<div dir="ltr"><div>Apologies for previous pollution on olsrd commit logs. Github to push my local merged branch directly to your repo.<br><br></div>Here is the PR for olsrd v0.6.6.1:<br><a href="https://github.com/opentechinstitute/olsrd/pull/22">https://github.com/opentechinstitute/olsrd/pull/22</a><br>
</div><div class="gmail_extra"><br><br><div class="gmail_quote">On Wed, Apr 23, 2014 at 4:54 PM, Ben West <span dir="ltr"><<a href="mailto:ben@gowasabi.net" target="_blank">ben@gowasabi.net</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div dir="ltr"><div><div><div>So far, the specific atheros repeater node where I'd been having apparent SmartGateway issues seems to be OK now after upgrading its local olsrd to v0.6.6.1 yesterday (along with the same on the affected gateway nodes). It does indeed look like the SmartGateway updates have yielded improvement for my situation.<br>
<br></div>I should point out that <b>option 'SmartGatewayUseCount' '1'</b> is needed to tell the newer SmartGateway code to only use one gateway at a time. The OLSR folks are now testing out multiple gateway nodes simultaneously in HEAD, but I believe it doesn't play wall with typical NAT configurations, so just use one gateway at a time for now.<br>
<br></div>The rationale for the timing tweaks from Ferry are somewhat opaque to me. The old ROBIN mesh firmware using olsrd v0.5 did deploy similar tweaks:<br><a href="https://trac.assembla.com/RobinMesh/browser/trunk/robin/robin-mesh/files/etc/config/olsr" target="_blank">https://trac.assembla.com/RobinMesh/browser/trunk/robin/robin-mesh/files/etc/config/olsr</a><br>
<br></div><div><div><table><tbody><tr><th><a href="https://trac.assembla.com/RobinMesh/browser/trunk/robin/robin-mesh/files/etc/config/olsr#L16" target="_blank">16</a></th><td>config "olsr" "interface"</td>
</tr><tr><th><a href="https://trac.assembla.com/RobinMesh/browser/trunk/robin/robin-mesh/files/etc/config/olsr#L17" target="_blank">17</a></th><td> option HelloInterval '2.0'</td></tr><tr><th><a href="https://trac.assembla.com/RobinMesh/browser/trunk/robin/robin-mesh/files/etc/config/olsr#L18" target="_blank">18</a></th>
<td> option HelloValidityTime '40.0'</td></tr><tr><th><a href="https://trac.assembla.com/RobinMesh/browser/trunk/robin/robin-mesh/files/etc/config/olsr#L19" target="_blank">19</a></th><td> option TcInterval '5.0' </td>
</tr><tr><th><a href="https://trac.assembla.com/RobinMesh/browser/trunk/robin/robin-mesh/files/etc/config/olsr#L20" target="_blank">20</a></th><td> option TcValidityTime '100.0'</td></tr><tr><th><a href="https://trac.assembla.com/RobinMesh/browser/trunk/robin/robin-mesh/files/etc/config/olsr#L21" target="_blank">21</a></th>
<td> option MidInterval '18.0'</td></tr><tr><th><a href="https://trac.assembla.com/RobinMesh/browser/trunk/robin/robin-mesh/files/etc/config/olsr#L22" target="_blank">22</a></th><td> option MidValidityTime '324.0'</td>
</tr><tr><th><a href="https://trac.assembla.com/RobinMesh/browser/trunk/robin/robin-mesh/files/etc/config/olsr#L23" target="_blank">23</a></th><td> option HnaInterval '18.0'</td></tr><tr><th><a href="https://trac.assembla.com/RobinMesh/browser/trunk/robin/robin-mesh/files/etc/config/olsr#L24" target="_blank">24</a></th>
<td> option HnaValidityTime '108.0'</td></tr></tbody></table><br></div><div>Antonio from ROBIN published a the calculations behind these timings on his wiki <a href="http://wiki-robin.meshroot.com/" target="_blank">http://wiki-robin.meshroot.com/</a> ,
but unfortunately that site now seems lost to the sands of time
(and likewise to <a href="http://archive.org" target="_blank">archive.org</a>).<br><br>It is likely these params are unnecessary. I was initially wanting to mimic Ferry's suggested config to rule out problems.<br><br>
</div><div>I can see about making the PR to the commotion branch later today.<br>
