Help:Why I can ping host1 and but not host2;same network.

[news2010a]

My client (network 192.168.1.0/24)----------[RouterA]-----Layer 3 switch----2950 switch (host1,host2)

from my client, I can ping host1. Host1 IP=11.2.53.104/26

from my client, ping to Host2 IP=11.2.53.105/26 times out.

From Layer3Switch and 2950 I can ping Host2 just fine.

I did trace from client and the traffic to Host2 dies on RouterA.

On RouterA, I confirm that I can ping Host1 OK and Host2 times out.

I examine access-lists existing there, but I can't see any that would allow 11.25.3.104/26

and it would block 11.2.53.105/26.

Question:

Any ideas on how I troubleshoot this further?

[Lucien Avramov]

debug ip icmp on the router A

Try to ping host2 and see if the icmp is sent.

Do also a show ip route 11.2.53.105 on your router A and post the output.

It sounds like either an ACL issue, or Router A does not have a route to host 2.

[news2010a]

On RouterA, I did it and route is present for host2. Same route is shown for host1.

Then using the below, I used cdp neig detail and got to a switchX. From switchX, ping to host2 also times out.

Trace from switchX goes to Layer3switch via trunk. I can't see anything wrong on the trunk. From Layer3Switch I can ping host2 as informed earlier.

RouterA#show ip route 11.2.53.105

Routing entry for 11.2.53.96/27 (before I said it was /26, but it is /27. Host is mask is /27 OK).

Known via "ospf 1", distance 110, metric 2020, type intra area

Last update from 10.2.55.233 on FastEthernet0/0.32, 1d00h ago

Routing Descriptor Blocks:

* 11.2.55.234, from 11.2.255.1, 1d00h ago, via FastEthernet0/0.32

Route metric is 2020, traffic share count is 1

11.2.55.233, from 11.2.255.1, 1d00h ago, via FastEthernet0/0.32

Route metric is 2020, traffic share count is 1

11.2.53.13, from 11.2.255.1, 1d00h ago, via FastEthernet4/0.19

Route metric is 2020, traffic share count is 1

11.2.55.226, from 11.2.255.1, 1d00h ago, via FastEthernet1/1.31

Route metric is 2020, traffic share count is 1

[Lucien Avramov]

from l3 switch, try to ping host 2 with a source interface: the one that connects to router A.

If that ping fails, host 2 doesnt have a route to router A subnet.

[news2010a]

I did the extended ping using host2 and source address of Layer3Switch which connects to RouterA and that fails. Host1 works.

So it sounds this would be a host configuration problem?

I already checked on the 2950switch the host is connected to and the ip-default gateway is correct.

[Lucien Avramov]

okay that is what I thought, that means host2 does not have a route to the subnet that is between layer3switch and router.

You can either configure a default route in host2 or define a default gateway. Most likely if its a linux it knows only about its own subnet by default.

Configure your host properly and then your issue should be fixed.

[iyde]

The ip default-gateway configured on the 2950 is the default gateway for management access to that particular switch. It has nothing to do with the end-client default gateway.

HTH

[PETER EIJSBERG]

Did you check the default gateway on host2?

[fjcardenas-1]

That sound like a default gateway problem in host 2. Make sure Host 2 has the same mask and gateway as host 1.