=3175
SUBSYSTEM=net
SYSTEMD_ALIAS=/sys/subsystem/net/devices/enp0s20u4u1
TAGS=:systemd:
USEC_INITIALIZED=861268742

So the MAC address originates from the kernel or systemd, not Network
Manager.

As there is a udev event, it should be possible to write a udev rule
to change the MAC address when this or one of the later events
arrives.

However, perhaps we can see if this MAC address comes from the kernel
or systemd; the event above references
/lib/systemd/network/99-default.link which says'

[Link]
NamePolicy=kernel database onboard slot path
MACAddressPolicy=persistent

Now, setting MACAddressPolicy to none has no effect, which
suggests the random address is coming from the kernel.

My guess is that the reason why your udev rpi script fails is that it
is triggered on more than one udev event, or races with other things
for access to the device.  Add more logging and debugging to it.

Reference:

https://www.freedesktop.org/software/systemd/man/systemd.link.html

On Fri, Jul 20, 2018 at 11:20:57PM -0400, Carrol Riddle wrote:
> The difficulty is that a new connection is started on each boot and starts as 
> "automatic" IPv4 instead of the desired "share with other computers" This 
> results from a randomized MAC address, which appears to be controlled by 
> Network Manager. If the MAC address persists on next boot,
> then the connection file will persist.
> 
> The MAC address in the cmdline.txt on Zero is vestige of efforts. It was left 
> since I suspect it will be part of the solution. Retested without and it 
> makes no difference when no other changes are made (i.e., only with 
> essentials in cmdline.txt). I have tried static IP on Zero via adding 
> ip=192.168.1.2 to cmdline.txt or in dhcpcd.config and udev rule on XO for 
> ip=198.168.1.1  .  Uncertain if udev rule needs additional specification 
> (parent ATTRS ?).
> 
> Reasoned that host_addr was telling Zero where to find existing MAC on XO and 
> udev rule then tried with hw to assure it was there.  Have tried adding 
> mac-cloned-address to NetworkManager.conf 
> 
> NetworkManager randomization of MAC addresses is widely discussed in search 
> results and variety of solutions proposed.  I have had periods where MAC was 
> persistent, but only have it fail a few hours later (just powered off). One 
> case of oddity was the persistence on every other boot, if the intervening 
> boot was without Zero attached.  However, my work with wired links to routers 
> (TP-Link MR3020 and Village Telco MP20) shows persistence in boot of router.  
> 
> A competent programmer might just modify the started connection file (using 
> awk, grep, sed ?) and restart NetworkManager.   :)  .  I tried replacing (cp) 
> with model files, but that has problems (date stamp).   
> /etc/NetworkManger/connections 
>

Re: Disable (?) NM random mac for shared XO wifi - USB0 / Chrome/ Pi Zero

[James Cameron]

Interesting.

But why do you need to give a MAC address on cmdline.txt on the RPi
and in the XO udev script?  g_ether should assign an address, and
cdc_ether should receive it from USB descriptors.

On Tue, Jul 10, 2018 at 10:50:44AM -0400, Carrol Riddle wrote:
> Soliciting info or pointers to resources for disabling mac address 
> randomization by NetworkManager for USB0 (eth1) on XO's.  Wifi mac addresses 
> are not randomized.
> 
> An XO can share its Wifi over wired ethernet link (USB0) where it can be used 
> by
> a Chromium browser on a linked Raspberry Pi Zero (not W).  SSH -X on the link
> allows XO keyboard and display to be used for Chromium.
> 
> The several setup steps are simple, but must be repeated with each boot since
> NetworkManager (version 0.9.8.* on XO's) supplies a different mac address for
> the link on each boot.  To provide a simple Sugar wrapper to run Chromium, a
> consistent mac address would be needed.
> 
> Methods from web searches have not been successful in disabling 
> randomization, 
> largely since they use features added after 0.9.8.   Using a udev rule to 
> "ifconfig
> usb0 hw ether address" has come the closest, but breaks after a few boots and 
> returns to
> randomization.
> 
> BACKGROUND
> 
> For those interested, details are below, but are only peripheral to question. 
>  This use of the XO's display, keyboard and wifi by Chromium / Zero provides 
> a low cost way to access a widely accepted browser. The Zero can be plugged 
> into XO USB for power and signals. The Zero is
> available for < $10, and with SD card and plug, total cost should be $15 - 20
> USD. 
> 
> 
> __
> 
> 
> MANUAL METHOD - XO (after Pi Zero setup)
> 
> In Sugar, connect to wifi . This needs to be done only the first time and is
> persistent between boots.  Switch to Gnome and disconnect from the "Wired
> connection ?" that just formed. Edit the connection :  IPv6 to ignore, IPv4 to
> require IPv4 and "shared to other computers"   and save.  Connect to this
> connection.  The connection info should show ip address starting with 10:  .
> This link is not persistent between boots.
> 
> systemctl enable sshd.service
> 
> PI ZERO SETUP (prior)
> 
> Raspberry Pi Zero with Raspbian Desktop installed.  To   /boot/config.txt on  
> Pi
> SD edit and add dtoverlay=dwc2 .  To /boot/cmdline.txt add after "rootwait"
>  modules-load=dwc2,g_ether host_addr=00:22:82:ff:ff:20 
> dev_addr=00:22:82:ff:ff:22  (Substitute local mac addresses.)   To 
> start with SSH, add dummy "ssh" file to /boot directory.  On XO, ssh 
> pi@raspberrypi.local with password raspberry   .   Configure Zero to boot to
> command line (CLI) with raspi-config  .  To use Raspbian desktop,  ssh -X
> pi@raspberry.local, and then /etc/X11/Xsession.  The Pi Tool Bar partially
> overlaps the Gnome Tool Bar, but selecting "medium display size" minimizes
> affects.
> 
> UDEV INFO FOR RULES  (but not able to make work)
> 
> DRIVERS=="cdc_ether"
> 
> ATTRS{idVendor=1d6b}
> 
> ATTRS{idProduct=0001}
> 
> RUN+="/etc/udev/scripts/rpi"
> 
> 
> Script rpi:
> 
> ifconfig usb0 down
> 
> ifconfig usb0 hw ether 00:22:82:FF:FF:20
> 
> ifconfig usb0 up
> 
> _
> 
> Carrol Riddle
> 
> ebox...@scishare.com
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: [Server-devel] [UKids] easiest way to automate install of a collection 12+ PDF's onto Sugar on XOs?

[James Cameron]

On Thu, Apr 19, 2018 at 08:26:27PM -0400, Adam Holt wrote:
> On Thu, Apr 19, 2018 at 8:17 PM, James Cameron <[1]qu...@laptop.org> wrote:
> 
> without deleting the identity
> key the activity collaboration feature will fail in strange ways.
> 
> What's the identity key, and how is it deleted?

A cryptographic identitifier of the Sugar user, randomly created at
first boot and stored in files;

.sugar/default/owner.key and
.sugar/default/owner.key.pub

Delete it using rm, as it says here;
http://wiki.laptop.org/go/Imaging/Side_effects

"Otherwise the cloned laptops won't be able to collaborate using activities, or 
the wrong names will be shown in Chat,"

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: [Server-devel] [UKids] easiest way to automate install of a collection 12+ PDF's onto Sugar on XOs?

[James Cameron]

On Thu, Apr 19, 2018 at 08:26:27PM -0400, Adam Holt wrote:
> On Thu, Apr 19, 2018 at 8:17 PM, James Cameron <[1]qu...@laptop.org> wrote:
> 
> without deleting the identity
> key the activity collaboration feature will fail in strange ways.
> 
> What's the identity key, and how is it deleted?

A cryptographic identitifier of the Sugar user, randomly created at
first boot and stored in files;

.sugar/default/owner.key and
.sugar/default/owner.key.pub

Delete it using rm, as it says here;
http://wiki.laptop.org/go/Imaging/Side_effects

"Otherwise the cloned laptops won't be able to collaborate using activities, or 
the wrong names will be shown in Chat,"

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: [Server-devel] [UKids] easiest way to automate install of a collection 12+ PDF's onto Sugar on XOs?

[James Cameron]

Which things are truly important depend on the circumstances, and only
the teacher will know.  For instance, without deleting the identity
key the activity collaboration feature will fail in strange ways.

On Sun, Apr 15, 2018 at 07:07:09PM -0400, Adam Holt wrote:
> On Sun, Apr 15, 2018 at 6:58 PM, Adam Holt <[1]h...@laptop.org> wrote:
> 
> On Mon, Apr 9, 2018 at 11:54 PM, Tony Anderson <[2]tony_ander...@usa.net>
> wrote:
>
> A teacher should never 'rm -rf /home/olpc/.sugar'. If the intent is to
> remove the Journal because of space considerations, 'rm -rf 
> /home/olpc/
> sugar/datastore' is sufficient. After this command the XO needs to be
> rebooted to create a new empty datastore.
> 
>
> 'rm -rf /home/olpc/sugar/datastore' is not sufficient to accomplish the
> required task (deleting the child's Sugar name).
> 
> 'rm /home/olpc/.sugar' is the only way we know.  (Unless there's a better
> approach ?)
> 
> There are many similar suggestions here:
> 
>    [3]http://wiki.laptop.org/go/Imaging/Side_effects
> 
> Which of the above are truly important for a teacher to type in at the
> beginning of the semester, to clean out Sugar on an XO.
> 
> Teachers much prefer something very short like "rm -rf /home/olpc/.sugar"
> (unless there's a better way?)
> 
> PS Naturally Gnome is not as easy to clean out, if students have left MP3's 
> and
> personal files lying around!
> 
> References:
> 
> [1] mailto:h...@laptop.org
> [2] mailto:tony_ander...@usa.net
> [3] http://wiki.laptop.org/go/Imaging/Side_effects

> ___
> Server-devel mailing list
> Server-devel@lists.laptop.org
> http://lists.laptop.org/listinfo/server-devel


-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: [Server-devel] [UKids] easiest way to automate install of a collection 12+ PDF's onto Sugar on XOs?

[James Cameron]

Which things are truly important depend on the circumstances, and only
the teacher will know.  For instance, without deleting the identity
key the activity collaboration feature will fail in strange ways.

On Sun, Apr 15, 2018 at 07:07:09PM -0400, Adam Holt wrote:
> On Sun, Apr 15, 2018 at 6:58 PM, Adam Holt <[1]h...@laptop.org> wrote:
> 
> On Mon, Apr 9, 2018 at 11:54 PM, Tony Anderson <[2]tony_ander...@usa.net>
> wrote:
>
> A teacher should never 'rm -rf /home/olpc/.sugar'. If the intent is to
> remove the Journal because of space considerations, 'rm -rf 
> /home/olpc/
> sugar/datastore' is sufficient. After this command the XO needs to be
> rebooted to create a new empty datastore.
> 
>
> 'rm -rf /home/olpc/sugar/datastore' is not sufficient to accomplish the
> required task (deleting the child's Sugar name).
> 
> 'rm /home/olpc/.sugar' is the only way we know.  (Unless there's a better
> approach ?)
> 
> There are many similar suggestions here:
> 
>    [3]http://wiki.laptop.org/go/Imaging/Side_effects
> 
> Which of the above are truly important for a teacher to type in at the
> beginning of the semester, to clean out Sugar on an XO.
> 
> Teachers much prefer something very short like "rm -rf /home/olpc/.sugar"
> (unless there's a better way?)
> 
> PS Naturally Gnome is not as easy to clean out, if students have left MP3's 
> and
> personal files lying around!
> 
> References:
> 
> [1] mailto:h...@laptop.org
> [2] mailto:tony_ander...@usa.net
> [3] http://wiki.laptop.org/go/Imaging/Side_effects

> ___
> Server-devel mailing list
> server-de...@lists.laptop.org
> http://lists.laptop.org/listinfo/server-devel


-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: [Server-devel] [UKids] easiest way to automate install of a collection 12+ PDF's onto Sugar on XOs?

[James Cameron]

OLPC OS on the XO is configured for ownership style "one laptop per
child".

If you always want to delete the child's Sugar name, you might either
change login scripts to delete it before starting, or assume a
default.  It simplifies getting started into a class.

e.g. in the OLPC OS 16.04 live build on the NL3, we have this in
/usr/bin/sugar;

gsettings set org.sugarlabs.user nick 'You'
gsettings set org.sugarlabs.user gender ''
gsettings set org.sugarlabs.user birth-timestamp 689659403
gsettings set org.sugarlabs.user group-label 'Adult'
gsettings set org.sugarlabs.user color '#808080,#c0c0c0'
gconftool-2 --set /desktop/sugar/user/color --type string '#808080,#c0c0c0'
gconftool-2 --set /desktop/sugar/user/nick --type string 'You'

On Sun, Apr 15, 2018 at 06:58:11PM -0400, Adam Holt wrote:
> On Mon, Apr 9, 2018 at 11:54 PM, Tony Anderson <[1]tony_ander...@usa.net>
> wrote:
> 
> A teacher should never 'rm -rf /home/olpc/.sugar'. If the intent is to
> remove the Journal because of space considerations, 'rm -rf /home/olpc/
> sugar/datastore' is sufficient. After this command the XO needs to be
> rebooted to create a new empty datastore.
> 
> 'rm -rf /home/olpc/sugar/datastore' is not sufficient to accomplish the
> required task (deleting the child's Sugar name).
> 
> 'rm /home/olpc/.sugar' is the only way we know.  (Unless there's a better
> approach ?)
> 
> References:
> 
> [1] mailto:tony_ander...@usa.net

> ___
> Devel mailing list
> de...@lists.laptop.org
> http://lists.laptop.org/listinfo/devel


-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: [UKids] easiest way to automate install of a collection 12+ PDF's onto Sugar on XOs?

[James Cameron]

OLPC OS on the XO is configured for ownership style "one laptop per
child".

If you always want to delete the child's Sugar name, you might either
change login scripts to delete it before starting, or assume a
default.  It simplifies getting started into a class.

e.g. in the OLPC OS 16.04 live build on the NL3, we have this in
/usr/bin/sugar;

gsettings set org.sugarlabs.user nick 'You'
gsettings set org.sugarlabs.user gender ''
gsettings set org.sugarlabs.user birth-timestamp 689659403
gsettings set org.sugarlabs.user group-label 'Adult'
gsettings set org.sugarlabs.user color '#808080,#c0c0c0'
gconftool-2 --set /desktop/sugar/user/color --type string '#808080,#c0c0c0'
gconftool-2 --set /desktop/sugar/user/nick --type string 'You'

On Sun, Apr 15, 2018 at 06:58:11PM -0400, Adam Holt wrote:
> On Mon, Apr 9, 2018 at 11:54 PM, Tony Anderson <[1]tony_ander...@usa.net>
> wrote:
> 
> A teacher should never 'rm -rf /home/olpc/.sugar'. If the intent is to
> remove the Journal because of space considerations, 'rm -rf /home/olpc/
> sugar/datastore' is sufficient. After this command the XO needs to be
> rebooted to create a new empty datastore.
> 
> 'rm -rf /home/olpc/sugar/datastore' is not sufficient to accomplish the
> required task (deleting the child's Sugar name).
> 
> 'rm /home/olpc/.sugar' is the only way we know.  (Unless there's a better
> approach ?)
> 
> References:
> 
> [1] mailto:tony_ander...@usa.net

> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel


-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: sugar-install-bundle fails to behave like Browse when installing .xol bundle

[James Cameron]

Your bundle does not meet the requirements.

I've downloaded it, and sugar-install-bundle says what is wrong;

sugar3.bundle.bundle.MalformedBundleException: All files in the bundle
must be inside a single directory whose name ends with '.activity'

Sure enough, "unzip -l" shows the directory name does not end with
'.activity'.

