Keyword search: 

FRP Network Protocols and Ports Used

11/26/2009 3:04 PM
You can subscribe to this wiki article using an RSS feed reader.
The following article describes FRP's processes as well as explaining what ports are used to effect operation and replication.

FRP has 3 main processes:
(1) FRP Replication Service (Engine),
(2) FRP Management service and
(3) FRP Helper service. 

The replication service performs the actual data replication.  The management service is the service that configures, monitors and maintains the running replication jobs.  The Helper sercvice ensures there is no collision of instructions comming from multiple sources for instruction sets to the replication servers.
For a comprehensive discussion of network configuration, firewalls and port forwarding see FRP Advance Network Configuration.

FRP Management Service (Console):
The FRP management service provides a web interface over port 9100.  Users can interact with the management service by pointing their browser the the management server at port 9100. All replication servers send execution logs and status messages to the management server using this port.  The repliation servers obtain the most up to date job list and schedule from the management server via this port.  This port must be open for incoming traffic on the management server and open for outgoing traffic on all the replication servers.

FRP Replication Service (Engine):
The FRP Replication service communicates with its peer engine(s) over the Engine standard TCP port 9200 (ENGINE <-> ENGINE). This service also receives control commands from the Management service via its standard TCP port 9200 (MANAGEMENT -> ENGINE).  Port 9200 must be open for incoming traffic on all replication servers. It should be open for outgoing connections on the management server.

The replication service sends execution logs and status messages to the Management service through the Management standard HTTP port 9100. If communications to the Management service over HTTP port 9100 is not enabled, FRP replication will still take place. The consequence is that the Management service will not be able to display job execution logs or the current server status.

In case of conflicts, users may configure the port used by any of the replication servers and change it from 9200 to a user-selected port. This can be done individually for any desired replication server.

Auto-Discovery:

The mechanism called "Auto-Discovery" allows the FRP Management service to automatically discover machines that are running FRP replication services. This feature is based on UDP multi-cast and works only on LANs.
FRP Management sends UDP packets to a configurable multi-cast address/port. Each FRP Engine listens to the UDP multi-cast address/port. Each packet sent contains the Management server host-name and HTTP port through which the Engine can register. If UDP is blocked then the machines can not be auto-discovered and the user needs to manually add the server using the Management Add Server screen.  Auto-discovery is performed over port 9110.

Helper Service:
This service assists in the startup and stopping of the management service and is not required on replication servers.
The three services in Windows are stopped using ports 9300, 9301 and 9311 within the server hosting the management server.

-Updated 11/9/2009
Tags:
Home: WIKI - Knowledge Base Index What's new: Recently changed articles