headers
Henning Rogge | 23 May 2013 20:51

Olsrd2, version 0.2.1

Hi,

I tagged the version 0.2.1 of Olsrd2 today and I think this might be
the first version interesting enough for other people to try out.

I think the software is still in an "Alpha" stage, its not even
feature complete and a few parts are prototypes that will be removed
later, but it already is doing its job.

What do I need to compile Olsrd2?
*********************************

1.) A reasonable current build system (gcc, ...).
2.) CMake version 2.8.5 or newer
3.) libnl3-dev or libnl-tiny

(yes, both the current stable and trunk of OpenWRT are okay for this)

How to compile Olsrd2 by hand?
******************************

1.) clone the OONF-API repository (Olsr.Org Network Framework)

git clone http://olsr.org/git/oonf_api.git

2.) go into the "build" subdirectory of the cloned repository and compile it

cd oonf_api/build
cmake ..
make
(Continue reading)

Permalink | Reply | 6 comments |
headers
Hans-Christoph Steiner | 21 May 2013 21:58

'debian/' folder should not be included in release tarballs!


Hey all,

I just tried updating the official Debian package.  Normally, its all
automated and I can just run "svn-upgrade" and it'll fetch the tarball from
olsr.org and setup the rest.

But recently, the 'debian/' folder has been added to the release tarballs,
breaking this process.  The 'debian/' folder must not be included in the
release tarballs.  That is the standard Debian procedure, and including it
breaks all the tools.

Its fine if there is a debian/ folder included in the olsrd git, but I don't
think anyone maintains it, so it should be removed in favor of the official
debian packaging:

svn://svn.debian.org/svn/collab-maint/deb-maint/olsrd/
http://packages.debian.org/source/olsrd

.hc

--

-- 
Olsr-dev mailing list
Olsr-dev <at> lists.olsr.org
https://lists.olsr.org/mailman/listinfo/olsr-dev
Permalink | Reply | 13 comments |
headers
Hans-Christoph Steiner | 2 May 2013 17:45

Test the official Android Adhoc mode!


Bruno and others are making some serious progress in getting adhoc mode
officially supported in Android!  His code has been integrated into
CyanogenMod, and is ready for testing:

> -------- Original Message --------
> Date: Thu, 02 May 2013 16:20:11 +0100
> From: Bruno Randolf <br1 <at> einfach.org>
> 
> The easiest way to test it is via the nightly builds of CyanogenMod 
> (http://get.cm/?type=nightly). Devices which have the kernel driver 
> support so far are:
> 
> * Nexus S ("crespo")
> * Galaxy Nexus ("maguro")
> * Nexus 7 ("grouper")
> * Samsung Galaxy S, Samsung Captivate, Samsung Vibrant and Samsung 
> Fascinate ("aries")
> 
> For many other popular Samsung devices (like Galaxy S3, Galaxy Tab 2, 
> etc..) the problem is that they use a different version of the bcmdhd 
> driver and our patch is not enough to add IBSS mode support. It would be 
> some work to port everything over, and this seems to be quite hard. I 
> don't have the devices, and I want to concentrate on the Nexus line and 
> AOSP inclusion anyways.
> 
> BTW: Anyone knows about IBSS mode support in the WLAN driver ("WCNSS", 
> "prima") for the Nexus 4?
> 
> I always try to document news and the current status of IBSS mode on
(Continue reading)

Permalink | Reply | 1 comment |
headers
Hans of Guardian | 28 Apr 2013 21:49

coordinating test infrastructure


I'm just heading back to NYC after a long weekend in Washington DC working with the OTI Commotion crew.  One
big piece of news I learned on my trip there is about the testing infrastructure that they have been working
on.  Additionally Ben West as been working on VM-based tests.  I watched Ferry and Henning's Battlemesh
talks and saw Ferry's Jenkins setup.

I can start with a piece of bad news: the Jenkins test builds that I had setup for olsrd have vanished with that
whole server, due to the collapse of the organization that was maintaining it.

There seems to be lots of people working on test infrastructure, so I think its a good time to discuss who is
doing what and how we can coordinate. I'll start with the various efforts I've heard something about:

* Ferry's Jenkins
* Ben West's VMs and python tests
* battlemesh shell scripts
* OTI lua scripts
* OTI+LTS system for wifi/mesh performance in changing physical space
* CONFINE setup? (I've only vaguely heard of this)

One very nice potential test bed for Android adhoc issues is some commercial services that host actual
Android devices in a Jenkins-accessible network setup.  Commotion is exploring whether there is any
funds to allocate to that.

What else is out there, and how can we coordinate and collaborate?

.hc
--

-- 
Olsr-dev mailing list
Olsr-dev <at> lists.olsr.org
https://lists.olsr.org/mailman/listinfo/olsr-dev
(Continue reading)

Permalink | Reply | 0 comments |
headers
Bastian Bittorf | 27 Apr 2013 10:44
Favicon
Gravatar

missing arproaming-plugin

the some problem like the OBAMP-plugin occured to
the arproaming-plugin, it was scrubbed while merging
branches. i can retrieve it via

git checkout old-master-branch -- lib/arproaming

what is the correct way to move to new master
without loosing history?

bye, bastian

--

-- 
Olsr-dev mailing list
Olsr-dev <at> lists.olsr.org
https://lists.olsr.org/mailman/listinfo/olsr-dev
Permalink | Reply | 9 comments |
headers
Bastian Bittorf | 26 Apr 2013 10:14
Favicon
Gravatar

Re: [PATCH] fix function olsrd_write_plparam() in olsrd.init

* Job <thomas <at> net.t-labs.tu-berlin.de> [26.04.2013 09:02]:

commit in
https://github.com/openwrt-routing/packages/commit/16f7810f11cbf74e433f74beb95f9861e4e56fa1

bye, bastian

--

-- 
Olsr-dev mailing list
Olsr-dev <at> lists.olsr.org
https://lists.olsr.org/mailman/listinfo/olsr-dev
Permalink | Reply | 0 comments |
headers
Steve Witt | 25 Apr 2013 22:13
Favicon

OBAMP

I ran across some information about the OBAMP multicast routing protocol 
that was added as a plugin to olsrd in 2009 as a Google Summer of Code 
project. I was looking through the olsrd git repo and noticed that the 
obamp source directory was present on a branch for a couple of years from 
2009 to 2011 or so and then it seems that it just disappears after a merge 
of that branch in 2011. I didn't see any comments as part of a commit when 
it was removed or any discussion on this mailing list about the reason 
that it was deleted. I was wondering if anyone might remember why it was 
dropped or anything else concerning this.

