14-5
Catalyst 2950 Desktop Switch Software Configuration Guide
78-11380-03
Chapter 14 Troubleshooting
Recovery Procedures
Recovery Procedures
These recovery procedures require that you have physical access to the switch:
• Recovering from lost member connectivity
• Recovering from a command-switch failure
• Recovering from a lost or forgotten password
• Recovering from corrupted software
Recovering from Lost Member Connectivity
Some configurations can prevent the command switch from maintaining contact with member switches.
If you are unable to maintain management contact with a member, and the member switch is forwarding
packets normally, check for these port-configuration conflicts:
• Member switches cannot connect to the command switch through a port that is defined as a network
port.
• Member switches must connect to the command switch through a port that belongs to the same
management VLAN. For more information, see the “Management VLAN” sectiononpage5-15.
• Member switches connected to the command switch through a secure port can lose connectivity if
the port is disabled due to a security violation. Secure ports are described in the “Enabling Port
Security” sectiononpage10-7.
Recovering from a Command Switch Failure
You can prepare for a command switch failure by assigning an IP address to a member switch that is
command-capable, making a note of the command-switch password, and cabling your cluster to provide
redundant connectivity between all member switches and the replacement command switch. Hot
Standby Router Protocol (HSRP) is the preferred method for providing a redundant command switch to
a cluster. For more information, see the “HSRP and Standby Command Switches” sectiononpage5-10
and the “Creating a Cluster Standby Group” sectiononpage5-20. For a list of command-capable
Catalyst desktop switches, refer to the Release Notes for the Catalyst 2950 Switch on Cisco.com.
If you have not configured a standby command switch, and your command switch loses power or fails
in some other way, management contact with the member switches is lost, and a new command switch
must be installed. However, connectivity between switches that are still connected is not affected, and
the member switches forward packets as usual. You can manage the members as standalone switches
through the console port or, if they have IP addresses, through CMS Device Manager.
These sections describe how to recover if a standby command switch was not available when the
command switch failed:
• “Replacing a Failed Command Switch with a Cluster Member” sectiononpage14-6
• “Replacing a Failed Command Switch with Another Switch” sectiononpage14-7
• “Recovering from a Failed Command Switch Without HSRP” sectiononpage14-8
To Next Page
Loading...
