Hello,
I have a problem with route monitoring.
Everything is OK when I use interface monitoring or gatway monitoring, but host monitoring
method doesnt work correctly.
My configuration:
Firmware Version: 2.27.02.11-14417 Aug 23 2010
ISP1 xdsl line -> modem -> WAN1
ISP2 adsl line -> modem -> WAN2
I've 2 routing tables.
1. main(default)
M 0.0.0.0/0 wan1 1.1.1.1 80
0.0.0.0/0 wan2 2.2.2.2 90
and
2. WWW-wan2
M 0.0.0.0/0 wan2 2.2.2.2 80
0.0.0.0/0 wan1 1.1.1.1 90
There is a routing rule to forward http traffic via WWW-wan2 table.
When I unplug adsl line from modem(ISP2), dfl-800 (based on host monitoring) recognize
route failure after ~30s, mark first route in WWW-wan2 table as X(disabled) and move
http trafic to second route (wan1).
X M 0.0.0.0/0 wan2 2.2.2.2 80
0.0.0.0/0 wan1 1.1.1.1 90
Unfortinatly after another 10 seconds, first route is enable again

I run sniffer to see what's happening.
When first route is disabled, dfl-800 keeps trying to ping monitored hosts via wan2.
When dfl-800 gets "destination unreachable" answers for his ICMP echo requests, route is re-enabled.
It looks like it dosen't matter if the answer is:
Type: 0 (Echo (ping) reply)
Code: 0 ()
or
Type: 3 (Destination unreachable)
Code: 0 (Network unreachable).
Monitoring configuration:
First route in WWW-wan2 table is monitored with:
Monitor Interface Link Status
Monitor Gateway using ARP
"enable host monitoring"
grace period 5 seconds
Minimum Number of Hosts Reachable: Specific 2
Monitored Hosts tab: hosts 3.3.3.3, 4.4.4.4, 5.5.5.5
method ICMP
Polling Interval: 10000ms
Samples: 10
Max Poll Fails: 2
Max Average Latency: 1000ms
Am I doing something wrong, or this is a bug in firmware?
Navi