Thanks...

--

-- 
Olsr-dev mailing list
Olsr-dev <at> lists.olsr.org
https://lists.olsr.org/mailman/listinfo/olsr-dev
Permalink | Reply | 9 comments |
headers
Saverio Proto | 25 Apr 2013 18:36
Picon
Gravatar

Re: [PATCH] fix olsr.init script function olsrd_write_plparam() as it wrongly removes dashes. olsrd deamon did not start when the nameservice plugin was active.

I loaded your same configuration and I cannot reproduce the bug :(
can you attach the version of /etc/init.d/olsrd you are using ?

I am using this one at commit 0015a5a216914ddf35abccefb530b0f7c651c857
of this repo: https://github.com/openwrt-routing/

If you are using a new version, we might had introduced the bug.

Bastian are you with us ?

Saverio

2013/4/25 Thomas Hühn <thomas <at> net.t-labs.tu-berlin.de>:
> Hi,
>
> My olsr.conf:
>
> root <at> Bluse-home:~# cat /etc/config/olsrd
>
> config olsrd 'olsrd'
> option IpVersion '4'
> option FIBMetric 'flat'
> option LinkQualityLevel '2'
> option LinkQualityAlgorithm 'etx_ff'
> option OlsrPort '698'
> option Willingness '7'
> option NatThreshold '1.0'
> option UseHysteresis 'no'
> option SmartGateway 'no'
> option LinkQualityFishEye '1'
(Continue reading)

Permalink | Reply | 0 comments |
headers
Saverio Proto | 25 Apr 2013 17:41
Picon
Gravatar

Re: [PATCH] fix olsr.init script function olsrd_write_plparam() as it wrongly removes dashes. olsrd deamon did not start when the nameservice plugin was active.

Can you send me the UCI configuration that was not working for you
before the patch ?

thanks

Saverio

2013/4/25 Thomas Huehn <thomas <at> net.t-labs.tu-berlin.de>:
> Signed-off-by: Thomas Huehn <thomas <at> net.t-labs.tu-berlin.de>
> ---
>  olsrd/files/olsrd.init | 5 ++++-
>  1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/olsrd/files/olsrd.init b/olsrd/files/olsrd.init
> index 7ffe33c..f4641aa 100644
> --- a/olsrd/files/olsrd.init
> +++ b/olsrd/files/olsrd.init
>  <at>  <at>  -153,8 +153,11  <at>  <at>  olsrd_write_plparam() {
>                 return 1
>         fi
>
> -       oldIFS="$IFS"; IFS='-_'; set -- $option; IFS="$oldIFS"
> +       oldIFS="$IFS"
> +       IFS='-_'
> +       set -- $option
>         option="$*"
> +       IFS="$oldIFS"
>         _option="$option"
>
>         if [ "$option" = 'hosts' ]; then
(Continue reading)

Permalink | Reply | 9 comments |
headers
Ferry Huberts | 24 Apr 2013 16:18
Gravatar

config files will move after the 0.6.6 release

A heads-up...

I'm planning to move the olsrd configuration files - that are currently
stored under /etc - under /etc/olsrd

This is more in-line with what distributions do and makes it easier to
manage the files, since we've been getting a few more configuration
files over time.

-- 
Ferry Huberts

--

-- 
Olsr-dev mailing list
Olsr-dev <at> lists.olsr.org
https://lists.olsr.org/mailman/listinfo/olsr-dev
Permalink | Reply | 2 comments |
headers
velt | 19 Apr 2013 18:51
Picon
Picon
Favicon

[PATCH 1/1] Fix setsockopt for setting Traffic Class in IPv6

From: Ronald in 't Velt <velt <at> xs4all.nl>

HiOLSR-developers,

For your consideration:

Setting Traffic Class (the counterpart of TOS in IPv4) on OLSR IPv6 packets is currently broken.
The code does not distinguish between the IPv4 and IPv6 cases, but it should, as the required 
'level' and 'optname' arguments to setsockopt differ.

I tested this patch by monitoring OLSR IPv6 packets with 'tcpdump -v': it appears to be working.

Note this is for Linux only.

This is my modest contribution to Dave Täht's quest to fix "TOS setting" in IPv6 :-)
See: http://www.bufferbloat.net/issues/249

Thanks,
Ronald

Signed-off-by: Ronald in 't Velt <velt <at> xs4all.nl>
---
 src/unix/ifnet.c |   14 +++++++++++---
 1 files changed, 11 insertions(+), 3 deletions(-)

diff --git a/src/unix/ifnet.c b/src/unix/ifnet.c
index 654c169..5ec8e32 100644
--- a/src/unix/ifnet.c
+++ b/src/unix/ifnet.c
 <at>  <at>  -783,9 +783,17  <at>  <at>  chk_if_up(struct olsr_if *iface, int debuglvl __attribute__ ((unused)))
(Continue reading)

Permalink | Reply | 1 comment |

Gmane