Cisco Cisco Prime Central 1.1 Guia Da Instalação

Disable SELinux:

# vi /etc/selinux/config

The following troubleshooting steps help solve common problems in an HA configuration.

The HA installation fails.

Check the log files to locate the problem and take the appropriate action. Log files contain detailed information about 

request processing and exceptions and are your best diagnostic tool for troubleshooting. See “Troubleshooting the Installation” 
in the 

.

Prime Central does not start in a clustered setup.

Check the /var/log/messages files for failure to either mount the shared storage or add the virtual IP address. If the 

shared storage failed to mount, shut down the cluster and verify that you can manually add the shared storage to a node. (Be 
sure to unmount it after your test.) 

If the virtual IP address was not added, verify that it is in the same subnet as the nodes and is not in use by any other computer 
in the network.

If you find that /usr/local/bin/pc.sh start failed, check /usr/local/bin/pc.log and /usr/local/bin/pc-start.log, which will tell you if 
the database or other Prime Central components failed to start. Then, to determine which component failed to start:

Stop the luci, ricci, rgmanager, and cman services on both nodes to shut down the cluster.

On the node where you originally installed Prime Central:

Mount the shared storage.

Add the virtual IP address.

Verify that all services have stopped:

Enter:

# su – primeusr

# emdbctl –start

# itgctl start

# portalctl start

Check the output from each of the preceding commands to locate the problem.

You receive the error “<err> 'fsck -p /dev/mapper/mpath2p1' failed, error=4; check /tmp/fs-vmpcfs.fsck.log.mq4986 

for errors.”

Enter the following command and reboot when it is finished running:

# fsck -f /dev/mapper/mpath2p1

You receive the error “Timeout exceeded while waiting for ‘/images/fm_status.sh’” in /var/log/messages.

Verify that you can use SSH to connect to each node and virtual machine without an authentication or password 

prompt. If SSH prompts for authentication or a password, the Prime Central and Fault Management services cannot start.

Your environment uses the wrong fencing device.

The examples in this guide use fence_manual and fence_virsh, which are test fencing devices and cannot be used for 

production. For information about which fencing device to use in your environment, see the 

.