Perhaps use the ntp client included in the OpenWRT BusyBox rather than ntpdate or install the full NTP client. See <a href="http://wiki.openwrt.org/doc/howto/ntp.client">http://wiki.openwrt.org/doc/howto/ntp.client</a><br>
<br>Bruce Ford<br><br><div class="gmail_quote">On Sat, Mar 3, 2012 at 10:11 AM, Ben West <span dir="ltr"><<a href="mailto:me@benwest.name">me@benwest.name</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0pt 0pt 0pt 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
I'm running OLSR v0.6.1 packaged with OpenWRT 10.03.1 on a collection of Ubiquiti M5 access points (Rocket, Bullet, Nanostation, Nano Loco).<div><br></div><div>All radios have /etc/init.d/ntpdate service enabled on boot to sync local time to a remote NTP server, although this only succeeds on non-gateway nodes if OLSRd (also enabled on boot) finds and sets a default route before ntpdate times out.</div>
<div><br></div><div>In the case where ntpdate does time out before setting a node's local time, I've noticed letting that node operate as-is in the mesh can cause other nodes to lose their default route. Presumably from the non-sync'ed node sending out HNA's with bad timestamps? Is this expected OLSR behavior?</div>
<div><br></div><div>Besides simply letting ntpdate run on all nodes as an hourly or daily cron job, would anyone have other recommendations on keeping all nodes' local time synced?<span class="HOEnZb"><font color="#888888"><br clear="all">
<div><br></div>-- <br>
<div>Ben West</div><div><a href="mailto:me@benwest.name" target="_blank">me@benwest.name</a></div><br>
</font></span></div>
<br>--<br>
Olsr-users mailing list<br>
<a href="mailto:Olsr-users@lists.olsr.org">Olsr-users@lists.olsr.org</a><br>
<a href="https://lists.olsr.org/mailman/listinfo/olsr-users" target="_blank">https://lists.olsr.org/mailman/listinfo/olsr-users</a><br></blockquote></div><br>