CCNP TSHOOT 6.0 - The Cisco Learning Network

CCNPv6 TSHOOT
Troubleshooting First Hop Redundancy Protocols
The figure illustrates an example of a method that you could follow to diagnose and resolve problems related to
FHRPs, such as HSRP, VRRP, and GLBP.
Sample First Hop Redundancy
Troubleshooting Flow
Determine if
outages are
caused by
FHRP or
other
protocols
If failover testing is
possible, ping
virtual and real IP
addresses during
failover
Determine if
FHRP
addresses
are used by
hosts
Investigate
FHRP
router-toroutercommunication
Investigate
FHRP role
selection
© 2009 Cisco Systems, Inc. All rights reserved. TSHOOT v1.0—39
The most common reason to start troubleshooting FHRP behavior is because during an outage or a test, network
connectivity is lost for longer than expected when a redundant device or link is temporarily disabled. In
redundantly configured IP networks, a number of different protocols usually need to reconverge to recover from a
failure. The FHRP that is used is just one of the protocols that could be the cause of the loss of connectivity.
Other protocols that need to converge as well—and could be the cause of the problem—are routing protocols and
Spanning Tree Protocol (STP).
So how do you determine if the FHRP is the problem?
If you have the opportunity to execute failover tests (for instance, during a scheduled maintenance window), a
good way to determine if the problem is caused by the FHRP or by another protocol is by sending multiple
continuous pings from a client that is using the virtual router as its default gateway. Ping to the virtual and real IP
addresses of the routers that participate in the FHRP, and ping to an IP address of a host that is one or more
router hops removed from the client. Observe and compare the behavior of the pings while you force a failover by
disabling a device or a link.
Based on the observed differences between the ping responses, you can draw conclusions about the likelihood
that the problem is related to the FHRP or to any other protocols that are involved in the convergence. Here are a
few examples:
• If you observe that the pings to the real IP address of the redundant router and the virtual IP address
of the FHRP both fail at the same time and resume at the same time when you disable the primary
router, assume that the problem is not related to the FHRP (because the FHRP does not affect the
pings to the real IP address). The most likely cause in this scenario is the Layer 2 convergence for
the VLAN, so you should start a Layer 2 troubleshooting procedure.
• If you observe that the pings to the real IP address of the redundant router do not suffer any packet
loss, but pings to the virtual IP address fail, this strongly suggests that there is a problem with the
FHRP.
All contents are Copyright © 1992–2010 Cisco Systems, Inc. All rights reserved. This document is Cisco Public Information. Page 20 of 26