Server configuration The server editor is used to monitor the system servers status, as well as add new servers to the system (see Hot backup setup), and start/stop the project in RUN-TIME (see Start/Stop the project in RUN-TIME). The editor is located on the “Servers” tab and looks as follows:
Add picture Server editor
The control panel is located at the top of the editor:
Server editor control panel
|Current state of RUN-TIME|
Start the project in RUN-TIME (see Start/Stop the project in RUN-TIME)
Add new server to the system (see Hot backup setup)
|Information about the volume of uncompressed archive for the current hour. Format -
|The name of the node to which the browser is currently connected. It has the following format:
The host name is set during the installation of Faceplate on the server (see Installation process description).
|Current server time|
Below the panel are server cards that reflect the current state of the servers.
Server status monitoring
For each configured server, information about its current status is displayed. Information is displayed on server cards, which are updated every 5 seconds. The server card looks like this:
Name of the node (see Hot backup setup). If the element is green, then the node is connected, active, all necessary information is copied and the node is ready to accept the role of wizard.
|Current server time|
|Server connection status|
|Data copy status|
|The amount of available and used RAM|
|The percentage of CPU utilization. For nodes running Linux OS, the percentage is calculated using the formula:
Where: Load_% is the calculated CPU utilization percentage Load_rup is the average amount of time spent by waiting queue processes. Factor - a correction factor. The time it takes for processes to wait their turn is an indirect measure of the load on the CPU. To convert this parameter to %, you need to consider the power and configuration of the processor. This is done by adjusting the correction factor. More information can be found on the link https://linux.die.net/man/3/cpu_sup.
For nodes running Windows, the correction factor is not involved in calculating the load on the CPU
|The number of active sessions on this node. Corresponds to the number of connections. The runtime environment uses 2 persistent connections and one dynamic connection per report generation. This means that when opening the runtime in the browser, 2 new sessions appear on the node. Another report appears during report generation.
The development environment uses one connection i.e. when a development environment is opened in the browser; one new session is added to the node.
Clicking on the button with the number of sessions will open a window containing information on open sessions:
|Disks status. Info as follows:
|The upper right corner has the button to delete the server. Before deleting, the server must be stopped and disabled.|
The right side of the card displays the trend by which you can track the dynamics of changes in the load on the server. The trend displays the percentage of RAM usage and the percentage of CPU usage. You can view data for:
- 10 minutes
- 3 hours
- 1 day
Hot backup setup
Increased fault tolerance of the system can be achieved by providing hot backup servers. There is one master in the hot standby mode; the other servers operate in stand-by mode. The master performs a survey of controller equipment, provides archiving and operation of the message system. Servers that are in stand-by mode, as well as the master, can be used to connect from the development or runtime environment. When a master-server fails for any reason (equipment failure, power outages, etc.), the first server in standby mode diagnoses the master’s loss and takes over: establishes connections to the controller equipment, starts system processes archiving and messaging system. Client connections are automatically switched to an available server. When the server is restored, the master role does not automatically return to it. If necessary, you can do it explicitly (see Assign new master).
To provide hot backup, the system must consist of at least 2 servers. Physical servers must be located in the same local network and each server must have access to port 4369. In hot backup mode, a large amount of information is transmitted between servers (proportionally to the scale of the project); therefore the communication channel between servers must have high bandwidth (at least 100 Mbps) / sec.)
To add a new server to the system:
Prepare a physical server, install Faceplate on it (see Installation on the backup server).
- Start the Faceplate application on a new server. At the first Faceplate launch, the application waits for the master-server to connect and copy necessary data to it.
Click “Add” button on control panel of the server editor of the current project (see Server editor control panel) and specify the name of the node for the server to be added. Node name has the following format: fp@<имя хоста>. Host name must match with the name defined in the process of installation (see Installation on the backup server).
As a result, a new server card will appear and the master server will start copying the necessary information to the new server. When the copy is completed, the host name of the added server will turn green (see Server status monitoring), which means the server is operating and ready to adopt the role of the master. ATTENTION! All actions can be performed on the working system, without the need to stop RUN-TIME.
If the server is added to the system and the connected status flag in its card remains false, the following checks must be performed:
Make sure that the domain name for the new server is the same as the host name in the server node name. Node name is defined during the Faceplate installation (see Installation on the backup server).
- Make sure that the new server is available from the master-server. In order to do this, you need to use ping command. When pinging, it is necessary to use domain name (host name).
- Make sure that the new server has the open 4369 port. Use the telnet utility for that.
- If the trouble is not resolved, restart the new server and launch the Faceplate again.
Assign new master
The server performing master role provides the RUN-TIME mode operation (see Start/Stop the project in RUN-TIME). The master interrogates with controller equipment, provides archive and message system operation. If the system has configured hot backup, the role of the master may be taken by servers in stand-by mode. The master role transfer between the servers is executed upon the failure of server performing master role, or upon clear assigning of the new master in server editor. To transfer the master role to another server, you need to click the master selection button on the control panel (see Server editor control panel):
A list of servers ready to accept the master role will be displayed. If none of the servers are ready to accept the master role, a message will be displayed:
Start/Stop the project in RUN-TIME
The server in RUN-TIME mode performing the master role contains processes for polling controller equipment (see Connections), maintaining archives (see Archiving system), message system operation (see Message system) and the number of other processes ensuring object management. When exiting the RUN-TIME mode, the listed processes are terminated and the system ceases to perform functions for managing an object — it goes into STOP mode. The project development environment functions independently of the state of the RUN-TIME mode. To run a project in RUN-TIME, click on the “RUN” button on the control panel (see Server editor control panel):
As a result, the processes that control the object will be launched on the master server, after which the button takes the following form:
To exit RUN-TIME mode, click on the STOP button.