Skip to contentSkip to conversation

1st default gateway is always dead!

LARTC Shanker Balan 2 participants 2 messages Jan 18, 2002

lore
Share & tools
lore
#0 lore raw

Hello:

I have been trying to use multiple default gateways over the last 2 days on my Debian box and have been fairly successful in understanding the fundas. Bert, thanx a lot for all those wonderful documentation at
http://www.ds9a.nl/2.4Routing/.

Something strange is happening when I try adding multiple default gateways:

# ip ro add default nexthop via 192.168.1.100 dev eth0 \
					nexthop via 192.168.3.254 dev eth2
# ip ro ls
192.168.3.0/24 dev eth2  proto kernel  scope link  src 192.168.3.24 
192.168.2.0/24 dev eth1  proto kernel  scope link  src 192.168.2.254 
192.168.1.0/24 dev eth0  proto kernel  scope link  src 192.168.1.24 
default 
		nexthop via 192.168.1.100  dev eth0 weight 1 dead
		nexthop via 192.168.3.254  dev eth2 weight 1 ^^^^

The first nexthop is _always_ marked dead for some reason though my primary gateway "192.168.1.100" _is_ reachable via eth0.

# ping 192.168.1.100 -c 3
PING 192.168.1.100 (192.168.1.100): 56 data bytes
64 bytes from 192.168.1.100: icmp_seq=0 ttl%5 time=0.5 ms
64 bytes from 192.168.1.100: icmp_seq=1 ttl%5 time=0.5 ms
64 bytes from 192.168.1.100: icmp_seq=2 ttl%5 time=0.5 ms

Since the first gw is marked as dead, all the traffic was routed out via eth2 to my secondary gateway "192.168.3.254". Load balancing was not happening.

Why is the first default nexthop marked dead even though it is reachable? This has happened to one other person according the LARTC archives, but no solution was discussed.

So i tried a hack, i specified the same gateway twice and it started to work! :)

# ip ro add default nexthop via 192.168.1.100 dev eth0 \
					nexthop via 192.168.1.100 dev eth0 \
					nexthop via 192.168.3.254 dev eth2
godzilla:~# ip ro ls
192.168.3.0/24 dev eth2  proto kernel  scope link  src 192.168.3.24 
192.168.2.0/24 dev eth1  proto kernel  scope link  src 192.168.2.254 
192.168.1.0/24 dev eth0  proto kernel  scope link  src 192.168.1.24 
default 
		nexthop via 192.168.1.100  dev eth0 weight 1 dead
		nexthop via 192.168.1.100  dev eth0 weight 1 ^^^^
		nexthop via 192.168.3.254  dev eth2 weight 1
# traceroute cnn.com 
traceroute: Warning: cnn.com has multiple addresses; using 207.25.71.28
traceroute to cnn.com (207.25.71.28), 30 hops max, 38 byte packets
 1  192.168.3.254 (192.168.3.254)  0.776 ms  0.657 ms  0.415 ms
 [..]
# traceroute in.yahoo.com
traceroute to vip2.in.yahoo.com (203.199.70.100), 30 hops max, 38 byte
packets
 1  192.168.1.100 (192.168.1.100)  0.858 ms  0.505 ms  0.380 ms
 [..]

Could anyone explain why the first nexthop is marked dead?

Thank you for your time.

-- Shanu
_______________________________________________
LARTC mailing list / [email protected]
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://ds9a.nl/lartc/

#1 lore raw

This bug is fixed, the patch and explanations from Bert can be found here:
http://marc.theaimsgroup.com/?l=lartc&m0885677229167&w=2

Best regards,
Pierrick.

Quoted 71 lines · Shanker Balan

Hello:

I have been trying to use multiple default gateways over the last 2 days on my Debian box and have been fairly successful in understanding the fundas. Bert, thanx a lot for all those wonderful documentation at
http://www.ds9a.nl/2.4Routing/.

Something strange is happening when I try adding multiple default gateways:

# ip ro add default nexthop via 192.168.1.100 dev eth0 \
					nexthop via 192.168.3.254 dev eth2
# ip ro ls
192.168.3.0/24 dev eth2  proto kernel  scope link  src 192.168.3.24 
192.168.2.0/24 dev eth1  proto kernel  scope link  src 192.168.2.254 
192.168.1.0/24 dev eth0  proto kernel  scope link  src 192.168.1.24 
default 
		nexthop via 192.168.1.100  dev eth0 weight 1 dead
		nexthop via 192.168.3.254  dev eth2 weight 1 ^^^^

The first nexthop is _always_ marked dead for some reason though my primary gateway "192.168.1.100" _is_ reachable via eth0.

# ping 192.168.1.100 -c 3
PING 192.168.1.100 (192.168.1.100): 56 data bytes
64 bytes from 192.168.1.100: icmp_seq=0 ttl%5 time=0.5 ms
64 bytes from 192.168.1.100: icmp_seq=1 ttl%5 time=0.5 ms
64 bytes from 192.168.1.100: icmp_seq=2 ttl%5 time=0.5 ms

Since the first gw is marked as dead, all the traffic was routed out via eth2 to my secondary gateway "192.168.3.254". Load balancing was not happening.

Why is the first default nexthop marked dead even though it is reachable? This has happened to one other person according the LARTC archives, but no solution was discussed.

So i tried a hack, i specified the same gateway twice and it started to work! :)

# ip ro add default nexthop via 192.168.1.100 dev eth0 \
					nexthop via 192.168.1.100 dev eth0 \
					nexthop via 192.168.3.254 dev eth2
godzilla:~# ip ro ls
192.168.3.0/24 dev eth2  proto kernel  scope link  src 192.168.3.24 
192.168.2.0/24 dev eth1  proto kernel  scope link  src 192.168.2.254 
192.168.1.0/24 dev eth0  proto kernel  scope link  src 192.168.1.24 
default 
		nexthop via 192.168.1.100  dev eth0 weight 1 dead
		nexthop via 192.168.1.100  dev eth0 weight 1 ^^^^
		nexthop via 192.168.3.254  dev eth2 weight 1
# traceroute cnn.com 
traceroute: Warning: cnn.com has multiple addresses; using 207.25.71.28
traceroute to cnn.com (207.25.71.28), 30 hops max, 38 byte packets
 1  192.168.3.254 (192.168.3.254)  0.776 ms  0.657 ms  0.415 ms
 [..]
# traceroute in.yahoo.com
traceroute to vip2.in.yahoo.com (203.199.70.100), 30 hops max, 38 byte
packets
 1  192.168.1.100 (192.168.1.100)  0.858 ms  0.505 ms  0.380 ms
 [..]
Quoted 9 lines

Could anyone explain why the first nexthop is marked dead?

Thank you for your time.

-- Shanu
_______________________________________________
LARTC mailing list / [email protected]
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://ds9a.nl/lartc/

_______________________________________________
LARTC mailing list / [email protected]
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://ds9a.nl/lartc/