$ unzip -l livr*
Archive:  livreshaiti.xol
  Length  DateTimeName
-  -- -   
0  2013-12-15 13:58   livreshaiti/
   935794  2013-12-15 13:14   livreshaiti/chita_pa_bay.pdf
  345  2013-12-15 14:04   livreshaiti/index.html
0  2013-12-15 13:32   livreshaiti/library/
  434  2008-03-24 11:04   livreshaiti/library/books.png
  293  2013-12-15 14:02   livreshaiti/library/library.info
  1286434  2013-12-15 13:15   livreshaiti/st_exupery_le_petit_prince.pdf
- ---
  2223300 7 files

On Wed, Apr 11, 2018 at 12:12:17PM -0400, Adam Holt wrote:
> Clicking on http://download.unleashkids.org/HaitiOS/bundles/livreshaiti.xol
> works in Sugar's Browse activity on OLPC OS 13.2.9, correctly loading books
> into Sugar.
> 
> However doing the same thing from the command-line ("sugar-install-bundle
> livreshaiti.xol") does not work :/
> 
> Does anyone have any tricks/ideas to making this happen programmatically?
> 
> (Context: we're mostly doing XO-1.5 work at the moment, and our scripts need 
> to
> work post-deployment, as educators seek continuous improvement.)
> 
> References:
> 
> [1] http://download.unleashkids.org/HaitiOS/bundles/livreshaiti.xol

> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel


-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: how to change Sugar's timezone+language after XO deployment

[James Cameron]

Supported method for configuring timezone is the "timezone" option in
the "[base]" section of the OS Builder configuration file.

Traceback you show is not fatal, is irrelevant, but is a bug that
should be fixed.

https://github.com/sugarlabs/sugar/issues/793

You should not use sudo for this command.  Sugar timezone is not a
system-wide configuration setting; it is local to the user.

On Wed, Apr 11, 2018 at 11:43:18AM -0400, Adam Holt wrote:
> Can commands like the following be made to work?
> 
>    sugar-control-panel -s timezone America/Port-au-Prince
>    sugar-control-panel -s languages French/France
> 
> If not, is there some other clever way to do this programmatically, perhaps
> using the gsettings command?
> 
> Thanks if poss!  Example error posted below -- changes do not take effect, 
> even
> after reboot.  Original doc @ 
> [1]http://wiki.laptop.org/go/Sugar_Control_Panel#
> Setting_The_Timezone
> 
> Adam
> 
> [olpc@xo-4a-c3-b5 ~]$ sudo sugar-control-panel -s timezone America/
> Port-au-Prince
> ERROR:root:Exception while loading extension:
> Traceback (most recent call last):
>   File "/usr/lib/python2.7/site-packages/jarabe/controlpanel/cmd.py", line 68,
> in load_modules
>     globals(), locals(), ['model'])
>   File "/usr/share/sugar/extensions/cpsection/frame/model.py", line 17, in
> 
>     from jarabe.frame import get_view
>   File "/usr/lib/python2.7/site-packages/jarabe/frame/__init__.py", line 16, 
> in
> 
>     from jarabe.frame.frame import Frame
>   File "/usr/lib/python2.7/site-packages/jarabe/frame/frame.py", line 30, in
> 
>     from jarabe.frame.friendstray import FriendsTray
>   File "/usr/lib/python2.7/site-packages/jarabe/frame/friendstray.py", line 
> 20,
> in 
>     from jarabe.view.buddymenu import BuddyMenu
>   File "/usr/lib/python2.7/site-packages/jarabe/view/buddymenu.py", line 37, 
> in
> 
>     import jarabe.desktop.homewindow
>   File "/usr/lib/python2.7/site-packages/jarabe/desktop/homewindow.py", line
> 28, in 
>     from jarabe.desktop.meshbox import MeshBox
>   File "/usr/lib/python2.7/site-packages/jarabe/desktop/meshbox.py", line 35,
> in 
>     from jarabe.view.buddyicon import BuddyIcon
>   File "/usr/lib/python2.7/site-packages/jarabe/view/buddyicon.py", line 19, 
> in
> 
>     from jarabe.view.buddymenu import BuddyMenu
> ImportError: cannot import name BuddyMenu
> To apply your changes you have to restart Sugar.
> Hit ctrl+alt+erase on the keyboard to trigger a restart.
> 
> References:
> 
> [1] http://wiki.laptop.org/go/Sugar_Control_Panel#Setting_The_Timezone

> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel


-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: [Server-devel] do RPi 3 B+ still need heat sinks?

[James Cameron]

On Tue, Apr 03, 2018 at 08:35:54PM -0400, Adam Holt wrote:
> On Tue, Apr 3, 2018 at 8:17 PM, James Cameron <[1]qu...@laptop.org> wrote:
> 
> On Tue, Apr 03, 2018 at 05:28:31PM -0400, Adam Holt wrote:
> > On Tue, Apr 3, 2018 at 5:01 PM, James Cameron <[1][2]qu...@laptop.org>
> wrote:
> >
> >     On Tue, Apr 03, 2018 at 10:33:12AM -0400, Adam Holt wrote:
> >     > On Mon, Apr 2, 2018 at 12:36 PM, Adam Holt <[1][2][3]
> h...@laptop.org> wrote:
> >     >
> >     >     On Mon, Apr 2, 2018 at 10:44 AM, Adam Holt <[2][3][4]
> h...@laptop.org>
> >     wrote:
> >     >
> >     >         1) Recap from 2016: heatsinks on the original RPi3 CPU are
> not a
> >     >         cure-all when the RPi 3 is enclosed in a case that lacks
> >     ventilation on
> >     >         a hot day.  But CPU heatsinks Do Work when taking the
> plastic top
> >     off
> >     >         the of the original RPi 3.  The CPU throttling problem
> >     "immediately"
> >     >         goes away on such hot days...bringing the temperature back
> down
> >     below
> >     >         80C...as measured by the command:
> >     >
> >     >            vcgencmd measure_temp
> >     >
> >     >         2) It's snowing today right outside my window, so I can't
> easily
> >     >         simulate a hot summer's day -- but can others who live in
> hot
> >     >         environments report back their readings above, when 
> running
> the
> >     new
> >     >         RPi3 B+ in various conditions?
> >     >           ☆ With heatsink on CPU -- and without?
> >     >           ☆ With motherboard fully enclose by a case -- and
> without?
> >     >
> >     >     My own results, with all 4 CPU's unloaded, in a chilly room:
> >     >
> >     >  RPi 3 with-heatsink-on-CPU / RPi 3 B+ / RPi 3 B+
> with-heatsink-on-CPU
> >     > 44-46C / 46-48C / 45-47C case's plastic top removed (w/o wind or
> active
> >     > ventilation)
> >     > 46-49C / 48-51C / 48-52C case's plastic top attached (contains 
> 100+
> small
> >     holes
> >     > on 1 end, allowing very little ventilation)
> >     >
> >     >     After I ran "yes > /dev/null &" 4 times, to fully load all 4
> cores of
> >     the
> >     >     CPU:
> >     >
> >     >  RPi 3 with-heatsink-on-CPU / RPi 3 B+ / RPi 3 B+
> with-heatsink-on-CPU
> >     > 80-82C / 70-71C / 71-72C case's plastic top removed (w/o wind or
> active
> >     > ventilation)
> >     > 82-84C / 75-78C / 79-82C case's plastic top attached (contains 
> 100+
> small
> >     holes
> >     > on 1 end, allowing very little ventilation)
> >     >
> >     > RESULT: attaching a heatsink to the RPi 3 B+ CPU does not help.  
> It
> might
> >     even
> >     > make things a bit worse, hmm.
> >
> >     Yes, your heatsink is no good.
> >
> >     The B+ CPU has a heatsink or heat spreader already, that silver
> >     coloured bevelled structure with the black dot and Broadcom logo.
> >
> >     What is the shape, size, and attachment method for your added
> >     heatsink?
> >
> > It's the standard Canakit 7-fin aluminum heatsink shown here, attached 
> to
> the
> > CPU with its own basic 3M self-adhesive sticker:
> >
> > [4][5]http://www.bestofjay.com/w/
> > raspberry-pi-3-overclock-heat-test-flirc-case-vs-canakit-heatsink/
> > 
> [5][6]http://11986-presscdn-0-77.pagely.netdna-cdn.com/wp-content/uploads
> /2016/04/
> > heat-sinks-installed.jpg
> 
> Given your results, probably wrong heatsink for the job.  No, I don't
> know of a better one.
> 
> Indeed.  It'd be really great if the Raspberry Pi Foundation would publish
> basic recommendations here, even if just community/testing suggestions.

Best tell them.

> Theory; the adhesive won't make good contact with outer bevel, and
> passive airflow above that outer bevel will be reduced.
> 
> Theory; bubble in adhesive if the heatsink was not angled slightly as
> you pushed it down.
> 
> Theory; the thermal control system is very differe

Re: [Server-devel] do RPi 3 B+ still need heat sinks?

[James Cameron]

On Tue, Apr 03, 2018 at 05:28:31PM -0400, Adam Holt wrote:
> On Tue, Apr 3, 2018 at 5:01 PM, James Cameron <[1]qu...@laptop.org> wrote:
> 
> On Tue, Apr 03, 2018 at 10:33:12AM -0400, Adam Holt wrote:
> > On Mon, Apr 2, 2018 at 12:36 PM, Adam Holt <[1][2]h...@laptop.org> 
> wrote:
> >
> >     On Mon, Apr 2, 2018 at 10:44 AM, Adam Holt <[2][3]h...@laptop.org>
> wrote:
> >
> >         1) Recap from 2016: heatsinks on the original RPi3 CPU are not a
> >         cure-all when the RPi 3 is enclosed in a case that lacks
> ventilation on
> >         a hot day.  But CPU heatsinks Do Work when taking the plastic 
> top
> off
> >         the of the original RPi 3.  The CPU throttling problem
> "immediately"
> >         goes away on such hot days...bringing the temperature back down
> below
> >         80C...as measured by the command:
> >
> >            vcgencmd measure_temp
> >
> >         2) It's snowing today right outside my window, so I can't easily
> >         simulate a hot summer's day -- but can others who live in hot
> >         environments report back their readings above, when running the
> new
> >         RPi3 B+ in various conditions?
> >           ☆ With heatsink on CPU -- and without?
> >           ☆ With motherboard fully enclose by a case -- and without?
> >
> >     My own results, with all 4 CPU's unloaded, in a chilly room:
> >
> >  RPi 3 with-heatsink-on-CPU / RPi 3 B+ / RPi 3 B+ with-heatsink-on-CPU
> > 44-46C / 46-48C / 45-47C case's plastic top removed (w/o wind or active
> > ventilation)
> > 46-49C / 48-51C / 48-52C case's plastic top attached (contains 100+ 
> small
> holes
> > on 1 end, allowing very little ventilation)
> >
> >     After I ran "yes > /dev/null &" 4 times, to fully load all 4 cores 
> of
> the
> >     CPU:
> >
> >  RPi 3 with-heatsink-on-CPU / RPi 3 B+ / RPi 3 B+ with-heatsink-on-CPU
> > 80-82C / 70-71C / 71-72C case's plastic top removed (w/o wind or active
> > ventilation)
> > 82-84C / 75-78C / 79-82C case's plastic top attached (contains 100+ 
> small
> holes
> > on 1 end, allowing very little ventilation)
> >
> > RESULT: attaching a heatsink to the RPi 3 B+ CPU does not help.  It 
> might
> even
> > make things a bit worse, hmm.
> 
> Yes, your heatsink is no good.
> 
> The B+ CPU has a heatsink or heat spreader already, that silver
> coloured bevelled structure with the black dot and Broadcom logo.
> 
> What is the shape, size, and attachment method for your added
> heatsink?
> 
> It's the standard Canakit 7-fin aluminum heatsink shown here, attached to the
> CPU with its own basic 3M self-adhesive sticker:
> 
> [4]http://www.bestofjay.com/w/
> raspberry-pi-3-overclock-heat-test-flirc-case-vs-canakit-heatsink/
> [5]http://11986-presscdn-0-77.pagely.netdna-cdn.com/wp-content/uploads/2016/04/
> heat-sinks-installed.jpg

Given your results, probably wrong heatsink for the job.  No, I don't
know of a better one.

Theory; the adhesive won't make good contact with outer bevel, and
passive airflow above that outer bevel will be reduced.

Theory; bubble in adhesive if the heatsink was not angled slightly as
you pushed it down.

Theory; the thermal control system is very different to the previous
version, and this invalidates your test method.  The system uses
thermal mass and core frequency scaling, and you've changed the
thermal mass.  It's non-linear.

Report your kernel version; the corresponding Raspbian release has
changes for the B+.

See
https://www.raspberrypi.org/blog/raspberry-pi-3-model-bplus-sale-now-35/
for more technical detail on the thermals; note how the core frequency
varies and the thermal mass is a resource.

Suggest you use sysbench for at least 15 minutes before reading the
temperature.  The "yes" test is very weak.

> FWIW these same heatsinks do lower the temperature of the original RPi 3 by
> many degrees IF the CPU 's under medium load AND the top of the plastic case 
> is
> removed :)

But irrelevant, as the CPU is a different model, has a flat surface,
and does not have a thermal control system that uses thermal mass.

Please remember; things change, and this breaks your knowledge.  You
must learn again.

> 
> > In any case: this result is completely different the original RPi 3
> (where past
> > experiments have shown that a heatsink-on-CPU greatly lowers its
> temperature,
> > When Th

Re: [Server-devel] do RPi 3 B+ still need heat sinks?

[James Cameron]

On Tue, Apr 03, 2018 at 10:33:12AM -0400, Adam Holt wrote:
> On Mon, Apr 2, 2018 at 12:36 PM, Adam Holt <[1]h...@laptop.org> wrote:
> 
> On Mon, Apr 2, 2018 at 10:44 AM, Adam Holt <[2]h...@laptop.org> wrote:
>
> 1) Recap from 2016: heatsinks on the original RPi3 CPU are not a
> cure-all when the RPi 3 is enclosed in a case that lacks ventilation 
> on
> a hot day.  But CPU heatsinks Do Work when taking the plastic top off
> the of the original RPi 3.  The CPU throttling problem "immediately"
> goes away on such hot days...bringing the temperature back down below
> 80C...as measured by the command:
> 
>    vcgencmd measure_temp
> 
> 2) It's snowing today right outside my window, so I can't easily
> simulate a hot summer's day -- but can others who live in hot
> environments report back their readings above, when running the new
> RPi3 B+ in various conditions?
>   ☆ With heatsink on CPU -- and without?
>   ☆ With motherboard fully enclose by a case -- and without?
> 
> My own results, with all 4 CPU's unloaded, in a chilly room:
> 
>  RPi 3 with-heatsink-on-CPU / RPi 3 B+ / RPi 3 B+ with-heatsink-on-CPU
> 44-46C / 46-48C / 45-47C case's plastic top removed (w/o wind or active
> ventilation)
> 46-49C / 48-51C / 48-52C case's plastic top attached (contains 100+ small 
> holes
> on 1 end, allowing very little ventilation)
> 
> After I ran "yes > /dev/null &" 4 times, to fully load all 4 cores of the
> CPU:
> 
>  RPi 3 with-heatsink-on-CPU / RPi 3 B+ / RPi 3 B+ with-heatsink-on-CPU
> 80-82C / 70-71C / 71-72C case's plastic top removed (w/o wind or active
> ventilation)
> 82-84C / 75-78C / 79-82C case's plastic top attached (contains 100+ small 
> holes
> on 1 end, allowing very little ventilation)
> 
> RESULT: attaching a heatsink to the RPi 3 B+ CPU does not help.  It might even
> make things a bit worse, hmm.

Yes, your heatsink is no good.

The B+ CPU has a heatsink or heat spreader already, that silver
coloured bevelled structure with the black dot and Broadcom logo.

What is the shape, size, and attachment method for your added
heatsink?

