A number of people emailed me asking to let them know if I found a solution or made any progress on this. I've had some time to work on it and think I have the RPM stuff figured out. Now I just need to actually use what RPM tracks in some "event-options" to actually have the system respond autonomously.
For more details and configuration code, see the my thread at the J-Net Forums, http://forums.juniper.net/t5/SRX-Services-Gateway/Dual-ISP-Failover-via-RPM/m-p/72700/highlight/false#M8459 On 1/24/2011 at 8:54 PM, "Crist Clark" <crist.cl...@globalstar.com> wrote: > I've got a site with multiple Internet links. I want to continuously > monitor Internet connectivity across all links although I only plan > to use one at a time for production traffic. This is fail over only, > not load balancing. > > Just routing by link up or down is not sufficient. All of the links > terminate as Ethernet on the device, an SRX 240H, with switches > between it and the CPE. > > I found the Real Time Monitoring (RPM) features in JUNOS, and it seemed > perfect. I could set up a few ICMP pings and HTTP GETs to some reliable > locations and then only fail over when the preferred ISP has more > failures than the backup(s). > > But I'm having problems getting this to work. I thought I could set up > a routing instance with the default route out each ISP then set up a > RPM test associated with each routing instance, after all, the knobs > seem to be in place to do this, but it does not work. From the research > I've done, it seems that a forwarding routing instance won't actually > affect the packets originating on the host itself? > > So what is the right way to do this? Am I on the right track? BTW, this > is running 10.0, but upgrading is definitely an option. -- Crist Clark Network Security Specialist, Information Systems Globalstar 408 933 4387 _______________________________________________ juniper-nsp mailing list juniper-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/juniper-nsp