Testing server, information processing system, and testing method

• US 9,026,858 B2
• Filed: 06/16/2014
• Issued: 05/05/2015
• Est. Priority Date: 12/21/2011
• Status: Expired due to Fees

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Rikitake, Susumu, Shimada, Ikuo
• Primary Examiner(s): KO, CHAE M
• Application Number:
  US14/305,036
• Publication Number:
  US 20140298082A1
• Time in Patent Office: 323 Days
• Field of Search: 714/41
• US Class Current: 714/41
• CPC Class Codes: G06F 11/2007 using redundant communication media
• CPC Class Codes: G06F 11/2215 to test error correction or detection circuits

Abstract

A testing server performs a test to check whether servers properly execute failover. The testing server includes a generation unit, a transmitting unit, a testing unit, a restoring unit, a judgment unit, and a power control unit. The generation unit generates an image file of an OS. The transmitting unit transmits the image file to to-be-tested servers. The testing unit injects a simulated fault into a server among the servers to which the image file is transmitted and performs a test. The restoring unit, each time the testing unit performs a test, restores a status of the to-be-tested server to a pre-failover status. The judgment unit judges whether the restoring unit properly restores the status. The power control unit, when the judgment unit judges that the status of the to-be-tested server is not properly restored, turns off power of the to-be-tested server and turns on the power again.

Claims

1. A testing server that performs a test to check whether a plurality of servers properly executes failover, the testing server comprising:

a generation unit that generates an image file of an OS (Operation System) that is executed by a to-be-tested server;

a transmitting unit that transmits the image file that is generated by the generation unit to to-be-tested servers;

a testing unit that injects a simulated fault into a server where a fault occurs among the servers to which the transmitting unit transmits the image file and that performs a test to check whether the to-be-tested server properly executes failover;

a restoring unit that, each time the testing unit performs a test, restores a status of the to-be-tested server to a pre-failover status;

a judgment unit that judges whether the restoring unit properly restores the status of the to-be-tested server; and

a power control unit that, when the judgment unit judges that the status of the to-be-tested server is not properly restored, turns off power of the to-be-tested server and then turns on the power again, wherein

with respect to the server whose power is turned on again by the power control unit, the transmitting unit transmits an image file and the testing unit performs a test in a repetitive manner.

2. The testing server according to claim 1, comprising:

a collection unit that collects information that indicates details of a device that is included in each server; and

a selection unit that selects a to-be-tested server in accordance with the information collected by the collection unit, wherein

the transmitting unit transmits the image file to the server that is selected as a subject for a test by the selection unit.

3. The testing server according to claim 2, comprising:

a receiving unit that receives, from a user, designations with respect to a type of simulated fault to be injected into the server and a type of failover to be executed by the server, wherein

the selection unit uses the information that is collected by the collection unit to select a plurality of servers that includes a server into which the type of simulated fault received by the receiving unit is injectable and that is capable of executing the type of failover received by the receiving unit.

4. The testing server according to claim 3, comprising a determination unit that divides the plurality of servers selected by the selection unit by multiple groups and that determines whether the type of failover received by the receiving unit is executable in the groups in parallel, wherein

when the determination unit determines that the type of failover received by the receiving unit is executable in the groups in parallel, the testing unit injects the type of simulated fault received by the receiving unit into each of the groups and performs a test to check whether failover is properly executed.

5. The testing server according to claim 4, wherein the testing unit determines an order in which the servers efficiently execute all types of failover received by the receiving unit, the servers being selected by the selection unit in accordance with the information that is collected by the collection unit and that indicates details of a device included in each of the servers, and injects a simulated fault into the server where a fault occurs such that failover is executed in the determined order.

6. The testing server according to claim 1, comprising:

a monitoring unit that monitors a status of a server to which the transmitting unit transmits an image file; and

an identifying unit that uses the status of each server that is monitored by the monitoring unit to identify a timing in which a simulated fault is injected into the server, wherein

the testing unit injects a simulated fault into the server at the timing identified by the identifying unit and performs a test to check whether failover is properly executed.

7. The testing server according to claim 1, wherein the transmitting unit transmits the image file generated by the generation unit to a RAM (Random Access Memory) that is included in the server.

8. An information processing system that includes a plurality of servers that executes failover when a fault occurs in any of the servers and a testing server that performs a test to check whether each of the servers properly executes failover, wherein

the testing server includes

a generation unit that generates an image file of an OS that is executed by a to-be-tested server;

a transmitting unit that transmits the image file that is generated by the generation unit to to-be-tested servers;

a testing unit that injects a simulated fault into a server where a fault occurs among the servers to which the transmitting unit transmits the image file and that performs a test to check whether the to-be-tested server properly executes failover;

a restoring unit that, each time the testing unit performs a test, restores a status of the to-be-tested server to a pre-failover status;

a judgment unit that judges whether the restoring unit properly restores the status of the to-be-tested server; and

a power control unit that, when the judgment unit judges that the status of the to-be-tested server is not properly restored, turns off power of the to-be-tested server and then turns on the power again,

with respect to the server whose power is turned on again by the power control unit, the transmitting unit transmits an image file and the testing unit performs a test in a repetitive manner, and

the server performs network booting with the image file.

9. A testing method that is performed by a testing server that performs a test to check whether a plurality of servers properly executes failover, the servers executing failover when a fault occurs in any of the servers, the testing method comprising:

generating an image file of an OS that is executed by a to-be-tested server;

transmitting the generated image file to to-be-tested servers;

injecting a simulated fault into a server where a fault occurs among the servers to which the image file is transmitted and performing a test to check whether the to-be-tested server properly executes failover;

each time a test is performed, restoring a status of the to-be-tested server to a pre-failover status;

judging whether the status of the to-be-tested server is properly restored;

when it is judged that the status of the to-be-tested server is not properly restored, turning off power of the to-be-tested server and then turning on the power again; and

with respect to the server whose power is turned on again, transmitting the image file and performing the test in a repetitive manner.