> In any case: this result is completely different the original RPi 3 (where 
> past
> experiments have shown that a heatsink-on-CPU greatly lowers its temperature,
> When The Case Is Open!)
>  
> 
> CONCLUSION: the RPi 3 B+ appears to be a real winner in CPU-intensive
> conditions, even in a "wintry" conditions (room is about 15C, and it
> continues to snow right outside the window).  The RPi 3 CPU is supposed to
> self-throttle at 80C, until it just about turns itself off at 85C.  I'd
> assume the RPi 3 B+ CPU does the same?  But do not know for sure.  Thanks
> to all who can add any similar data points, in warmer climate/conditions.
> 
> CLARIFS:  Both RPi's were running a near-final prerelease of
> Internet-in-a-Box 6.5 on Raspbian Lite to explore real-world conditions.  
> I waited 10+ min in all above 8 experiments before taking "steady state"
> temp readings.  Still, fluctuations in CPU activity (and temperature)
> arise, even long after I ran "killall yes" to end the most intensive CPU
> activity.
> 
> 3) Prelim thermal analysis of RPi 3 B+:
> 
> [3]https://cdn-images-1.medium.com/max/1600/1*0jU89Yu_6miI-CODB
> MuHAw.png
> [4]https://medium.com/@ghalfacree/benchmarking-the-raspberry-pi
> -3-b-plus-44122cf3d806
> 
> --
> [5]
> [6]Unsung Heroes of OLPC, interviewed live @ [7]http://unleashkids.org
> !
> 
> --
> Unsung Heroes of OLPC, interviewed live @ [8]http://unleashkids.org !
>
> 
> References:
> 
> [1] mailto:h...@laptop.org
> [2] mailto:h...@laptop.org
> [3] https://cdn-images-1.medium.com/max/1600/1*0jU89Yu_6miI-CODBMuHAw.png
> [4] 
> https://medium.com/@ghalfacree/benchmarking-the-raspberry-pi-3-b-plus-44122cf3d806
> [5] 
> https://medium.com/@ghalfacree/benchmarking-the-raspberry-pi-3-b-plus-44122cf3d806
> [6] 
> https://medium.com/@ghalfacree/benchmarking-the-raspberry-pi-3-b-plus-44122cf3d806
> [7] http://unleashkids.org/
> [8] http://unleashkids.org/

> ___
> Server-devel mailing list
> Server-devel@lists.laptop.org
> http://lists.laptop.org/listinfo/server-devel


-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: [Server-devel] do RPi 3 B+ still need heat sinks?

[James Cameron]

On Tue, Apr 03, 2018 at 09:07:22AM -0400, Adam Holt wrote:
> On Tue, Apr 3, 2018 at 1:53 AM, James Cameron <[1]qu...@laptop.org> wrote:
> 
> Temperature readings are not comparable between RPi 3 and RPi 3 B+
> because of throttling differences, and the metal heat spreader.
> 
> Ambient temperature is critical for the separate chip used for
> Ethernet and USB; it has a lower maximum operating temperature than
> the CPU.
> 
>  
> Do you happen to know this chip's max operating temperature spec?

Only 70 degrees C.

http://www.microchip.com/wwwproducts/ProductCompare/LAN7515/LAN9514

This looks like a binning threshold.

Let me spin a story.  The manufacturer of the chip tests the operating
temperature range using some method, and separates the results into
four bins;

- operates above 86, into the bin labelled "excellent",

- operates above 70 and below 86, into the bin labelled "industrial",

- operates above 60 and below 71, into the bin labelled "commercial",

- doesn't operate above 70, into the bin labelled "throw away".

Then what happens next is often surprising;

1.  if a customer orders the industrial version, it will be filled
from the "industrial" and then the "excellent" bin,

2.  if a customer orders the commercial version, it will be filled
from the "commercial", "industrial" and then the "excellent" bins.

The chips can be encapsulated and labelled at any time, before or
after this binning.

So you as customer of customer only knows it will operate to 70, and
it might operate to 85.

The chip heats from being powered, and from Ethernet and USB activity.

The heat from the CPU will bleed across to the chip.  In a normal
enclosure, this isn't a problem.  In a sealed enclosure with no air
flow, and high CPU, Ethernet and USB activity, it could be more
interesting.

It would be reasonably easy to test for; if the CPU temperature does
not fall rapidly when activity slows, it is because the board and
air nearby has reached equilibrium with the CPU.

> Your best comparison will be between timings of equal workloads with
> equal cooling in identical environment.
> 
> You can't do this honestly with temperature alone.
>
> --
> James Cameron
> [2]http://quozl.netrek.org/
> ___
> Server-devel mailing list
> [3]Server-devel@lists.laptop.org
> [4]http://lists.laptop.org/listinfo/server-devel
> 
> --
> [5]
> [6]Unsung Heroes of OLPC, interviewed live @ [7]http://unleashkids.org !
> 
> References:
> 
> [1] mailto:qu...@laptop.org
> [2] http://quozl.netrek.org/
> [3] mailto:Server-devel@lists.laptop.org
> [4] http://lists.laptop.org/listinfo/server-devel
> [5] http://lists.laptop.org/listinfo/server-devel
> [6] http://lists.laptop.org/listinfo/server-devel
> [7] http://unleashkids.org/

> ___
> Server-devel mailing list
> Server-devel@lists.laptop.org
> http://lists.laptop.org/listinfo/server-devel


-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: [Server-devel] do RPi 3 B+ still need heat sinks?

[James Cameron]

Temperature readings are not comparable between RPi 3 and RPi 3 B+
because of throttling differences, and the metal heat spreader.

Ambient temperature is critical for the separate chip used for
Ethernet and USB; it has a lower maximum operating temperature than
the CPU.

Your best comparison will be between timings of equal workloads with
equal cooling in identical environment.

You can't do this honestly with temperature alone.

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

[Server-devel] Test

[James Cameron]

Test message.  Random words.

akimbo halfpennies lama griffins regressed happen priors maternal
davis torrens devalues sven discombobulates convection determinant
equestrian grungier mornings launching stallone werewolves ayala
karroo spontaneity convex salivated waterspout awing creditors
manipulates dispensation brigands earthier evocation repudiating
crater busts scrubbers nutrias buffering parkway sawmill courtlier
griffins cloakroom subtle gayle aberration la submarines astrophysics
spineless feats overcasts proper tucson imagine brobdingnag marc
percentile marks thousand tailspins engorges qatar emulation pet
spattering pedro cleat overspreading disease windmilling saigon
chances informally corkscrews charismatics chagrins microscopy stabbed
discredit watchwords indispensables hyphenated pandemic smugger
uncoordinated combating sequencer trailing agricola striking
inextricably allegories funnels importuning scions calvinist
underwriting chartering trackers dewdrops tutoring muscle retorted
barefooted nautili ports esoterically guess hoarding bludgeons buggies
previewers doggy admonished flooder currants swedes shortcakes subsume
ruination stumping paramedic dispossessed crucifixes socialise

-- 
James Cameron
http://quozl.netrek.org/


signature.asc
Description: PGP signature
___
Server-devel mailing list
server-de...@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Test

[James Cameron]

Test message.  Random words.

akimbo halfpennies lama griffins regressed happen priors maternal
davis torrens devalues sven discombobulates convection determinant
equestrian grungier mornings launching stallone werewolves ayala
karroo spontaneity convex salivated waterspout awing creditors
manipulates dispensation brigands earthier evocation repudiating
crater busts scrubbers nutrias buffering parkway sawmill courtlier
griffins cloakroom subtle gayle aberration la submarines astrophysics
spineless feats overcasts proper tucson imagine brobdingnag marc
percentile marks thousand tailspins engorges qatar emulation pet
spattering pedro cleat overspreading disease windmilling saigon
chances informally corkscrews charismatics chagrins microscopy stabbed
discredit watchwords indispensables hyphenated pandemic smugger
uncoordinated combating sequencer trailing agricola striking
inextricably allegories funnels importuning scions calvinist
underwriting chartering trackers dewdrops tutoring muscle retorted
barefooted nautili ports esoterically guess hoarding bludgeons buggies
previewers doggy admonished flooder currants swedes shortcakes subsume
ruination stumping paramedic dispossessed crucifixes socialise

-- 
James Cameron
http://quozl.netrek.org/


signature.asc
Description: PGP signature
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

[Server-devel] Test

[James Cameron]

Test message.  Random words.

akimbo halfpennies lama griffins regressed happen priors maternal
davis torrens devalues sven discombobulates convection determinant
equestrian grungier mornings launching stallone werewolves ayala
karroo spontaneity convex salivated waterspout awing creditors
manipulates dispensation brigands earthier evocation repudiating
crater busts scrubbers nutrias buffering parkway sawmill courtlier
griffins cloakroom subtle gayle aberration la submarines astrophysics
spineless feats overcasts proper tucson imagine brobdingnag marc
percentile marks thousand tailspins engorges qatar emulation pet
spattering pedro cleat overspreading disease windmilling saigon
chances informally corkscrews charismatics chagrins microscopy stabbed
discredit watchwords indispensables hyphenated pandemic smugger
uncoordinated combating sequencer trailing agricola striking
inextricably allegories funnels importuning scions calvinist
underwriting chartering trackers dewdrops tutoring muscle retorted
barefooted nautili ports esoterically guess hoarding bludgeons buggies
previewers doggy admonished flooder currants swedes shortcakes subsume
ruination stumping paramedic dispossessed crucifixes socialise

-- 
James Cameron
http://quozl.netrek.org/


signature.asc
Description: PGP signature
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: can one "dd" XO-1.5 internal microSD cards, to another XO-1.5?

[James Cameron]

On Mon, Feb 19, 2018 at 10:13:51PM -0800, Hal Murray wrote:
> > Better would be to use SD cards.
> 
> The original goal was to copy microSD cards.

Yes, but original goal unwise; it will damage the microSD card
connector and display cable.  At this stage of life there is
contact oxidation, spring tension reduction, and brittle polymers.

I've observed these in my test units that have had way less use than
units in the hands of children.

> Would it be reasonable to copy microSD to SD, move the SD card to
> the target XO, and then copy SD to microSD?  I expect an
> OpenFirmware wizard could do it with no effort but I'm not that
> level of wizard.

No, not really.

An order of magnitude faster to use Linux to copy, as it can do
asynchronous I/O.  Open Firmware only does asynchronous I/O during
fs-update.

To access my wizardry, use the tools we have refined and kept running
for years, or get me to do it.  Can also sign the build; overcomes the
next common problem; locked laptops.

The .zd format and fs-update is the fastest method to bulk load the
microSD card without having to remove and replace it.

A clone can be made with Linux, then converted to .zd using the
zhashfs program.

There is also rsync from a teacher laptop, see
http://wiki.laptop.org/go/Tiny_Core_Linux/Reversion

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: can one "dd" XO-1.5 internal microSD cards, to another XO-1.5?

[James Cameron]

Stripped of your superlatives, yes.

It isn't just .sugar, there's also .config.

http://wiki.laptop.org/go/Imaging/Side_effects describes the exact
steps to do before duplication.

Get it wrong and you will break collaboration of Sugar activities.

Certainly unfortunate that said Ace Teacher isn't Ace enough to find
or develop expertise.

The microSD card connector has a limited life, as does the display
cable, so I'm all for opening and closing it many times so you have to
buy new laptops!  ;-)  Better would be to use SD cards.

On Mon, Feb 19, 2018 at 05:10:58PM -0500, Adam Holt wrote:
> James & All,
> 
> Will it work if one extricates an XO-1.5's internal microSD card and then
> places it inside another XO-1.5?
> 
> e.g. allowing an Ace Teacher to make her own customized images offline in
> Haiti, with the power of a screwdriver, then simply re-assembling her very
> small collection of XO-1.5's after the "dd" duplication of microSD's is done.
> 
> If this works, I will ask her to run "rm -rf /home/olpc/.sugar" first, in 
> order
> to delete all personal info on the master/template laptop, if she really 
> cares!
>   Context for this potential Lifesaver: the Ace Teacher does not and cannot
> realistically use these more professional tools that we are would normally
> advocate for:
> 
> [1]http://wiki.laptop.org/go/OS_Builder
> [2]http://wiki.laptop.org/go/Tiny_Core_Linux
> [3]http://wiki.laptop.org/go/Customization_stick (deprecated!)
> 
> References:
> 
> [1] http://wiki.laptop.org/go/OS_Builder
> [2] http://wiki.laptop.org/go/Tiny_Core_Linux
> [3] http://wiki.laptop.org/go/Customization_stick

> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel


-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: "Can't open boot device" XO laptop firmware/OFW issue?

[James Cameron]

As it ages, the microSD card takes longer and longer to erase and
write blocks.

Eventually the time it takes to erase and write may exceed the few
seconds between end of reflashing and the reboot power cycle.

In those situations, use fs-update instead of the four button method,
and leave the system on for five minutes before the first reboot.

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: "Can't open boot device" XO laptop firmware/OFW issue?

[James Cameron]

Either there is an SD card in the external slot with unsupported partition
table, or there is a fault with the internal microSD card or slot.

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: [Server-devel] serious Raspbian WiFi flaw discovered: works for 10SEC then cycles OFF/ON

[James Cameron]

On Tue, Jan 02, 2018 at 06:57:43AM +, Peter Robinson wrote:
> >> We need to report this to the Raspberry Pi Foundation to see if they can
> >> fix it -- WiFi connections are repeatedly failing, sometimes even very soon
> >> after booting, profoundly affecting Internet-in-a-Box 6.5 !
> >>
> >> But first a big Thanks In Advance to all who can reproduce this & offer
> >> your own experiences/perspectives prior to Thursday's [*] call:
> >>
> >>http://minutes.iiab.io
> >>
> >> Here are 2 (known) ways to reproduce the WiFi bug with near certainty:
> >>
> >>https://github.com/iiab/iiab/issues/638#issuecomment-354639673
> >
> >
> > CLARIF: I do *not* mean to blame Raspbian, as we seek a solution here!!
> >
> > At this this point it's entirely possible the underlying/root cause is
> > dhcpcd...and/or even RPi3 firmware / WiFi module etc?
> 
> Is it reproducible with upstream kernel/firmware on something like an
> up to date Fedora install, there was some new firmware pushed upstream
> that fixed some CVEs in the Broadcom firmware (CVE-2016-0801,
> CVE-2017-0561, CVE-2017-9417), I'm not sure which firmware's were
> affected/updated but it might be worth checking to see if it's fixed
> on other distros that are closer to upstream.

Agreed.  Well worth a try with Fedora.  Raspbian is a long way behind,
so it is not easy to test recent fixes.

Most recent data from Adam; his Raspberry Pi is being connected by
wireless to one router with IP of 192.168.0.1, then connected by
ethernet to a physically different router with same IP of 192.168.0.1.

I'm thinking that's an exciting and interesting thing to do, and I'm
fascinated to learn how it will pan out.

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: The needed patches to run new systemd on 3.3 kernel

[James Cameron]

Thanks; indeed useful where kernel is not ported.

On Thu, Dec 28, 2017 at 11:30:22PM -0600, Ethan wrote:
> Just thought it would be useful for the XO-1/1.5

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: "Show sample projects" button in Physics activity v32.1 on OLPC OS 13.2.9

[James Cameron]

On Tue, Dec 26, 2017 at 07:51:47AM -0500, Adam Holt wrote:
> Just FYI the Physics activity has a button in the top-right ("Show
> sample projects") that momentarily flashes a large black rectangle
> onto the screen, but does not proceed.
> 
> This was tested in Physics activity version 32.1 on OLPC OS 13.2.9
> on XO-1.5 and XO-1.75, by right-clicking and also with a (regular)
> left-clicking on the button.
> 
> Is there a known workaround to load sample projects possibly?

Thanks.  This was reported by Jammily and fixed in August
https://github.com/sugarlabs/physics/issues/32

Physics-32.2 has been uploaded for OLPC OS 13.2.x.  Use alt-shift-m My
Settings, Software Update.

> And/or should we document this in
> http://wiki.laptop.org/go/Release_notes/13.2.9#Known_Problems ?

Wouldn't want to set a high expectation; resource intensive to
add notes for each fix to each activity.  History of page
http://wiki.laptop.org/go/Activities/13.2.9 could help.