</div><div><br></div>
</div></div><div class="gmail_extra"><div><div class="h5"><br><br><div class="gmail_quote">On Wed, Apr 23, 2014 at 3:06 PM, Josh King <span dir="ltr"><<a href="mailto:jking@opentechinstitute.org" target="_blank">jking@opentechinstitute.org</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hey Ben,<br>
<br>
How about you open a pull request against the 'commotion' branch in<br>
upstream? I'll go ahead and test and then merge that into the rc2<br>
testing images I'm working on.<br>
<br>
Also, I'm interested in the defaults you got from Ferry in the config<br>
file below. Is there a rationale behind those that exists somewhere?<br>
<div><div><br>
On 04/22/2014 12:25 PM, Ben West wrote:<br>
> Thus far only deployed to a handful of atheros 802.11g nodes running<br>
> r40431 OpenWRT AA (i.e. not commotion-router proper, nor yet on ar71xx<br>
> target, nor with olsrd-mdp plugin). The specific failure I was seeing<br>
> on an atheros repeater node has not recurred yet, although I only<br>
> reflashed it a couple hours ago. I can at least say that v0.6.6.1 is<br>
> working just fine on those nodes previously running olsrd v0.6.5-4. No<br>
> obvious regression.<br>
><br>
> Here is the /etc/config/olsrd I'm using, for your reference, based on<br>
> advice from Ferry Huberts:<br>
><br>
> config olsrd<br>
> # uncomment the following line to use a custom config file instead:<br>
> #option config_file '/etc/olsrd.conf'<br>
><br>
> option 'IpVersion' '4'<br>
> option 'LinkQualityLevel' '2'<br>
> option 'LinkQualityAlgorithm' 'etx_ffeth'<br>
> option 'SmartGateway' 'yes'<br>
> option 'SmartGatewayUseCount' '1'<br>
> option 'Pollrate' '0.1'<br>
> option 'TcRedundancy' '2'<br>
> option 'MprCoverage' '5'<br>
> option 'LinkQualityFishEye' '0'<br>
><br>
> config 'LoadPlugin'<br>
> option 'library' 'olsrd_arprefresh.so.0.1'<br>
><br>
> config 'LoadPlugin'<br>
> option 'library' 'olsrd_dyn_gw.so.0.5'<br>
> option 'HNA' '0.0.0.0 0.0.0.0'<br>
><br>
> config 'LoadPlugin'<br>
> option 'library' 'olsrd_nameservice.so.0.3'<br>
> #option 'resolv_file' '/tmp/resolv.conf.auto'<br>
> option 'sighup_pid_file' '/var/run/dnsmasq.pid'<br>
> option 'suffix' '.mesh'<br>
><br>
> config 'LoadPlugin'<br>
> option 'library' 'olsrd_txtinfo.so.0.1'<br>
> option 'port' '2006'<br>
> option 'Accept' '127.0.0.1'<br>
><br>
> config 'Interface'<br>
> list 'interface' 'wan'<br>
> option 'Mode' 'ether'<br>
><br>
> config 'Interface'<br>
> list 'interface' 'mesh'<br>
> option 'Ip4Broadcast' '255.255.255.255'<br>
> option 'Mode' 'mesh'<br>
> #peformance-tuned params from Ferry Huberts <<a href="mailto:mailings@hupie.com" target="_blank">mailings@hupie.com</a><br>
</div></div>> <mailto:<a href="mailto:mailings@hupie.com" target="_blank">mailings@hupie.com</a>>><br>
<div>> option HelloInterval '2.0'<br>
> option HelloValidityTime '20.0'<br>
> option TcInterval '5.0'<br>
> option TcValidityTime '30.0'<br>
> option MidInterval '5.0'<br>
> option MidValidityTime '50.0'<br>
> option HnaInterval '5.0'<br>
> option HnaValidityTime '15.0'<br>
><br>
><br>
><br>
><br>
> On Tue, Apr 22, 2014 at 11:13 AM, Josh King <<a href="mailto:jking@opentechinstitute.org" target="_blank">jking@opentechinstitute.org</a><br>
</div><div><div>> <mailto:<a href="mailto:jking@opentechinstitute.org" target="_blank">jking@opentechinstitute.org</a>>> wrote:<br>
><br>
> Hey Ben,<br>
><br>
> That sounds great. Have you tried it out on any images yet? I was going<br>
> to start rolling 1.1rc2 images maybe later today, and if that release<br>
> fixes some SmartGateway bugs then it might be worth including.<br>
><br>
> On 04/22/2014 11:56 AM, Ben West wrote:<br>
> > Howdy,<br>
> ><br>
> > I'd been having problems with the v0.6.5-4 version of olsrd in<br>
> OTI's git<br>
> > repo having sporadic gateway problems using the SmartGateway feature,<br>
