Preliminary Information

Before upgrading rvrd on a production server, we recommend upgrading in a test or development environment using the same procedure.

We recommend that you upgrade or reconfigure any rvrd only during scheduled downtime (not during production hours).

When upgrading routing daemons in an enterprise, we recommend upgrading them one at a time. Verify that each upgraded daemon is properly forwarding messages and cooperating with any redundant routing daemons (for fault tolerance). Only then upgrade the next routing daemon.

Backup

Before upgrading any rvrd, make backup copies of all rvrd store files and log files (throughout your network).

Configuration Changes

Determine whether you must modify the routing daemon’s configuration. If so, prepare the new configuration before starting an upgrade. For more information, see Reconfiguring an Upgraded Routing Daemon Reconfiguring an Upgraded Routing Daemon Reconfiguring an Upgraded Routing Daemon.

Message Flow

Upgrading a routing daemon involves stopping the daemon, making changes, and starting the new daemon. While the daemon is stopped, it cannot forward messages. Two scenarios are possible:

•

When redundant routing daemons cooperate for fault-tolerant service, you can upgrade one while the others continue to forward messages. This technique lets you upgrade without disrupting message flow.

•

Otherwise, you might need to stop the flow of messages that require forwarding through the stopped daemon. (see Stopping Messages that Require Routing).

Clients

Routing daemon executables can also serve as ordinary daemons. If any client applications connect specifically to the daemon you are upgrading, you must stop those client processes before upgrading. (You may restart them such that they connect to a different daemon process during the upgrade; but see Message Flow.)

Did you find this helpful?

Yes No