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Announcing OLPC OS 13.2.9 with Sugar 0.112

[James Cameron]

G'day,

We're pleased to announce the release of OLPC OS 13.2.9 for XO-1,
XO-1.5, XO-1.75 and XO-4.

It is Sugar 0.112 on Fedora 18, with updated activities Browse-157.4,
Calculate-44, Chat-84, Clock-19, GetBooks-17, Help-20.1,
ImageViewer-63, Implode-18, Jukebox-33, Log-38, Maze-27, Memorize-53,
MusicKeyboard-8.9, Paint-66, Physics-32.1, Pippy-71.1, Read-118.1,
StopWatch-19.1, Terminal-45.2, TurtleBlocks-215 and Write-98.3.

Details of new features, fixes, known issues, and how to download,
install or upgrade can be found in the release notes:

http://wiki.laptop.org/go/Release_notes/13.2.9

Our thanks to all contributors, testers, upstreams, and those who
have provided feedback of any kind.

Build file names: 32021o0 32021o1 32021o2 32021o4

-- 
James Cameron
http://quozl.netrek.org/


signature.asc
Description: PGP signature
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: [Server-devel] PRE-Released: Internet-in-a-Box 6.5 Preview 2

[James Cameron]

On Fri, Nov 10, 2017 at 03:43:50PM -0500, Adam Holt wrote:
> On Fri, Nov 10, 2017 at 12:26 PM, Adam Holt <[1]h...@laptop.org> wrote:
> 
> Including [2]IIAB 6.5 PRE-Release Notes :-)
> 
> But just a warning that Raspbian's mirrors (download servers) were barely
> working at all over the past ~12 hours or so, so you might have much 
> better
> luck installing on Ubuntu 16.04 LTS, Debian 9.2.1 or possibly CentOS 7.4 
> in
> the coming hours?
> 
>    [3]https://github.com/iiab/iiab/releases
> 
> Reconfirmed: disturbingly [4]http://mirrordirector.raspbian.org is 
> hemorrhaging
> (works less than 50% of the time today, as [5]http://
> downforeveryoneorjustme.com/mirrordirector.raspbian.org confirms) for almost
> 24h now.

Only the redirector affected.  Can temporarily use a nearby mirror
site if you know the URL and can change configuration.  Not an option
for ease of use though.

> Consequence: those installing on Raspbian are having to run "./iiab-install"
> 10+ Times In A Row today, just to finally get a regular install (e.g. 
> [6]http:/
> /download.iiab.io/6.5/rpi/load.txt) to incrementally/finally complete on
> Raspberry Pi 3.  What normally would just happen on its own within about an
> hour.
> 
> Better to take the day off !
> 
> ( Possibly someone knows how we can build more resiliency into Raspbian 
> package
> downloads/updates in future, using [7]https://www.raspbian.org/RaspbianMirrors
> or similar? )

Make a preinstalled image and deliver that instead.  That's what we
did with the XO.  With everything already included, you just have to
get one file, and don't need precious internet.

> 
> What's new?  Our prototype replacement for ./runansible allows 
> installation
> of [8]IIAB 6.5 Preview 2 to progress far more efficiently, far more
> readably -- and to recover roughly where it left off during a large
> net-based install, avoiding repetition!  So once you have [9]all your OS
> prereqs in place, you can now run IIAB's new 9-stage / incremental
> installer as follows:
> 
>    cd /opt/iiab/iiab
>    ./iiab-install
> 
> NEW: it will even warn you if your Ansible and/or Raspbian kernel are not
> recent enough, and guide you to fixing these 2 most essential prereqs! 
> Finally, most installers/implementers will still want to use a 1-line
> install/load scripts -- to walk away from the keyboard and have most
> everything taken care of.  As such, YES all our install/load scripts now
> take advantage of "./iiab-install" 9-stage incremental/recoverable 
> installs
> too:
> 
>    [10]http://download.iiab.io/6.5/rpi/
> 
> RECAP #1: if you suffer from a connectivity failure during installation of
> IIAB, the vast majority of problems are quickly solved by re-trying "./
> iiab-install" (after Raspbian mirroring sites hopefully come back to life
> later today!)
> 
> RECAP #2: Raspbian folk...until the new Raspbian OS (point release) is
> officially released in coming days by the Raspberry Pi Foundation & All,
> avoid kernel compatibility hassles using this temporary measure to prep
> your OS properly, just prior to installing:
> 
>    apt update
>    apt install raspberrypi-kernel
>    reboot
>    (then install IIAB as above, either way!)
> 
> Thanks all who've helped make this an incredibly high-quality pre-release 
> already, just a few short weeks away from its final release sometime later
> this November!!
> 
> --
> Unsung Heroes of OLPC, interviewed live @ [11]http://unleashkids.org !
> 
> References:
> 
> [1] mailto:h...@laptop.org
> [2] https://github.com/iiab/iiab/wiki/IIAB-6.5-Release-Notes
> [3] https://github.com/iiab/iiab/releases
> [4] http://mirrordirector.raspbian.org/
> [5] http://downforeveryoneorjustme.com/mirrordirector.raspbian.org
> [6] http://download.iiab.io/6.5/rpi/load.txt
> [7] https://www.raspbian.org/RaspbianMirrors
> [8] https://github.com/iiab/iiab/releases
> [9] 
> https://github.com/iiab/iiab/wiki/IIAB-Installation#do-everything-from-scratch
> [10] http://download.iiab.io/6.5/rpi/
> [11] http://unleashkids.org/

> ___
> Server-devel mailing list
> Server-devel@lists.laptop.org
> http://lists.laptop.org/listinfo/server-devel


-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

activation.laptop.org unavailable on Sunday 5th November

[James Cameron]

activation.laptop.org and antitheft.laptop.org will be unavailable for
up to 12 hours on Sunday 5th November 2017.  We will be upgrading the
database server.

+BCC all registered users active in past year.

-- 
James Cameron
http://quozl.netrek.org/


signature.asc
Description: PGP signature
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: [Server-devel] [UKids] Re: [XSCE] Internet-in-a-Box 6.4 is 79% Complete according to GitHub :-)

[James Cameron]

You might mirror the content to your own servers, or use other mirrors.

On Fri, Sep 22, 2017 at 10:16:10AM -0400, Adam Holt wrote:
> Thanks César Natarén for confirming KA Lite's server to download
> exercises+assessments is once again working today:
> 
> [1]https://github.com/iiab/iiab/issues/264
> [2]https://github.com/iiab/iiab/issues/257
> 
> (The fact that KA Lite's server was unavailable for downloads 2 days out of 
> the
> last 4 is indeed very worrisome.  We really do need a better way to trap such
> commonplace glitches, so that newcomers are not completely overwhelmed by
> poorly explained Big Red Errors!)
> 
> On Sep 22, 2017 12:50 AM, "James Cameron" <[3]qu...@laptop.org> wrote:
> 
> There will always be connection failures, and they aren't always the
> fault of [4]learningequality.org.  Locally broken WiFi can do it too.
> 
> Ask ka-lite team to add retry when the download is obviously not a Zip
> file, or to retry anyway.
> 
> They've certainly had problems with content pack downloads.
> 
> On Fri, Sep 22, 2017 at 12:37:23AM -0400, Adam Holt wrote:
> > On Thu, Sep 21, 2017 at 11:59 PM, Rick Castorani <[1][5]
> rcastor...@gmail.com>
> > wrote:
> >
> >     The script just finished installing on an RPI 3. Ran into a kalite
> error on
> >     the last command from [2][6]download.iiab.io/6.4/rpi/load-vpn.txt
> >
> >     See error in attachment (sorry for the pic, ran script locally on 
> the
> pi).
> >     Pi did not reboot due to the error.​
> >     [3] IMG_20170921_215647.jpg
> >
> > Thanks Rick for speaking for everyone (not just Internet-in-a-Box
> implementers,
> > but all KA Lite users) facing this together!
> >
> > Your Internet-in-a-Box install is 100% complete (feel free to reboot!)
> but yeah
> > we need to do better explaining this extreme annoyance to newcomers --
> just
> > like on Monday / Sept 18 -- when KA Lite's servers were similarly
> > misconfigured-or-down for almost the entire day.
> >
> > Meaning nobody can auto-install KA Lite's (mandatory to use the 
> product!)
> > English assessments/exercises content pack from the GUI ([4][7]http://
> box:8008 ->
> > Login -> Manage tab -> Language tab) *or* from the command line...as you
> > basically did with "kalite manage retrievecontentpack download en".  For
> now
> > all we can do is wait until KA Lite's web server(s) come back online
> properly.
> >
> > But those who can help KA Lite & all of us nail what to do about KA
> Lite's
> > sometimes unreliable servers please do post to:
> >
> > [5][8]https://github.com/iiab/iiab/issues/264 "kalite manage
> retrievecontentpack
> > download en" fails on new install (bad zip file?)
> > [6][9]https://github.com/iiab/iiab/issues/257 "kalite manage
> retrievecontentpack
> > download en" fails nightly
> > [7][10]https://github.com/learningequality/ka-lite/issues
> >
> > Error posted here, a bit more readably than Rick's photo:
> >
> > [INFO] [2017-09-21 22:03:37,292] kalite: Downloading content pack from
> [8]http:
> > //[11]pantry.learningequality.org/downloads/ka-lite/0.17/content/
> contentpacks/
> > en.zip
> > Traceback (most recent call last):
> >   File "/usr/local/kalite/venv/local/lib/python2.7/site-packages/kalite/
> > packages/bundled/django/core/management/base.py", line 224, in
> run_from_argv
> >     self.execute(*args, **options.__dict__)
> >   File "/usr/local/kalite/venv/local/lib/python2.7/site-packages/kalite/
> > packages/bundled/fle_utils/django_utils/command.py", line 141, in 
> execute
> >     super(LocaleAwareCommand, self).execute(*args, **kwargs)
> >   File "/usr/local/kalite/venv/local/lib/python2.7/site-packages/kalite/
> > packages/bundled/django/core/management/base.py", line 263, in execute
> >     output = self.handle(*args, **options)
> >   File "/usr/local/kalite/venv/local/lib/python2.7/site-packages/kalite/
> > distributed/management/commands/retrievecontentpack.py", line 90, in
> handle
> >     self.download(*args, **options)
> >   File "/usr/local/kalite/venv/local/lib/python2.7/site-packages/kalite/
> > distributed/management/commands/retrievecontentpack.py", line 147, in
> download
> >     z

Re: [Server-devel] [UKids] Re: [XSCE] Internet-in-a-Box 6.4 is 79% Complete according to GitHub :-)

[James Cameron]

uot; group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [15]unleashkids+unsubscr...@googlegroups.com.
> For more options, visit [16]https://groups.google.com/d/optout.
> 
> --
> Unsung Heroes of OLPC, interviewed live @ [17]http://unleashkids.org !
> 
> References:
> 
> [1] mailto:rcastor...@gmail.com
> [2] http://download.iiab.io/6.4/rpi/load-vpn.txt
> [3] 
> https://drive.google.com/file/d/10jhRj47EEpjSWL8MR7JHgJDOkfy8LMBMIg/view?usp=drive_web
> [4] http://box:8008/
> [5] https://github.com/iiab/iiab/issues/264
> [6] https://github.com/iiab/iiab/issues/257
> [7] https://github.com/learningequality/ka-lite/issues
> [8] 
> http://pantry.learningequality.org/downloads/ka-lite/0.17/content/contentpacks/en.zip
> [9] mailto:h...@laptop.org
> [10] https://github.com/iiab/iiab/milestone/1
> [11] https://github.com/iiab/iiab/releases/
> [12] http://download.iiab.io/6.4/rpi/
> [13] http://unleashkids.org/
> [14] http://unleashkids.org/
> [15] mailto:unleashkids+unsubscr...@googlegroups.com
> [16] https://groups.google.com/d/optout
> [17] http://unleashkids.org/

> ___
> Server-devel mailing list
> Server-devel@lists.laptop.org
> http://lists.laptop.org/listinfo/server-devel


-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: [Server-devel] "Raspbian reserves 4% of all disk space"

[James Cameron]

No sympathy; you asked for this problem, and you got it.

Fix the underlying problem, which is either training, confidence,
remote support, brain drain, or SD card is too small for the job,

Automatically delete least used content once the limit is hit.

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: Using a official kernel

[James Cameron]

Please don't break threads; open the message you want to reply to, and
then press reply.

https://en.wikipedia.org/wiki/Conversation_threading

On Wed, Jul 26, 2017 at 12:32:22AM -0400, Jhon Diaz wrote:
> I used a official kernel 3.10

That doesn't tell me much.  We had several 3.10
kernel RPMs, and each was built from different git hashes.

http://rpmdropbox.laptop.org/f20-xo1/

> turns out its a problem with busybox in the actrd.img ill try a
> custom initramfs

No point doing that, because it won't be signed, therefore it won't be
run by the firmware.

You'll have to find out what the problem actually is.

> i can provide the log file if you want me to

Of what?

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

jffs2 mounting by actrd, [Was: I need help]

[James Cameron]

On Wed, Jul 26, 2017 at 12:11:23AM -0400, Jhon Diaz wrote:
> When i try to boot my build using actrd, dracut gives me a warning

actrd.zip should only be loaded when a laptop is being booted
securely, when it does not have a developer key, or when the ❌ (cross
mark) game pad key is used to force secure boot when in non-secure
mode.  Purpose of actrd.zip is to obtain a lease or deny use of the
laptop.

http://wiki.laptop.org/go/Firmware_security has more details, as does
the kernel git repository directory olpc/.

What was the warning?  If it goes past too fast, you might be able to
catch it with a serial console, see
http://wiki.laptop.org/go/Serial_adapters

> then spits this out
> mount: mounting /dev/mtd0 on /sysroot failed: Block device required
> I don't know what to do?

Debug it.

Does your kernel have jffs2 support enabled?

Look at the code being run.  You'll find it in the OLPC dracut source
repository.  http://dev.laptop.org/git/users/quozl/dracut-modules-olpc/

Use an SD card or USB drive to boot the laptop and look carefully at
the jffs2 filesystem.

http://wiki.laptop.org/go/Tiny_Core_Linux
http://wiki.laptop.org/go/Cheat_codes

Also, we have several different actrd.zip files, make sure you are
using the right one for the kernel.  It should be in the kernel RPM.

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: How do i test my jffs2 images?

[James Cameron]

When you use mkfs.jffs2 you will have an .img file.

Firmware uses a CRC file to check the IMG file.

Generate a CRC file using the sumtool and crcimg commands, see how we
do it in our builder;


http://dev.laptop.org/git/projects/olpc-os-builder/tree/modules/jffs2_image/image.50.makeimg.sh

sumtool is part of mtd-utils package.

crcimg is a binary from the crcimg package on Fedora 18, or there is a
perl script crcimg.pl in xodist,

git clone http://dev.laptop.org/~quozl/xodist.git

Put the two files on a USB drive, e.g. file.img and file.crc

On the XO-1, get to the Ok prompt in the firmware, then

ok copy-nand u:\file.img

Also at the ok prompt you can explore the filesystem

ok dir nand:\
ok dir nand:\boot\

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Working script to produce a debian chroot for usb

[James Cameron]

There's a Debian 8 (Jessie) guide on the Wiki.

http://wiki.laptop.org/go/Debian_Jessie_on_XO

Agreed, Debian 9 (Stretch) is minimal version now, so I cannot
recommend Jhon's Debian 6 (Squeeze) method as it is unsupported and
has security vulnerabilities.  It should be updated to Stretch.

An image installable with the "four game key buttons held down" method
will require either;

- signing with the deployment keys,

- injecting new deployment keys, or;

- signing for the built-in OLPC keys.

I'll be happy to sign an image of Debian 9 if it can be reproduced
simply, has our actrd.zip file in /boot, and has Sugar or MATE
desktops.

