topic problems with ip route add default in ExtremeSwitching (EOS)

problems with ip route add default

chrisfipe_f — Thu, 17 Nov 2016 09:08:00 GMT

I have two iproute add default but with different metric , when I ingress sh iproute in console display me the route with lower metric without # , sombody why this??

RE: problems with ip route add default

Im_JeongJin — Thu, 17 Nov 2016 10:11:00 GMT

Hi Chrisfipe..

I could see correct iproute info in my lab.

* BD1.53 # show iproute
Ori Destination Gateway Mtr Flags VLAN Duration
#s Default Route 192.168.55.1 10 UG---S-um--f CRBT 0d:0h:0m:52s

but in case there is no port assigned on the vlan.. ( in my case CRBT ), "show ip route" output didn't show # mark.

* BD1.63 # sh iproute
Ori Destination Gateway Mtr Flags VLAN Duration
s Default Route 192.168.55.1 10 -G---S-um--- CRBT 0d:0h:0m:2s

* BD1.64 # sh vlan "CRBT"
VLAN Interface with name CRBT created by user
Admin State: Enabled Tagging: 802.1Q Tag 10
Virtual router: VR-Default
Primary IP : 192.168.55.2/24
IPv6: None
STPD: None
Protocol: Match all unfiltered protocols
Loopback: Disabled
NetLogin: Disabled
QosProfile: None configured
Egress Rate Limit Designated Port: None configured
Flood Rate Limit QosProfile: None configured
Ports: 0. (Number of active ports=0)

RE: problems with ip route add default

Im_JeongJin — Thu, 17 Nov 2016 10:11:00 GMT

Here is the result, when switch has two different path for default route.

* BD1.79 # sh iproute
Ori Destination Gateway Mtr Flags VLAN Duration
s Default Route 192.168.55.1 10 -G---S-um--- CRBT 0d:0h:0m:3s
#s Default Route 192.168.157.1 100 UG---S-um--f U-CG 0d:0h:1m:17s

even 192.168.55.1 configured with lower metric, it has no port on that VLAN, so cannot used as best path for default route.

you can also check with the command "debug hal show ipv4fib" to see which path used for forwarding the traffic.

* BD1.80 # debug hal show ipv4Fib
HAL IPv4 FIB:
VrId Prefix/Mask Gateway Vlan MAC Port Or Ref Ins Flags Bucket
===================================================================
2 000.000.000.000/ 0 192.168.157.001 1000026 00:00:00:00:00:00 00000000 3 1 1 F--- 9999

RE: problems with ip route add default

Prashanth_KG — Thu, 17 Nov 2016 10:47:00 GMT

Hi

When we have 2 default routes with different metrics, we will not be able to load share the traffic using both the routes. So, in this scenario one route would be active and the other route will be backup and will be used only when the primary route is not present/active.

# sign indicates if a particular route is in used for forwarding the traffic or not. So, it is expected not to see the sign for the backup route.

#s Default Route 10.2.0.2 10 UG---S-um--f- pe2_syd1_to_pe1_syd2 0d:0h:0m:56s
s Default Route 10.2.0.22 30 UG---S-um---- pe2_syd1_to_pe1_syd1 0d:0h:0m:43s

Hope this helps!

RE: problems with ip route add default

Henrique — Thu, 17 Nov 2016 19:33:00 GMT

When you have a default route as well as any other static route, you should see the # when the GW vlan is UP (at least 1 port UP). If the vlan is down, then you won't see the #.

To confirm if the route is being used you can check the "f" flag that means "Provided to FIB".

Example 1:

Ori Destination Gateway Mtr Flags VLAN Duration
s Default Route 10.10.0.181 1 -G---S-um---- Link1 0d:0h:1m:34s
#s Default Route 10.10.0.189 2 UG---S-um--f- Link2 0d:0h:1m:25s

Route with lower metric through vlan Link1 (metric 1) is down and the next best route will be used (with higher metric).

Example 2:

Ori Destination Gateway Mtr Flags VLAN Duration
#s Default Route 10.10.0.181 1 UG---S-um--f- Link1 0d:0h:12m:12s
#s Default Route 10.10.0.189 1 UG---S-um---- Link2 0d:0h:0m:5s

Both routes are UP, but the route with lower metric is being used. Again, you can confirm that based on the "f" flag.