Container system and monitoring method for container system

1. A container system, comprising:

a plurality of servers, arranged in N areas, where N is a positive integer; and

a master server node, coupled to the servers, wherein the master server node selects one of the servers in an i th area to be a main node of the i th area, and the main node collects a plurality of pieces of temperature information of the servers in the i th area and transmits a piece of highest temperature information to the master server node, where i is a positive integer and 1≦i≦N,

wherein when the master server node fails to connect to the main node, the master server node selects one of other servers connected to the master server node in the i th area to be a replacing node, and designates the replacing node to be the main node of the i th area to collect the temperature information in the i th area continuously.

2. The container system according to claim 1 , wherein the master server node is coupled to the servers through an Ethernet and a plurality of switches.

3. The container system according to claim 1 , further comprising:

N heat dissipation devices, correspondingly arranged in the N areas, respectively; and

a heat dissipation controller, coupled to the master server node and the heat dissipation devices, wherein the master server node transfers a heat dissipation control request to the heat dissipation controller according to the respective highest temperature information of the N areas, so as to adjust heat dissipation efficiencies of the N heat dissipation devices corresponding to the N areas.

4. The container system according to claim 1 , wherein the master server node reads a server address mapping table to resolve the servers arranged in the i th area, sends a test request to the servers in the i th area in sequence to find servers capable of serving as the main node, and records the servers capable of serving as the main node in the server address mapping table.

5. The container system according to claim 4 , wherein when the master server node fails to connect to the main node, the master server node reads the server address mapping table, sends the test request to the servers capable of serving as the main node in sequence, and sets one of servers returning a test response to be the replacing node.

6. The container system according to claim 1 , wherein the master server node sends a report request to the main node of each area in sequence, so that the main node collects a plurality of pieces of temperature information of the servers in the i th area and transmits the highest temperature information, and the master server node judges the main node as failing to be connected when the report request is sent to the main node, but the number of times that the highest temperature information returned by the main node is not received exceeds a disconnection preset value.

7. The container system according to claim 1 , wherein the master server node sends a report request to the main node of each area in sequence, so that the main node collects a plurality of pieces of temperature information of the servers in the i th area and transmits the highest temperature information, and the master server node judges the main node as failing to be connected when the report request is sent to the main node, but the time period that the highest temperature information returned by the main node is not received exceeds a disconnection preset period.

8. A monitoring method for a container system, wherein the container system comprises a plurality of servers arranged in N areas, where N is a positive integer, the monitoring method comprising:

resolving and selecting one of the servers in an i th area to be a main node of the i th area;

sending a report request to the main node of each area in sequence, so that the main node collects a plurality of pieces of temperature information of the servers in the i th area and transmits a piece of highest temperature information; and

when the main node fails to be connected, selecting one of other connected servers in the i th area to be a replacing node, and adjusting the replacing node to be the main node of the i th area to collect the temperature information in the i th area continuously.

9. The monitoring method according to claim 8 , wherein the step of resolving and selecting one of the servers in the i th area to be the main node of the i th area comprises:

reading a server address mapping table to resolve the servers arranged in the i th area; and

sending a test request to the servers in the i th area in sequence to find servers capable of serving as the main node, and recording the servers capable of serving as the main node in the server address mapping table.

10. The monitoring method according to claim 8 , wherein the step of resolving and selecting one of the servers in the i th area to be the main node of the i th area further comprises:

setting one of servers returning a test response to be the main node.

11. The monitoring method according to claim 9 , wherein the step of selecting one of the other connected servers in the i th area to be the replacing node comprises:

reading the server address mapping table, and sending the test request to the servers capable of serving as the main node in sequence; and

setting one of servers returning a test response to be the replacing node.

12. The monitoring method according to claim 8 , wherein the step of judging that the main node fails to be connected comprises:

judging the main node as failing to be connected when the report request is sent to the main node, but the number of times that the highest temperature information returned by the main node is not received exceeds a disconnection preset value.

13. The monitoring method according to claim 8 , wherein the step of judging that the main node fails to be connected comprises:

judging the main node as failing to be connected when the report request is sent to the main node, but the time period that the highest temperature information returned by the main node is not received exceeds a disconnection preset period.