On Sun, Jul 23, 2017 at 09:27:23PM -0500, Sebastian Silva wrote:
> Squeeze? :-/ That's Debian 6, which was supported until Feb 2016.
> 
> That doesn't help much. Also you don't say what model XO this is for.
> 
> I'd be very interested in a Debian 9 builder that one could install with
> the game buttons on XO1 and 1.5. There are still many of those in Peru.
> 
> Regards,
> 
> Sebastian
> 
> 
> 
> On 23/07/17 20:32, Jhon Diaz wrote:
> > Well i did it
> > https://github.com/electimon/xobuild
> > It makes a debian squeeze system with lxde installed in
> > /tmp/xobuild.0/build/ just copy it to your ext3 formatted drive the
> > password is olpc with a user named olpc
> > its in alpha so please report any bugs if it doesn't work for you.
> > Im adding more options hopefully supporting ubuntu and opensuse later on.
> > ___
> > Devel mailing list
> > Devel@lists.laptop.org
> > http://lists.laptop.org/listinfo/devel
> 
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: [Server-devel] Internet-in-a-Box 6.3 PRE-RELEASE installs w/o a hitch onto Raspbian!

[James Cameron]

G'day Adam,

Congratulations on the continued development and interest in this
derivative of XS.

As this is your second re-branding, perhaps it is time for your team
to re-consider your use of wiki.laptop.org and h...@laptop.org?

It must be confusing for your customers to see laptop.org mixed into
the resources you refer to.

Also, faq.iiab.io redirects to wiki.laptop.org, and isn't available
over HTTPS, (invalid certificate, unconfigured virtual host), so
you'll lose anybody using automatic encryption.

We still have our wiki.laptop.org instance running, with daily
backups, so you can continue to use it, but you're the only one
editing those pages.  Let me know if anybody else needs an account.

I've adjusted our disclaimer template again.

-- 
James Cameron
http://quozl.netrek.org/

On Tue, Jun 13, 2017 at 06:28:06AM -0400, Adam Holt wrote:
> George Hunt loudly curses my name every day ...
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: [Server-devel] Internet-in-a-Box 6.3 PRE-RELEASE installs w/o a hitch onto Raspbian!

[James Aubury]

Sounds good!
I'll give it a test to my Pi3 HDD setup in the next couple of days and see 
what happens.

On Tuesday, 13 June 2017 11:28:07 UTC+1, holt wrote:
>
> George Hunt loudly curses my name every day when he remembers he gave up a 
> comfortable retirement on the gold course to... *write software for 
> children + medical students in emerging countries instead ???*
>
> But the communities in the Dominican Republic and Haiti love him 
> nonetheless, as in coming weeks they will receive the product of his epic 
> work...
>
> Thanks in no small part to Tim Moody & Josh Dennis pulling off a very 
> challenging refactoring over the past weeks, rebasing a decade's worth of 
> code into far-more future-proof repositories, under our new 
> "Internet-in-a-Box" name ( http://github.com/iiab !)
>
>
> *So for those who now want a sneak preview install on Raspbian Pixel or 
> Raspbian Lite immediately, our 1-line installer is:*
>curl download.iiab.io/6.3/rpi/load.txt | bash
>
> For me this completes in 1.5 hours, essentially flawlessly bringing us all 
> a very powerful new version of Kiwix -- but do let us know what works for 
> you!  And just for fun, read the instructions here too, so you learn 
> from+with others:
>
>http://wiki.laptop.org/go/IIAB/6.3
>
> Know Issues:
>
>- your offline documentation (http://box/info) has a couple glitches 
>we'll be working on in coming weeks
>- some commands within /usr/bin have not yet been convert from prefix 
>"xs-" to prefix "iiab-"
>- a reboot is needed after installation, if you want OpenVPN to work
>
>
> PS final release date is undecided, but hopefully this will happen well 
> within the coming weeks!  There is so much to offer from our repertoire of 
> rapid community action / deployment efficiencies (even an 
> smartphone-friendly "Instant Off" button requested by medical clinics!)  
> But much more about our rising UX focus in coming weeks, as we write up our 
> release notes~  FYI far cleaner TeamViewer integration onto Raspbian, as 
> summarized here within http://FAQ.IIAB.IO here is another 
> transformational step forward allowing churches & small NGO's (even 
> sometimes Peace Corps voluntters in the field etc) to *co-develop their 
> community's very own Internet-in-a-Box*, even while separated by large 
> distances:
>
>FAQ #21 How can I remotely manage my Internet-in-a-Box? 
> 
>
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: [Server-devel] ssh.service error on CentOS 7.3

[James Cameron]

On Fri, May 26, 2017 at 09:40:05AM +0100, Peter Robinson wrote:
> It's sshd.service not ssh.server as the error indicates. Make sure you
> have openssh-server package installed.

Good catch.  Possibly non-portable code; the name is ssh.service on
Debian derived systems, but sshd.service on Fedora derived systems.

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: [Server-devel] Apache 2.4.6 on CentOS and 2.4.10 on Debian/Raspbian

[James Cameron]

No, that's the wrong approach.

Version numbers cannot be compared, because both CentOS and Debian
have backported later changes.

Instead, look at the change log for 2.4.25 and every prior version
back to the version you have, for changes that are important to you,
especially changes with a CVE number or tagged SECURITY.

That gives you a list of changes you want to have.

Then, focus on the changes that are likely to impact server
operations, such as privelege escalation or denial of service.

Then, look at the change log for the CentOS and Debian packages,
looking for where they have backported the changes.  For Debian you'll
find this in /usr/share/doc/apache2/changelog.Debian.gz

It is a complex process, which is why most people delegate it to
CentOS and Debian security teams.

And to answer your question; the particularly important risks that
Internet-in-a-Box may face are all the SECURITY and CVE tagged changes
in the 2.4 series change log;

http://www.apache.org/dist/httpd/CHANGES_2.4

The most important one appears to be CVE-2016-8740 for a denial of
service vulnerability.

Risk is high if the server is accessed from the internet.

Risk is medium if the server is accessed by local public wireless.

Risk is low if the server is accessed by password protected wireless.

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: [Server-devel] ssh.service error on CentOS 7.3

[James Cameron]

Explore /etc/profile.d and look at each script added by xsce to find
one that tries to do "/sbin/service" commands or mentions ssh.

If that doesn't help, start a bash shell without profile
(--noprofile), and then execute each file in turn until you find the
culprit.

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: [Server-devel] ditch all *.img.md5.txt ? (*.img.zip.md5.txt far more useful)

[James Cameron]

Too much noise is a good argument.

Solutions can be;

0.  omit the checksums if they are used less than 10% of the time
(check your logs),

1.  use one file to contain all checksums, e.g. MD5SUMS,

2.  don't use Apache indexes, but instead use a script or template to
generate friendly HTML with file name, checksum, on one line,

Also, it surprises me you-all aren't using any digital signatures on
the files; but then I think of your user base who wouldn't necessarily
have the time to be able to verify one, let alone an MD5.

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: [Server-devel] "systemctl restart xsce-cmdsrv" &/or other stuff when Admin Console crashes?

[James Cameron]

My gut feel is electrostatic lockup, radio interference, cosmic rays,
or a momentary DC bus collapse leading to memory corruption or
filesystem remount readonly.

You might look at logs before the event, to see if the logs stop
suddenly or if they proceed to your reboot.

Next time it happens, collect dmesg and system logs.

To avoid it, use the authorised USB power supply adapter, keep the Pi
cool, and protect the top of the Pi from any nearby hands or radios.

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: Help finding xo upgrade kit

[James Cameron]

Thanks for your interest.

We make upgrade kits in large quantities as special orders for
schools and other organisations.

We would not normally keep any stock ourselves, because of the
warehousing costs and the low demand.

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: [Server-devel] in plain english: How to Add Content to Internet-in-a-Box

[James Cameron]

On Thu, Mar 16, 2017 at 09:19:02PM -0400, Adam Holt wrote:
> Today I've overhauled most of the "Add Content" section at the
> bottom of our Internet-in-a-Box (formerly XSCE) install doc:
> 
> https://github.com/XSCE/xsce/wiki/IIAB-Installation#add-content

It is, frankly, frightening.  Write for your audience, and close the
loop with your audience.  Make the process easier, cleaner, and write
without disclaimers, warnings, or redundant expressiveness.  It should
also be less text, more pictures.

As an example of simplified content bundling, have a look at Endless
OS content apps, and Khan Academy Lite's built-in video acquisition UX.

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: [IAEP] do OLPC browsers support WebRTC microphone audio input?

[James Cameron]

G'day Jim,

I can recommend our OLPC NL3 with OLPC OS 16.04.2 and Firefox-52 or
Chrome-56 for WebRTC microphone input.

Several other combinations do not work.  Here are details with
footnotes.

- OLPC OS 16.04.2 on NL3 has Sugar Browse-200.1 [A] and Firefox-52
  [W], and can install Chrome-56 [W].

- OLPC OS 13.2.8 on XO-4 has Sugar Browse-157.3 [A], and can install
  Firefox-25 [B].  Chrome is not available.

Footnotes:

Failure mode "A"; pocketsphinx.js acoustic scores demo does not work;
the start button is insensitive.  Because the WebRTC layer does not
support the feature.

Failure mode "B"; "Would you like to share your microphone with
snd95.github.io" is asked, the microphone light turns on, but failure
mode "A" then occurs.

Working mode "W", works fine.

p.s.

Questions about OLPC OS are best posted by subscribing to
devel@lists.laptop.org as an authoritative source.  Let me know if you
have trouble subscribing.  I expect my reply will be bounced by IAEP.

Thanks to Adam for cross posting and Sam for data about OLPC OS
13.2.8.

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: [Server-devel] shutdown/bootup snafus+delays itemized (IIAB/XSCE Raspian on RPi3)

[James Cameron]

On Sun, Feb 26, 2017 at 02:28:56PM -0500, Adam Holt wrote:
> Conclusion: apt-get or "cd /opt/schoolserver/xsce; git pull" similar
> must have applied some changes in recent days, as the shutdowns have
> now reverted to the Linux-style stream of small messages scrolling
> up the screen.

Yeah, starting to sound self-inflicted.

Break your changes into smaller pieces and test shutdown times after
each piece.

You'll eventually find which piece is causing it.

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: [Server-devel] RPi3 "poweroff" was consistently ~15sec, now consistently ~1.5min

[James Cameron]

Ninety second delay is double a systemd default timeout for services
not responding to shutdown request.

Three minutes is a quadruple.  We saw similar with OLPC OS and systemd
and fixed the offending services.

It will take a systemd debugging session to find which services are
the cause.

Cause might or might not be MariaDB, but it's a good candidate to
investigate; stop it first using systemctl, then see if shutdown is
faster.  Pre-shutdown shutdowns can be a quick and easy test, but may
still need in-depth debugging.

When shutdown does not work properly, RTC synchronisation may also be
affected.

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: [Server-devel] auditing our "IIAB/XSCE 6.2 Networking" overview

[James Cameron]

On Wed, Feb 22, 2017 at 01:38:57PM -0500, Adam Holt wrote:
> Are all the reserved ports here still in actual/ongoing use?
> 
> Protocol   Port  Service
>   TCP  3128squid / dansguardian

Sugar 0.110 can use this now.  My Settings, Proxy.

>   TCP  3130squid

Sugar 0.110 can use this now.  My Settings, Proxy.

>   TCP  5000xs-authserver

Sugar 0.110 does not use this.  No reference to 5000 found.

>   TCP4369,47893,5280,5222,5223 ejabberd-xs

Sugar 0.110 can use this now, via Telepathy Gabble.

>   TCP  8000sugar-stats-server

Sugar 0.110 does not use this.  A custom branch of Sugar may have used
it in the past, but the changes did not make it into Sugar.

>   TCP  8080idmgr

Sugar 0.110 does use this during registration.
src/jarabe/desktop/schoolserver.py

> Any suggestions for ports we should remove from this list, as
> infrequently used in 2017 and onwards?

Widen your net.  Please ask on sugar-devel@ if anybody needs
8000/sugar-stats-server, and remove it if you get no response.

You might also refine that list of open ports to show which are open
on an external network interface.

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: [Server-devel] Debian LTS Clarification & UbuCon @ SCaLE March 1-2 (free)

[James Cameron]

On Tue, Feb 21, 2017 at 07:20:51AM -0500, Adam Holt wrote:
> 1) in the address bar type: chrome://flags
> 2) Go down to the setting: "Maximum TLS version enabled"
> 3) change it from "default" [TLS 1.3] to "TLS 1.2"

Thanks.  This suggests Google uses different default versions on web
servers depending on geography, or my connection path may contain a
man in the middle version downgrade attack.  The former seems more
likely; and is a problem for any project seeking to localise; because
testing has to be in the deployment country.

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: [Server-devel] Debian LTS Clarification & UbuCon @ SCaLE March 1-2 (free)

[James Cameron]

On Mon, Feb 20, 2017 at 07:05:01PM -0500, Adam Holt wrote:
> On Mon, Feb 20, 2017 at 6:05 PM, James Cameron <[1]qu...@laptop.org> wrote:
> 
> No sign of ERR_SSL_PROTOCOL_ERROR here with same version of Chromium
> package on Raspbian.  Please check your system clock is correct; SSL
> depends on it.  Booting while disconnected, then later connecting, can
> cause clock not to be set.  Common problem with Raspberry Pi because a
> real-time clock was omitted to lower cost.
> 
> This RPi3 has an RTC, and is online with Ethernet these days, and so shows the
> correct time for both reasons (RTC & ntp).
> 
> I rebooted to see if that might would help, but no.
> 
> Changing the Timezone briefly to Europe/London and the back to 
> America/New_York
> (within X -> Raspberry menu in top-left -> Preferences -> Raspberry Pi
> Configuration -> Localisation -> Set Timezone) raised my hopes perhaps 
> falsely,
> as [2]https://google.com suddenly started working for 10 min after that, but
> most all other Google sites remained blocked with ERR_SSL_PROTOCOL_ERROR.
> 
> Cause/effect are unclear, as I cannot reproduce that getting [3]https://
> google.com to appear again, so the above may have been a fluke.

Same version of Chromium is working fine on a Raspberry Pi 3 on my lab
bench, without RTC, with NTP, so there must be something in your
environment causing problem.

There are many possible causes; my guess about clock was just the first one.

Dig deeper into the message; see if you can elicit further detail from
Chromium.

You might also try using "wget" to reproduce the error in a Terminal
window, e.g.

$ wget https://google.com/

This may give more detail as to the reason for the protocol error.

So find a URL that causes the error in Chromium, then copy and paste
the URL for use by wget.

As far as OpenSSL is concerned, both wget and Chromium are callers;
they use OpenSSL.

> I will post to [4]https://www.raspberrypi.org/forums/viewtopic.php?f=63=
> 174870 to see if others hopefully have ideas?

It does not seem likely to be unique to Raspberry Pi 3, so you might
widen your engagement to OpenSSL and Chromium supporting sites.

> You may also use "apt" instead of "apt-get" like this;
> 
> apt update
> apt full-upgrade
> 
> You may also find "apt-get clean" to be more comprehensive than
> "apt-get autoclean".
> 
> There's also a package for automated unattended upgrades, see
> "unattended-upgrades".
> 
> 3 excellent tips above.  Thank you James.

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: [Server-devel] Debian LTS Clarification & UbuCon @ SCaLE March 1-2 (free)

[James Cameron]

No sign of ERR_SSL_PROTOCOL_ERROR here with same version of Chromium
package on Raspbian.  Please check your system clock is correct; SSL
depends on it.  Booting while disconnected, then later connecting, can
cause clock not to be set.  Common problem with Raspberry Pi because a
real-time clock was omitted to lower cost.

You may also use "apt" instead of "apt-get" like this;

apt update
apt full-upgrade

You may also find "apt-get clean" to be more comprehensive than
"apt-get autoclean".

There's also a package for automated unattended upgrades, see
"unattended-upgrades".