> > specifically with repeater nodes seeing their route to the Internet<br>
> > occasionally broken in the presence of multiple gateway nodes.<br>
> ><br>
> > Some responses on the olsr-users listserv indicated quite a few<br>
> > improvements to SmartGateway in the v0.6.6 release, especially for<br>
> > situations with multiple gateway nodes, so I made a merged branch:<br>
> ><br>
> > <a href="https://github.com/opentechinstitute/olsrd/tree/commotion" target="_blank">https://github.com/opentechinstitute/olsrd/tree/commotion</a><br>
> > +<br>
> > <a href="http://olsr.org/git/?p=olsrd.git;a=shortlog;h=refs/heads/release-0.6.6" target="_blank">http://olsr.org/git/?p=olsrd.git;a=shortlog;h=refs/heads/release-0.6.6</a><br>
> > =<br>
> > <a href="https://github.com/westbywest/olsrd/tree/commotion-0.6.6" target="_blank">https://github.com/westbywest/olsrd/tree/commotion-0.6.6</a><br>
> ><br>
> > (New branch name "commotion-0.6.6")<br>
> ><br>
> > Posting here in case there is interest to push this branch upstream to<br>
> > OTI's repo too.<br>
> ><br>
> > --<br>
> > Ben West<br>
> > <a href="http://gowasabi.net" target="_blank">http://gowasabi.net</a><br>
> > <a href="mailto:ben@gowasabi.net" target="_blank">ben@gowasabi.net</a> <mailto:<a href="mailto:ben@gowasabi.net" target="_blank">ben@gowasabi.net</a>><br>
</div></div>> <mailto:<a href="mailto:ben@gowasabi.net" target="_blank">ben@gowasabi.net</a> <mailto:<a href="mailto:ben@gowasabi.net" target="_blank">ben@gowasabi.net</a>>><br>
> > <a href="tel:314-246-9434" value="+13142469434" target="_blank">314-246-9434</a> <tel:<a href="tel:314-246-9434" value="+13142469434" target="_blank">314-246-9434</a>> <tel:<a href="tel:314-246-9434" value="+13142469434" target="_blank">314-246-9434</a> <tel:<a href="tel:314-246-9434" value="+13142469434" target="_blank">314-246-9434</a>>><br>
<div>> ><br>
> ><br>
> > _______________________________________________<br>
> > Commotion-dev mailing list<br>
> > <a href="mailto:Commotion-dev@lists.chambana.net" target="_blank">Commotion-dev@lists.chambana.net</a><br>
</div>> <mailto:<a href="mailto:Commotion-dev@lists.chambana.net" target="_blank">Commotion-dev@lists.chambana.net</a>><br>
<div>> > <a href="https://lists.chambana.net/mailman/listinfo/commotion-dev" target="_blank">https://lists.chambana.net/mailman/listinfo/commotion-dev</a><br>
> ><br>
><br>
> --<br>
> Josh King<br>
> Lead Technologist<br>
> The Open Technology Institute<br>
> <a href="http://opentechinstitute.org" target="_blank">http://opentechinstitute.org</a><br>
> PGP Fingerprint: 8269 ED6F EA3B 7D78 F074 1E99 2FDA 4DA1 69AE 4999<br>
><br>
><br>
><br>
><br>
> --<br>
> Ben West<br>
> <a href="http://gowasabi.net" target="_blank">http://gowasabi.net</a><br>
</div>> <a href="mailto:ben@gowasabi.net" target="_blank">ben@gowasabi.net</a> <mailto:<a href="mailto:ben@gowasabi.net" target="_blank">ben@gowasabi.net</a>><br>
> <a href="tel:314-246-9434" value="+13142469434" target="_blank">314-246-9434</a><br>
<div><div><br>
--<br>
Josh King<br>
Lead Technologist<br>
The Open Technology Institute<br>
<a href="http://opentechinstitute.org" target="_blank">http://opentechinstitute.org</a><br>
PGP Fingerprint: 8269 ED6F EA3B 7D78 F074 1E99 2FDA 4DA1 69AE 4999<br>
<br>
</div></div></blockquote></div><br><br clear="all"><br>-- <br>Ben West</div></div><div><a href="http://gowasabi.net" target="_blank">http://gowasabi.net</a><br><a href="mailto:ben@gowasabi.net" target="_blank">ben@gowasabi.net</a><br>
<a href="tel:314-246-9434" value="+13142469434" target="_blank">314-246-9434</a><br></div>
</div>
</blockquote></div><br><br clear="all"><br>-- <br>Ben West<div><a href="http://gowasabi.net" target="_blank">http://gowasabi.net</a><br><a href="mailto:ben@gowasabi.net" target="_blank">ben@gowasabi.net</a><br>314-246-9434<br>
</div>
</div>