Cisco Cisco TelePresence Management Suite (TMS) Version 15 Installation Guide
The probes on the NLB will now pick up that the upgraded server is the active node. Users and managed systems can
now use Cisco TMS again.
now use Cisco TMS again.
1.
Upgrade the second Cisco TMS node.
2.
The installer will detect that the database has already been upgraded, and offer to continue using the updated
database: select Yes. The installer will then update the binaries and leave the database alone.
database: select Yes. The installer will then update the binaries and leave the database alone.
3.
Log in to the Cisco TMS web application on the active server, and force a manual failover. Verify that the second
server becomes active within one minute.
server becomes active within one minute.
4.
Enable file replication.
Check that the network settings have not been changed during the install process:
1.
In the Cisco TMS application go to Administrative Tools > Configuration > Network Settings.
2.
Check that the following IP address(es) and hostnames are set to the NLB's virtual IP address(es) and hostname:
—
Event Notification > SNMP Traphost IP Address
—
Advanced Network Settings for Systems on Internal LAN: all fields
—
Advanced Network Settings for Systems on Public Internet/Behind Firewall
Recovery in Case of a Failing Node
No immediate action is required in the event of a server failure. If the passive node goes down, the active node
continues its operations as normal. If the active node goes down, the passive node will become active automatically, and
the NLB will detect the failure and direct all traffic to the new active node.
continues its operations as normal. If the active node goes down, the passive node will become active automatically, and
the NLB will detect the failure and direct all traffic to the new active node.
Expect a delay of approximately one minute while the failover takes place. During this time the 'Unable to connect to
Cisco TMS' error will be shown on the web page.
Cisco TMS' error will be shown on the web page.
Troubleshoot the failing node’s software and hardware as you normally would, and bring it back online once it is
operational.
operational.
Troubleshooting with Managed Systems in a Redundant Deployment
Capturing a Wireshark trace on the active Cisco TMS server will show all incoming traffic as coming from the NLB. You
will therefore be unable to easily identify the communication between Cisco TMS and the managed systems you are
interested in.
will therefore be unable to easily identify the communication between Cisco TMS and the managed systems you are
interested in.
Before you begin troubleshooting, do the following:
1.
For all the systems you are investigating, temporarily set the management address to the address of the active
Cisco TMS server.
Cisco TMS server.
2.
Take the Wireshark capture.
3.
Once you have finished taking the trace, set the management address of all the systems back to the address of
the NLB.
the NLB.
Logs
■
Log files must be gathered from both Cisco TMS servers.
■
If changing the log levels, ensure that this is done on both servers. The exception to this is if you are changing the
log level for a very short time, for example if you increase the log level, reproduce the problem and then
immediately decrease the log level again.
log level for a very short time, for example if you increase the log level, reproduce the problem and then
immediately decrease the log level again.
Deploying a Hot Standby
Keeping an additional Cisco TMS server as a warm spare in case of failure is known as the "Hot Standby" redundancy
model.This requires manual intervention if there is a failure on the primary Cisco TMS server, and is therefore a
model.This requires manual intervention if there is a failure on the primary Cisco TMS server, and is therefore a
38
Cisco TelePresence Management Suite Installation and Upgrade Guide
Setting up a Redundant Deployment