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: [Server-devel] Debian LTS Clarification & UbuCon @ SCaLE March 1-2 (free)

[James Cameron]

On Sat, Feb 18, 2017 at 03:58:26PM -0500, Adam Holt wrote:
> Thanks to Alex Perez:
> 
> "The Debian project is extending its famous development process to offer
> long-term support."
> http://www.linux-magazine.com/Issues/2017/194/Debian-LTS
> 
> I'm having trouble understanding if this is really different from
> the promises made in very recent years,

Which promises are they?  My guess is that you don't actually need to
compare promises in detail.

> but hopefully experts can clarify how the rubber's increasingly
> meeting the road, delivering on these proactive security promises :)

Probably you mean "how to use LTS?"

1.  recognise that long-term support is mostly security updates which
close vulnerabilities that are detected after release,

2.  set up and test the automated installation of the security updates
(using http://security.debian.org/ in sources.list); if network
bandwidth for updates is a problem, avoid deploying certain large
packages, set large packages to hold using dpkg, switch to using
ostree, delta packages, or use other tricks,

3.  make a list of packages for which you need long-term support; such
as the list of packages in your product, (sudo dpkg-query -W),

4.  subtract from the list any that the Debian release notes
specifically exclude; such as WebKit or VirtualBox; and figure out how
to self-support those packages,

5.  the remaining packages are long-term supported; so ensure the
debian-security-support package is installed and is configured to
inform the user somehow.

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: [Server-devel] Apache 2.4.10 from July 2014 is part of XSCE 6.2: is this safe?

[James Cameron]

Security updates to the Jessie apache2 package have been happening, as
patches are backported, and they are in Rasbian, see
https://anonscm.debian.org/cgit/pkg-apache/apache2.git/log/?id=refs/heads/jessie
for the list of changes, many of which have CVEs.

But yeah, Nginx vs Apache2, doesn't matter for your use case; choose
based on preference and skills of your technical users.  Flexibility
traded for performance.

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: Best GCompris for OLPC OS 13.2.8?

[James Cameron]

What was wrong with my offer?

You're not doing anything wrong; it's just the way it is.

The way yum works; both the downloaded .rpm files and the unpacked
files have to exist at the same time briefly.  yum calculates this
and tells you it won't work.

Removing the swap file will save you no more than 64MB, and will make
the laptop slower afterwards.

If you are insisting, for some reason, on ignoring my recommendations;
let me give you one more to ignore; mount a USB drive over /var so
that yum uses it for the downloaded .rpm files.  That way, the
downloads won't count against the root filesystem space.

On Mon, Feb 13, 2017 at 12:40:31AM -0500, Adam Holt wrote:
> Thanks James & Nathan.  This "model" XO-1 has 256MB of disk available on fresh
> install of 13.2.8, which never falls below 240MB when trying many different
> ways to install the "7+ MB" TuxMath.
> 
> So I'm surely doing something wrong, and will keep working on it this week,
> until I rediscover something like the reliable TuxMath install recipe we had
> for Haiti earlier.
> 
> Then, trying to install the "142+ MB" GCompris will be a different story of
> course, given its OS library needs installing Tux Paint etc alongside, which
> may require me deleting the 70+ MB /home/olpc/Library (and other /tmp /var/tmp
> or swap adjustments?  Nathan also suggests "swapoff -a"  and "rm /var/swap").
> 
> I'm traveling today but will rededicate myself to cracking these 2 painful/
> important obstacles Tuesday onwards!
> 
> On Feb 13, 2017 5:02 AM, "James Cameron" <[1]qu...@laptop.org> wrote:
> 
> Yes, the XO-1 JFFS2 NAND filesystem may report less space after the
> laptop has been used for a bit.  Fresh install gets the best space.
> 
> (Because JFFS2 compresses data when writing, it cannot know how much
> free space is available, as because it depends on how compressible the
> data is.  So JFFS2 provides an estimate, and the estimate is
> pessimistic, and yum takes it at face value.)
> 
> Yes, one XO-1 may show different space to another XO-1 freshly
> installed.
> 
> (Because NAND may have bad blocks that are skipped by JFFS during
> reflashing.  You will have noticed these as differently coloured.
> Good chance of more bad blocks as the laptops age.)
> 
> Yes, removing files apart from Activities and Library won't fix space.
> 
> (Because olpc-update keeps a hard link mirror of the filesystem in
> /versions, and your changes to removing files won't remove them from
> the copy.)
> 
> Yes, customising the operating system is really time consuming; we
> have instead provided the tools for remastering the operating system,
> and these tools are easy to use.  Adding .xo and doing yum installs
> are supported.  See OS_Builder on the Wiki.
> 
> If you lack resources to do this, then I can assist to a limited
> extent, as I did for Haiti in January last year.  Additional benefit
> of using me is that the build may be signed, which installs using the
> four game key method without having to create deployment keys.
> 
> Workaround for yum lack of space is to use .rpm files directly.  Also
> much faster.
> 
> --
> James Cameron
> [2]http://quozl.netrek.org/
> 
> References:
> 
> [1] mailto:qu...@laptop.org
> [2] http://quozl.netrek.org/

> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel


-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Best GCompris for OLPC OS 13.2.8?

[James Cameron]

Yes, the XO-1 JFFS2 NAND filesystem may report less space after the
laptop has been used for a bit.  Fresh install gets the best space.

(Because JFFS2 compresses data when writing, it cannot know how much
free space is available, as because it depends on how compressible the
data is.  So JFFS2 provides an estimate, and the estimate is
pessimistic, and yum takes it at face value.)

Yes, one XO-1 may show different space to another XO-1 freshly
installed.

(Because NAND may have bad blocks that are skipped by JFFS during
reflashing.  You will have noticed these as differently coloured.
Good chance of more bad blocks as the laptops age.)

Yes, removing files apart from Activities and Library won't fix space.

(Because olpc-update keeps a hard link mirror of the filesystem in
/versions, and your changes to removing files won't remove them from
the copy.)

Yes, customising the operating system is really time consuming; we
have instead provided the tools for remastering the operating system,
and these tools are easy to use.  Adding .xo and doing yum installs
are supported.  See OS_Builder on the Wiki.

If you lack resources to do this, then I can assist to a limited
extent, as I did for Haiti in January last year.  Additional benefit
of using me is that the build may be signed, which installs using the
four game key method without having to create deployment keys.

Workaround for yum lack of space is to use .rpm files directly.  Also
much faster.

-- 
James Cameron
http://quozl.netrek.org/


signature.asc
Description: Digital signature
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: [Server-devel] overhauling our FAQ for Internet-in-a-Box/XSCE 6.2 !

[James Cameron]

Adam wrote:
> And older x86 machines like XO-1 and XO-1.5 may run i386 !? (316MB ISO)
> https://www.debian.org/distrib/

No.

XO-1 and XO-1.5 are not a PC.  XO-1 and XO-1.5 require an Open
Firmware bootloader configuration and custom kernel.  The ISO 9660
based Debian network installer image you reference does not have these
features.

Shortest path mapped already is debootstrap on another system to
create a custom filesystem on an SD card or USB drive that can then be
inserted into the XO.

http://wiki.laptop.org/go/Debian_Jessie_on_XO

Please also ensure hardware redundancy; the XO-1 and XO-1.5 units are
beyond end of expected life and will be failing rapidly, statistically
speaking.  Can your server software work with multiple server
instances?  If not, stick to one known good server.

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: [Server-devel] How can I delete 10 years of git history in the XSCE repo?

[James Cameron]

Adam Holt wrote:
> James Cameron wrote:
> > George Hunt wrote:
> > > The XSCE repo, inherited from laptop.org has grown to 80MB+.
> >
> >Yes, but what problem are you trying to solve?
> 
> Preliminary Repo Re-org document here [...]
> 
> https://docs.google.com/document/d/1wY2AwSfr0w712FhL9VY51Tpf9E39cBnOclVPU7_yzeg/edit

Thanks.  Reviewed.  Doesn't answer my question.  Reorganisation there
has nothing significant to do with the size of the repository.

With that degree of reorganisation it may be better to start with
fresh repositories.  History of change can still be found in the old
repository.

I've not been watching the difficulties of your multiple committers,
but my assessment, based on limited information (server-devel@ and the
xsce/xsce.git commit log merge commits), is that your situation can be
improved another way; ensure each contributor is able and willing to
squash, interactive rebase, and resolve any kind of git-conflicts,
using the best tools for the job.

At Sugar Labs the Google Code-In just finished had quite a few
contributors who had difficulty with these steps.  Contributors vary
considerably in their willingness to engage at that level.  Your
reorganisation looks like it will avoid conflict by avoiding
collaboration.  On the other hand, if that is the group's decision,
then go for it.

Nadia Eghbal's keynote talk yesterday at linux.conf.au is really
relevant.  Nadia is in Community Programs at GitHub.  GitHub is the
closed-source system you're using for contributor collaboration.

https://linux.conf.au/schedule/presentation/106/
https://www.youtube.com/watch?v=W2AR1owg0ao

Hope that helps!

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: [Server-devel] How can I delete 10 years of git history in the XSCE repo?

[James Cameron]

On Wed, Jan 18, 2017 at 08:13:38PM -0800, George Hunt wrote:
> The XSCE repo, inherited from laptop.org has grown to 80MB+.

Yes, but what problem are you trying to solve?

A --depth 1 clone of xsce/xsce.git master is 1.64 MiB, and a --depth 1
clone of georgehunt/xsce.git base-6.2 is 941.93 KiB.  So the 80 MiB of
a full clone should have no impact unless the full history is needed;
almost nobody needs that.

Make sure your guidance for new developers mentions --depth properly,
and never script anything that clones without --depth.

> What's wanted is a repo that includes all of the commits starting from
> https:/github.com/xsce/xsce --branch release-6.0 and including all of
> the commits on https://github.com/georgejhunt/xsce --branch
> base-6.2.

Yes, easily said, not so easily done.

I've reproduced your problem, but with the release-6.0 branch alone.

That branch can't be easily reduced, because of merge conflicts that
were pushed before 6.0.  Rebasing across merge conflicts is fraught.

This blocks you from using the simple technique.

You have to either resolve each conflict, or get into rewriting
history.  Choose from the available mergetools.

It may be easier to start a fresh repository based on release-6.0,
then add a single commit to bring it to base-6.2.  Or add your ~637
commits as patches using "git format-patch" and "git am".

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: Firefox on XO

[James Cameron]

Firefox-26 from Fedora 18 using yum was built by the Fedora project,
but Firefox-50 downloaded from Mozilla was built by the Mozilla
project.

Firefox-26 from Fedora 18 has different build configuration settings
to Firefox-50 from Fedora latest.  You might compare those settings by
looking at the RPM spec file.

You can find the Fedora packaging files for Firefox here;

https://src.fedoraproject.org/cgit/rpms/firefox.git/tree/ (the current
master branch),

https://src.fedoraproject.org/cgit/rpms/firefox.git/tree/?h=f18 (the
last Fedora 18 build),

Firefox might be rebuilt with these configuration settings changed;
see rpmbuild.

But I don't think you'll be likely to find a recent build of Firefox
suited for Fedora 18.  Everyone else will have upgraded by now.

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Firefox on XO

[James Cameron]

G'day Terry,

You didn't mention which version of Firefox.

With firefox-26.0-2 from the Fedora 18 repository your no-sound issue
with those sites reproduces.

Then I installed apulse, a PulseAudio emulation for ALSA, but the
problem persisted.  Looking at /proc/$(pidof firefox)/smaps the
libpulse libraries are not mapped.

Which Firefox are you trying, and how are you trying it?

References:

https://github.com/i-rinat/apulse (source)
https://github.com/vitvegl/apulse-rpm (for the spec file)

dependencies: gcc-c++ gcc-sh-linux-gnu cmake alsa-lib-devel glib2-devel 
alsa-lib glib2

On Sun, Jan 01, 2017 at 08:52:39AM +1000, T Gillett wrote:
> Hi All
> 
> I have installed Firefox on the current 13.2.8 build on XO-1.5 and it works OK
> except that there is no sound on some websites.
> Other sites work OK.
> 
> This seems to be a known problem:
>     [1]https://bugzilla.mozilla.org/show_bug.cgi?id=803042
>     [2]http://forums.fedoraforum.org/showthread.php?t=277040
>     [3]https://ask.fedoraproject.org/en/question/65978/fedora-21-fi
> refox-3603-and-flash-no-sound-system-sound-fine/
>     [4]http://unix.stackexchange.com/questions/144395/
> firefox-html5-asoundrc-no-sound
> 
> The issue seems to revolve around getting FF to use ALSA for sound, and 
> without
> installing PulseAudio.
>  
> I am wondering if anyone using Firefox on XO has encountered and/or solved 
> this
> problem.
> 
> Thanks in advance for any help on this.
> Terry
> 
> 
> 
> Some sites for testing the audio. Open any of these in a new tab and you 
> should
> hear audio.
> 
> 1.    [5]http://student.readingeggs.com//sample_lessons/#/sample/lesson/6/
> activity/1
> No sound on XO.
> 
> 2.    [6]http://vt-motoli.org/CG/cg_03/cg_03.html#l-en
> No sound on XO.
> 
> 3.    [7]http://robbbenson.com/01%20Eve.mp3
> Sound OK.
> 
> References:
> 
> [1] https://bugzilla.mozilla.org/show_bug.cgi?id=803042
> [2] http://forums.fedoraforum.org/showthread.php?t=277040
> [3] 
> https://ask.fedoraproject.org/en/question/65978/fedora-21-firefox-3603-and-flash-no-sound-system-sound-fine/
> [4] 
> http://unix.stackexchange.com/questions/144395/firefox-html5-asoundrc-no-sound
> [5] 
> http://student.readingeggs.com//sample_lessons/#/sample/lesson/6/activity/2
> [6] http://vt-motoli.org/CG/cg_03/cg_03.html#l-en
> [7] http://robbbenson.com/01%20Eve.mp3

> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel


-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Announcing OLPC OS 13.2.8 with Sugar 0.110

[James Cameron]

On Wed, Dec 14, 2016 at 01:20:54PM -0800, Alex Perez wrote:
> Thanks for the release. It is much appreciated. So far, I’ve flashed
> it just fine on several production XO-1’s I have, both on the
> internal NAND as well as to a 32GB Samsung (Class 6) SD card,
> however in the case of the SD card install, once the flash
> completes, and I’m booted into the 13.2.8 environment , I
> consistently get MMC timeout errors, and the kernel hangs. This is
> easily reproducible by simply running ‘yum update’ from the CLI.
> 
> The exact message emitted by the kernel is “mmc0: Timeout waiting
> for hardware interrupt. mmcblk0: error -110 sending stop command,
> original cmd response 0x900, card status 0xe00.

This is a known symptom on some XO-1 units.  It also varies by SD
card, and is very likely if automatic power management is enabled.  My
build does not enable automatic power management; can you confirm you
did not?

> I can re-boot into the on-board instance of the OS on NAND, and run
> fsck against the filesystem on the SD card, and no filesystem-level
> errors are detected there.
> 
> Do you have any suggestions for follow-up troubleshooting?

It is difficult because SD card is tiny computer with closed-source.

Try shutdown, remove battery, wait ten seconds, restore battery, power
on.  Reason: ensure no CMOS lockup due to ESD.

Try different cards.  Reason: cards vary.

Try different XO-1.  Reason: SD slot circuit varies.

Try read the card fully; e.g. "sudo dd if=/dev/mmcblk0 of=/dev/null".
Reason: card firmware will rewrite any flash blocks that are close to
read disturb threshold count.  As a result timing of future read will
be changed.

Try fs-update on same card in an XO-1.5, XO-1.75, or XO-4, and leave
laptop at ok prompt for a couple of minutes, then move the card to the
XO-1.  Reason: card firmware may not be capable of recovery from
sudden power loss after write; the XO-1 fs-update pauses for only ten
seconds before power donw.

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: [Server-devel] [XSCE] xsce on centos/rpi3

[James Cameron]

Yes, the Rpi3, clones, and similar devices are encouraging.

We have everything needed for Sugar 0.110 on the Rpi3 within Raspbian
as a desktop option or as applications within the desktop.

I've done tests in my lab as part of testing Sugar for the OLPC
laptops.  My collection of Rpi3 have helped with my recent work on
OLPC OS 16.04.2 on Ubuntu 16.04 and OLPC OS 13.2.8 on Fedora 18.

But my focus is still on OLPC laptops, and so I'm not really in a
position to fully support Rpi3 yet.  The OLPC laptops have the best
Sugar experience, in my biased opinion.

Developers who wish to be involved on a Sugar on Raspbian push should
be talking on sugar-devel@ and taking the lead from Jonas Smedegaard
and Sebastian Silva.  Their work is packaging and integrating, using
resources such as the Debian git repositories, package tracker, bug
tracker, package archive, and mailing lists.  Tip of the iceberg
occasionally shows on sugar-devel@ or Sugar git repositories;
especially noisy last couple of days.

On Tue, Dec 13, 2016 at 04:20:11AM -0500, Adam Holt wrote:
> On Tue, Dec 13, 2016 at 3:07 AM, Adam Holt <[1]h...@laptop.org> wrote:
> 
> On Mon, Dec 12, 2016 at 9:04 PM, Anish Mangal <[2]anis...@umich.edu> 
> wrote:
>
> Hi,
> 
> Wondering if anyone has tried installing XSCE on centos on rpi3, and
> faced issues. Planning to try this out now that I have rpi3's
> 
> Not yet.  Most of us are using Raspbian (see Nov 30's [3]https://
> www.raspberrypi.org/blog/a-security-update-for-raspbian-pixel/)
> 
> Just FYI a new development is that a growing number of kickass deployments are
> asking me for the full Raspian ("Pixel" takes up 3GB more disk "Lite" version,
> likewise its zip/download is 1.1GB larger) as it provides an increasingly
> well-rounded learning suite for remote teachers/techs/kids to learn Linux &
> much more:
> 
>    [4]http://downloads.raspberrypi.org/raspbian/release_notes.txt
> 
> Aside: currently lacking Sugar regrettably, and not quite as well packaged for
> educators as those other Brits @ [5]https://twitter.com/teamkano.  Naturally
> there are other deployments that do not want these 3GB of learning/IT tools, 
> as
> things boot faster without, and traditionalists do not want visual tools (X
> Windows).
> 
> In essence it took half a decade, but the $35 computer is now unleashing a lot
> of field innovation across these 2 categories primarily:
> 1) "one RPi3 per child" deployments who require modern visual tools on each
> RPi3 (and can afford a cheap LCD per child/seat)
> 2) "one RPi3 server per classroom" deployments (who may or may not be able to
> afford an LCD or projector for the teacher)
> 
> Of course there are many of both types of learning communities above and
> others, who do not always see eye to eye (imagine that, another culture war,
> what a surprise ;)  So our Mission is to encourage maximum cross-fertilization
> across these most practical scenarios--wherever we can, despite
> not-invented-here bubbles of innovation--embracing creative 
> cross-collaboration
> / proven best practices as a larger force than infantile infighting, across an
> entire planet of fantastically passionate grassroots implementers~
> 
> In many cases Sugarizer-on-XSCE/IIAB can provide a partial answer.  But Tony
> Anderson is correct to remind us Sugarizer does not yet provide nearly as rich
> an experience as full Sugar.  Meantime all educators cry out for platform
> stability that constructionism ironically relies on, so [6]http://
> wiki.laptop.org/go/Release_notes/13.2.8 on XO laptops is a big step forward
> with Sugar 0.110, thanks to James Cameron!
> 
> or FedBerry ([7]http://fedberry.org) as HW/networking support (Bluetooth,
> WiFi, etc) is far better.
> 
> In future CentOS may catch up; these 3 people are certainly trying:
> [8]https://lists.centos.org/pipermail/centos/2016-December/162276.html
>  
> 
> Best,
> Anish
> 
> --
> Unsung Heroes of OLPC, interviewed live @ [9]http://unleashkids.org !
> 
> --
> Unsung Heroes of OLPC, interviewed live @ [10]http://unleashkids.org !
> 
> References:
> 
> [1] mailto:h...@laptop.org
> [2] mailto:anis...@umich.edu
> [3] https://www.raspberrypi.org/blog/a-security-update-for-raspbian-pixel/
> [4] http://downloads.raspberrypi.org/raspbian/release_notes.txt
> [5] https://twitter.com/teamkano
> [6] http://wiki.laptop.org/go/Release_notes/13.2.8
> [7] http://fedberry.org/
> [8] https://lists.centos.org/pipermail/centos/2016-December/162276.html
> [9] http://unleashkids.org/
> [10] http://unleashkids.org/

> ___
> Server-devel mailing list
> Server-devel@lists.laptop.org
> http://lists.laptop.org/listinfo/server-devel


-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Announcing OLPC OS 13.2.8 with Sugar 0.110

[James Cameron]

G'day,

We're pleased to announce the release of OLPC OS 13.2.8 for XO-1,
XO-1.5, XO-1.75 and XO-4.

It is Sugar 0.110 on Fedora 18, with updated activities Speak-52,
Measure-53, Maze-26.1, Implode-17, GetBooks-16.2, Clock-18.1, and
Chat-83.

Details of new features, fixes, known issues, and how to download,
install or upgrade can be found in the release notes:

http://wiki.laptop.org/go/Release_notes/13.2.8

Our thanks to all contributors, testers, upstreams, and those who
have provided feedback of any kind.

Build file names: 32020o0 32020o1 32020o2 32020o4

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

[Server-devel] BBR: Congestion-Based Congestion Control

[James Cameron]

School server networks fail to perform fairly or effectively because
the loss-based congestion control methods in TCP networking are not a
good match for the wireless environment.

Wireless access points or devices with too much buffering memory make
the problem worse.  A case of more not being good.

Former OLPCer Jim Gettys warned us about bufferbloat years ago.

I'd like to hear from anyone deploying the new BBR congestion-base
controls, which should be available in 4.9 kernels.  It is a
server-side feature, so there's no great need for it in laptops that
operate as clients.

References:

http://queue.acm.org/detail.cfm?id=3022184
https://lwn.net/Articles/701165/
http://queue.acm.org/detail.cfm?id=2071893

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: OSBuilder for 13.2.7

[James Cameron]

It is not possible to use the G1G1 group as is.  Many activities would
be downgraded and some won't work properly in 13.2.7.  You found one.

Much ambiguity on the Wiki page, so I've made many edits:

http://wiki.laptop.org/go/OS_Builder/Add_additional_Activities

Please re-read the page and choose a method.

Briefly, cause of the error is in content collection Biology; version
10 released around 2010 contains a library.info file with protection
from read.  As there was no substantial change from version 9 to
version 10, you can use version 9 instead.

Because you added the letters G1G1, our Wiki page
http://wiki.laptop.org/go/Activities/G1G1/13.2 was used for versions,
and this was last updated by Daniel Drake in May 2013.  You might have
got the idea to use G1G1 from the Wiki page, hence the edits.

You can also find documentation on sugar_activity_group in the
modules/sugar_activity_group/README file of the builder repository.

Hope that helps.

On Sun, Oct 02, 2016 at 09:31:17PM +0200, Lionel Laské wrote:
> 
> Hi James,
> 
> Just a follow up on this thread because I've got an issue customizing my 
> build.
> I've changed the line:
> 
> [sugar_activity_group]
> 
> url=[1]http://wiki.laptop.org/go/Activities
> 
> to:
> 
> [sugar_activity_group]
> 
> url=[2]http://wiki.laptop.org/go/Activities/G1G1
> 
> My objective is to include more activities.
> I've got an issue at the end of the build (see message below).
> It seams related to content (Biology here) included in Browse activity.
> 
> Is it still possible to use G1G1 group or should I mention and download myself
> locally all activities ?
> 
> Best regards.
> 
>          Lionel.
> 
> caution: excluded filename not matched:  mimetype
> 
> caution: excluded filename not matched:  mimetype
> 
> caution: excluded filename not matched:  mimetype
> 
> Traceback (most recent call last):
> 
>   File "/bin/olpc-library-update", line 175, in 
> 
>     cp.readfp(codecs.open(info_file_path, 'r', 'utf-8'))
> 
>   File "/usr/lib/python2.7/codecs.py", line 881, in open
> 
>     file = __builtin__.open(filename, mode, buffering)
> 
> IOError: [Errno 13] Permission denied: 
> u'/home/olpc/Library/Biology/library/[3]
> library.info'
> 
> ERROR:root:Error creating Live CD : %post script failed with code 1
> 
> INFO:root:Unmounting directory /oob/olpc-os-builder/build/imgcreate-u5YUwK/
> install_root
> 
> WARNING:root:Unmounting directory /oob/olpc-os-builder/build/imgcreate-u5YUwK/
> install_root failed, using lazy umount
> 
> Unmounting directory /oob/olpc-os-builder/build/imgcreate-u5YUwK/install_root
> failed, using lazy umount
> 
> INFO:root:lazy umount succeeded on 
> /oob/olpc-os-builder/build/imgcreate-u5YUwK/
> install_root
> 
> lazy umount succeeded on /oob/olpc-os-builder/build/imgcreate-u5YUwK/
> install_root
> 
> INFO:root:Losetup remove /dev/loop0
> 
>  * Caught error, cleanup and then bail out.
> 
>  * Running part cleanup base cleanup.50.cleanup.sh...
> 
> ERROR: Failure in BuildStage: module base, part [4]build.40.imagecreate.py,
> error code 1
> 
> bash-4.2#
> 
> 2016-09-23 2:33 GMT+02:00 James Cameron <[5]qu...@laptop.org>:
> 
> On Thu, Sep 22, 2016 at 10:12:54PM +0200, Lionel Laské wrote:
> > Next issue seems to be related to a 404 error loading a Fedora package.
> > Following is the detail.
> > [...]
> > yum.Errors.NoMoreMirrorsRepoError: failure: repodata/repomd.xml from
> fedora:
> > [Errno 256] No more mirrors to try.
> > [56][6]https://fr2.rpmfind.net/linux/fedora-secondary/releases/18/
> Everything/
> > armhfp/os/repodata/repomd.xml: [Errno 14] curl#22 - "The requested URL
> returned
> > error: 404 Not Found"
> 
> Thanks, the problem did reproduce on our build system, and is fixed in
> the v7.0 branch of the olpc-os-builder repository.  Please pull my
> patch "configs: fedora repository mirrors".
> 
> [7]http://dev.laptop.org/git/projects/olpc-os-builder/commit/?h=v7.0=
> 7e2b9a90ea0182f045e8c983e75200f0c4cb6c21
> 
> Then repeat your build.
> 
> My build ran to completion.  It had to be restarted a few times to
> deal with temporary failures of [8]download.sugarlabs.org, but that's
> outside my control.
> 
> Compared to 13.2.7, the build has Sugar 0.109.0.2, with activity
> updates Chat-82, Clock-18.1, Measure-53, and Speak-52.  Also a new
> kernel and olpc-powerd package for camera support.
> 
> Result is the same as if 13.2.7 is installed followed by "yum upgrade"
> and Sugar Software Update.
> 
>

Re: [Server-devel] Rpi3 wifi boost hack

[James Cameron]

Thanks, that's interesting.

The disclaimers are correct.

I've inspected an rpi3 just now, and as the author observed the U.FL
pads are not fully clear of mask; my guess is they were set up for
testing only, not production.

I'm licensed to make this sort of change, but I'd only do it if I'd
exhausted other link budget possibilities.  I'm not offering.

If anyone does this, use a spectrum analyser to verify success, and
especially avoid generating out-of-band interference.  And use more
flux, and a board preheat.  ;-)

On Tue, Sep 27, 2016 at 09:13:43AM +0530, Anish Mangal wrote:
> Hi,
> 
> I came across this. Apparently, it may be possible to connect external 
> antennas
> to the rpi3. This will allow the pi3 wifi to have much better sensitivity (and
> thus provide wider & customizable wifi coverage) for a little extra expense.
> 
> [1]http://www.dorkbotpdx.org/blog/wramsdell/external_
> antenna_modifications_for_the_raspberry_pi_3
> 
> Cheers,
> Anish
> 
> References:
> 
> [1] 
> http://www.dorkbotpdx.org/blog/wramsdell/external_antenna_modifications_for_the_raspberry_pi_3

> ___
> Server-devel mailing list
> Server-devel@lists.laptop.org
> http://lists.laptop.org/listinfo/server-devel


-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: OSBuilder for 13.2.7

[James Cameron]

On Thu, Sep 22, 2016 at 10:12:54PM +0200, Lionel Laské wrote:
> Next issue seems to be related to a 404 error loading a Fedora package.
> Following is the detail.
> [...]
> yum.Errors.NoMoreMirrorsRepoError: failure: repodata/repomd.xml from fedora:
> [Errno 256] No more mirrors to try.
> [56]https://fr2.rpmfind.net/linux/fedora-secondary/releases/18/Everything/
> armhfp/os/repodata/repomd.xml: [Errno 14] curl#22 - "The requested URL 
> returned
> error: 404 Not Found"

Thanks, the problem did reproduce on our build system, and is fixed in
the v7.0 branch of the olpc-os-builder repository.  Please pull my
patch "configs: fedora repository mirrors".

http://dev.laptop.org/git/projects/olpc-os-builder/commit/?h=v7.0=7e2b9a90ea0182f045e8c983e75200f0c4cb6c21

Then repeat your build.

My build ran to completion.  It had to be restarted a few times to
deal with temporary failures of download.sugarlabs.org, but that's
outside my control.

Compared to 13.2.7, the build has Sugar 0.109.0.2, with activity
updates Chat-82, Clock-18.1, Measure-53, and Speak-52.  Also a new
kernel and olpc-powerd package for camera support.

Result is the same as if 13.2.7 is installed followed by "yum upgrade"
and Sugar Software Update.

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: OSBuilder for 13.2.7

[James Cameron]

On Sat, Sep 17, 2016 at 09:20:05PM +0200, Lionel Laské wrote:
> 
> Thanks for your answer.
> I've launched 'make' because it's mentioned here [1].

Thanks for the explanation, but please reload the page if it still
says just "make".  It was edited after your mail.

> But the zhashfs is present in the bin directory.

Good.

> BTW it don't works when I'm tried. The build fail on KspostStage
> (see full log below).  Any idea ?
> [...]
> Examining org.sugarlabs.MusicKeyboard v8.2: http://download.sugarlabs.org/
> activities/4564/music_keyboard-8.2.xo
> 
> HTTP error: 404

During an outage of activities.sugarlabs.org in July, builds failed
because the activity download links on wiki.laptop.org were invalid.
The links to activity bundles on download.sugarlabs.org continued to
function.  So I changed our page to use the better URLs.

I made a typo.  Digit transposition.  It wasn't noticed because my
build system used cached files, and other people doing builds use
their own activity lists.

You can see the edit history of the activity list page here:

http://wiki.laptop.org/index.php?title=Activities/13.2.7=history

Now that I've fixed the typo, try the build again, so we can find out
what next will fail.

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: OSBuilder for 13.2.7

[James Cameron]

Okay, I see the problem.

Instead of "make", please type "(cd bin && make)" then proceed with the next 
steps on the page.

For your interest, in the bin directory you'll find the zhashfs program source, 
which has to be built before starting the builder.  Only has to happen once 
after cloning.  All the rest of olpc-os-builder is in Python and doesn't need 
compilation.

On Sat, Sep 17, 2016 at 11:17:21AM +0200, Lionel Laské wrote:
> Hi all,
> 
> I would like to generate OSBuilder for XO-4 to create my custom build on a 
> XO-4
> using OLPC 13.2.7.
> I've installed OLPC 13.2.7 on the XO-4 following instructions here [1].
> I've installed os-build tool following instructions here [2].
> But when I'm launching "make", I've got a message telling:
> 
>    make[1]: Nothing to be done for 'all'.
> 
> Could you confirm that the procedure described here [2] is compatible with
> 13.2.7 ?
> If no, is there anyone that could help me to update the procedure ?
> If yes, what this message means ?
> 
> Thanks in advance for your help.
> 
> Best regards from France.
> 
>          Lionel.
> 
> [1] [1]http://wiki.laptop.org/go/Release_notes/13.2.7#XO-4
> [2] [2]http://wiki.laptop.org/go/OS_Builder/Run_on_XO_hardware
> 
> References:
> 
> [1] http://wiki.laptop.org/go/Release_notes/13.2.7#XO-4
> [2] http://wiki.laptop.org/go/OS_Builder/Run_on_XO_hardware

> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel


-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: OSBuilder for 13.2.7

[James Cameron]

I've never typed make.  I run the builder directly.

On Sat, Sep 17, 2016 at 11:17:21AM +0200, Lionel Laské wrote:
> Hi all,
> 
> I would like to generate OSBuilder for XO-4 to create my custom build on a 
> XO-4
> using OLPC 13.2.7.
> I've installed OLPC 13.2.7 on the XO-4 following instructions here [1].
> I've installed os-build tool following instructions here [2].
> But when I'm launching "make", I've got a message telling:
> 
>    make[1]: Nothing to be done for 'all'.
> 
> Could you confirm that the procedure described here [2] is compatible with
> 13.2.7 ?
> If no, is there anyone that could help me to update the procedure ?
> If yes, what this message means ?
> 
> Thanks in advance for your help.
> 
> Best regards from France.
> 
>          Lionel.
> 
> [1] [1]http://wiki.laptop.org/go/Release_notes/13.2.7#XO-4
> [2] [2]http://wiki.laptop.org/go/OS_Builder/Run_on_XO_hardware
> 
> References:
> 
> [1] http://wiki.laptop.org/go/Release_notes/13.2.7#XO-4
> [2] http://wiki.laptop.org/go/OS_Builder/Run_on_XO_hardware

> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel


-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: [Server-devel] (probably) the world's highest solar powered schoolserver and mesh setup

[James Cameron]

Good reading, thanks.  Looking forward to more.

The clear bottle packaging of the routers in the photographs is
interesting.  Looking at the climate data for Leh;

https://en.wikipedia.org/wiki/Leh#Climate

my guess is that the bottle will make a good thermal environment;
spending the most time within the temperature range of the router.
Router radios become less sensitive as temperature increases.

Charting inside and outside temperature would be interesting.

At high altitudes my caution would suggest a watchdog circuit for the
DC to DC converter, to switch it off and then on again if the router
isn't responding.  Especially where batteries are used.  Where no
battery is used, a node will restart next morning, and that may be
enough to handle the more frequent single event upsets.

http://www.dfrsolutions.com/pdfs/2004_HighAltitude_Hillman-Blattau.pdf

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: Removing Adobe Flash on XO-1, XO-1.75 and XO-4 laptops (CVE-2015-7645)

[James Cameron]

An update from Adobe;

https://blogs.adobe.com/flashplayer/2016/08/beta-news-flash-player-npapi-for-linux.html#sthash.zmBvj3cS.R31QTmNV.dpbs

thanks to

http://www.theregister.co.uk/2016/09/06/adobe_reverses_decision_to_kill_npapi_flash_plugin_for_linux/

This will likely work on XO-1.5.  On XO-1, XO-1.75 and XO-4 the
situation below is unchanged.

On Mon, Oct 19, 2015 at 01:36:32PM +1100, James Cameron wrote:
> G'day,
> 
> To keep learners safe, please remove the Adobe Flash plugin on XO-1,
> XO-1.75 or XO-4 laptops:
> 
>   sudo rm /usr/lib/mozilla/plugins/libflashplayer.so
> 
> Adobe Security Bulletin APSA15-05 [1] is for a critical remote control
> vulnerability (CVE-2015-7645).  An attacker can take control of the
> laptop.
> 
> For XO-4 and XO-1.75, there will be no fix.  Adobe do not provide a
> player for Linux 32-bit ARM platform [2].
> 
> For XO-1, there will be no fix.  Adobe do not provide a player
> compatible with the AMD Geode processor on the XO-1, which doesn't
> have SSE (XMM) registers [3].
> 
> For XO-1.5, a fix may be available from Adobe.  Use the steps
> on our Wiki [4].
> 
> --
> 
> References:
> 
> 1.
> https://helpx.adobe.com/security/products/flash-player/apsa15-05.html
> 
> 2.
> https://get.adobe.com/flashplayer/otherversions/
> 
> 3.
> http://lists.laptop.org/pipermail/devel/2012-October/036209.html
> 
> 4.
> http://wiki.laptop.org/go/Adobe_Flash#Installation_on_XO-1.5
> 
> -- 
> James Cameron
> http://quozl.linux.org.au/

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: [Server-devel] "Gigabyte BRIX more scalable than Intel NUC"

[James Cameron]

On Mon, Sep 05, 2016 at 10:25:09AM -0400, Adam Holt wrote:
> On Terry's recommendation to look at replaceable WiFi modules that
> can serve "almost 50 kids", my current interest is to explore the
> $279.99 Gigabyte BRIX GB-BSi3H-6100:
> 
> http://www.gigabyte.com/products/product-page.aspx?pid=5691

Looks good.

> If we go with the above CPU/chassis -- or anything similar folks
> recommend -- which WiFi module (PCIe M.2 presumably) to consider for
> maximum community support isn't an easy question of course: any
> recommendations for Linux support?

At OLPC in the past months I've been testing two M.2 cards and two
MiniPCI Express cards.  Not the special ones for XO-4, but industry
standard ones.  And not for access point mode, just as client.
Realtek is the chipset vendor.  Don't know if the cards are available
retail, as we're focused on the volume side of things.  M.2 is
certainly a more forward looking connector design to choose from;
higher density, more bus features.

I'm worried about your approach.  Don't fixate on product selection?
While it might seem to be the only thing you can control, your problem
is affected by a whole stack of components; wireless module,
motherboard host controller, kernel driver for host controller, kernel
driver for wireless module, kernel driver for access point mode,
programs for access point mode, and more.  I'd draw you a picture but
my crayon is hiding.

> Bluetooth 4.2 Low Energy support would be an Optional Bonus, as we
> have a long-term desire to provide teacher smartphones' full control
> over their "personal" school server.

Bad idea.  Bluetooth LE will use the same radiofrequency spectrum as
the wireless access point.  Co-existence means switching between the
two modes, and that switching will deafen one or the other mode.
Switching also takes time, albeit a very small time, but when the time
is shared across 32 clients it becomes very important.  Number of
devices, or download speeds, will be negatively affected by use of
Bluetooth.  Smartphones should be used with WiFi instead.

Did you test with Bluetooth turned off on the NUC?  Bluetooth consumes
shared resources; time, RF spectrum, antenna, radio, module memory,
and bus.

> PS the above unit comes with an "Intel® Dual Band Wireless-AC 3165"
> which we'll throw out as nec!  Or perhaps it suffices among the
> smallest installations, with only 12 simultaneous WiFi connections?

You can't tell.  It might work better with that motherboard.

> Thanks to anybody who can contribute to this R
> discussion/evaluation, as Intel NUC's soldered-down internal WiFi
> (likewise limited to 12 WiFi connections) has in the end become too
> constricting~

This isn't an R discussion; since your first post on 26th August,
I've seen no data from engineers as to what the underlying problem is.
No wireless packet traces, no kernel logs, no error logs, nothing.

I agree you are doing an evaluation.  On that basis, I agree with
Anish that a wireless router will do much better.  If equipment loss
is a problem, glue it to the inside of the case.  If the case isn't
big enough, bolt it to the outside of the case, or get a bigger case.

p.s. the Rpi3 wireless is working great for me too.

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: [Server-devel] [XSCE] CentOS 7 quirk: VGA cable not re-attachable

[James Cameron]

On Mon, Aug 29, 2016 at 10:13:44PM -0400, Adam Holt wrote:
> - Fun Curiosity: you can reliably pull the power cable out of the
> LCD screen (or power cycle the LCD monitor with its own ON/OFF
> button) and VGA video *will* quite survive amazingly, promptly
> re-appearing on screen, anytime later when the monitor's power (or
> power cord) restores electricity!

That's normal, not curious.

Please learn this; your monitor has a device that is powered over the
cable by the computer, even if the power cable is out.  The device
says "here I am and this is what I can do!".

Key words for further educational searches; Display Data Channel
(DDC), Extended Display Identification Data (EDID).

My guess is that the driver for the DisplayPort or VGA adapter needs
an update, and an examination of the kernel events may show
differences between CentOS 7.2 and Fedora 22 at the instant of
plugging.  Give it to someone technical; someone who has good
understanding of the whole stack of components.  Maybe you can get the
fix backported to CentOS 7.2.

VGA connectors are not hotpluggable.  You'll notice the pins are all
the same length.  While it often works, there's no guarantee.

Oh, look, Wikipedia says that.

"Hot pluggable: Depends"

"The VGA interface is not engineered to be hotpluggable (so that the
user can connect or disconnect the output device while the host is
running), although in practice this can be done and usually does not
cause damage to the hardware or other problems. However, nothing in
the design ensures that the ground pins form the first makek or the
last break in the connection, so hotplugging may introduce surges in
signal lines which may or may not be adequately protected against
damage. Also, depending on the hardware and software, monitor
detection is sometimes unreliable when hotplugging a VGA connection."

... complete with obligatory typo.  ;-)

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: [Server-devel] [XSCE] CentOS 7 quirk: VGA cable not re-attachable

[James Cameron]

Only tested one monitor?  Try another; the EDID communication path
over VGA could have problems, or the data in monitor Flash unusual.

Also check age of monitor.  Mix not old and new wineskins.

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: [Server-devel] are NUC5i3RYH & NUC6i3SYH really limited to 12 WiFi clients??

[James Cameron]

Doesn't sound right.

What happened to stop the test?

Isn't the CentOS 7.2 kernel used in your test way older than the RPi3
kernel?

It would appear that CentOS 7.2 released with kernel 3.10 dated 30
June 2013, with minimal changes patched into it since.  Perhaps it
needs another fix.

RPi3 kernel with Rasbian is 4.4 dated 1 November 2015.

Bisect the problem broadly.  Try the latest kernel.

There have been many wireless driver and wireless networking changes
between the two kernels.  There's a possibility it may be one of them
you have hit.

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: [Server-devel] quasi-automatic security updates: state-of-art headed where?

[James Cameron]

On Mon, Aug 15, 2016 at 09:56:43AM -0400, Adam Holt wrote:
> "yum update" on CentOS regularly tells me that "yum-cron" should
> likely be installed+run instead.  Does anyone have experience /
> informal recommendations (or just tricks-of-the-trade, no matter how
> hacky) around yum-cron or dnf-automatic or similar, in Debian-land
> or all around?  Ideally some such/ similar would:
> 
> (1) be patient enough to deal with developing world servers being
> offline for many months at a time, yet smart enough to grab security
> updates quickly when Internet appears unpredictably every few weeks
> or months (or years, such server are almost inherently
> re/distributed re/donated re/sold Without centralized
> control...Internet-of-Things / IoT "anarchy" fears are indeed
> relevant+real ;)

Already implemented; a scriptlet is run by Network Manager when the
connection appears, and begins download of updates.

Downside is that the connection may have been made for a reason other
than downloading updates, and it can be quite irritating to the site
user to share their bandwidth with the updates.

As bandwidth increases for the 1% more such upstream policy decisions
are taken.  Removing agency from the owner.

> (2) Ideally only download "security" updates (however informally
> managed, anything closer to LTS than constant upgrades of
> major/minor packages!)  Is CentOS catching up to RHE on this front,
> or does Red Hat intentionally differentiate its products such that
> Red Hat Enterprise gets security updates faster or in a cleaner way?

Having been a Red Hat Enterprise support engineer; yes, Red Hat
customers do get security updates faster than CentOS does (of cour$e),
but the delay will vary.  From point of view of a CentOS user, worst
delay is where a Red Hat customer has asked for the issue to be fixed
and been given an early fix, and best delay is for a well known and
announced update.

> (3) Avoid bloating smaller (e.g. 120GB) SSD drives & 128GB MicroSD
> cards driving RPi3's and similar?  Sometimes I see ~100MB/week of
> updates from "yum update" which makes me wonder if yum is smart
> enough to fully delete not just deprecate older/unused packages?? 
> Hopefully I am wrong to fear disk bloat.  Or does this truly
> represent an estimated ~5GB/year of disk bloat in recent years,
> purely for OS-level updates?  (Part of a larger challenge on how to
> manage bloat of log files, content files, tmp files, user files for
> sure!)

Already implemented; old version of files are deleted when update is
applied, and downloads are deleted afterwards.  If you have disk
bloat, it will be some other cause.

> (4) Email the owner of the machine (offline is increasingly online,
> no matter how we slice it: offline servers will increasingly be at
> risk for years to come sad to say!)  Some kind of interface to Gmail
> or a similar online notification service should be possible when the
> server reappears online, without running a mail server heaven
> forbid?  For moments when truly-more-critical-intervention's
> required -- with the obvious risk that excessive "nagware" and
> "liabilityware" warnings will be ignored or far worse -- when local
> (often less-literate) operators even exist at all within developing
> world schools, trying their best!

Already implemented; several things provide this kind of service, and
a mail server is not required on the system.

> PS there's no perfect solution for sure -- Apple/Microsoft/Google
> spend many, many millions on their security-auto-updating infra
> (nevermind associated UX's) we just don't have.  Offline security
> updates on a quasi-monthly basis may be one answer, if older
> quasi-offline solutions from the 1990s also/still have legit hope &
> lessons for us all?  What should such services cost, if it comes
> down to money?

As much as the market will pay.  Which can be from zero to lots.

> As real-world challenges evolve: what practical non-puritanical
> compromises are evolving out there in CentOS-land / Debian-land /
> Etc as "offline increasingly becomes online" ?  (as serious online
> risks increasingly reach offline servers, IoT devices, etc...what
> quasi-automatic security update regimes should we be looking at for
> the coming decade, realistically?)  Thanks for ideas facing up to
> these existential challenges uniting us all, knowing there's a
> serious diversity of opinions / tripwires / threat models /
> mitigations out there quite naturally!

Go for a shorter planning horizon than a decade.

-- 
James Cameron
http://quozl.netrek.org/
___
Server-devel mailing list
Server-devel@lists.laptop.org
http://lists.laptop.org/listinfo/server-devel

Re: "Physics with XO" by Guzmán Trinidad now available in English

[James Cameron]

On Mon, Jul 04, 2016 at 01:16:26AM -0500, Sebastian Silva wrote:
> If you find anything that can be improved, do let us know; it will
> be appreciated.

Quoting from
https://icarito.gitbooks.io/physics-with-xo/content/09_precision_accuracy_ADC.html

> It is important to highlight that at the moment of finishing this
> investigation we detected an important problem with this model: When
> the TB Activity attempts to measure voltage or resistance in one
> channel, it randomly obtains the readings belonging to one channel
> or the other; this problem makes it impossible, for the moment, to
> use the applications that involve these measurements. In queries
> made about this issue, we were informed that it is a low level error
> ("bug") that is being worked on.

And also from https://icarito.gitbooks.io/physics-with-xo/content/annexes.html

> At the time of finishing this work, this model consistently fails in
> the following way: when running any TB program it may measure the
> correct channel (generally CHL) or it will randomly return
> measurements coming from the other channel (CHR) (this error is
> generally known as swapping). This causes programs that measure
> volume, frequency, resistance and voltage to either function
> properly or not. Having made the query to the experts, we were
> informed that they were working on a solution involving low level
> modifications to the XO.

Both of these were the same underlying cause, were fixed last
September, and the fix is in OLPC OS 13.2.7.  Please upgrade kernel or
operating system.

http://dev.laptop.org/git/olpc-kernel/commit/?h=arm-3.0-wip=95c24cb3859cd01a9689675304ee85179770bbbf

(removed CC for lists not subscribed).

-- 
James Cameron
http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel