Software-based watchdog method and apparatus

• US 20030221141A1
• Filed: 05/22/2002
• Published: 11/27/2003
• Est. Priority Date: 05/22/2002
• Status: Active Grant

• Current Assignee: American Power Conversion Corporation (Schneider Electric SE)
• Original Assignee: American Power Conversion Corporation (Schneider Electric SE)
• Inventors: Wenisch, Thomas F.
• Granted Patent: US 7,162,714 B2
• Time in Patent Office: Days
• US Class Current: 714/47
• CPC Class Codes: G06F 11/1438 Restarting or rejuvenating

Abstract

In a computer system which allows simultaneous operation of multiple processes, a software watchdog process operates to monitor a primary process through operating system calls. If the response to an operating system call shows that the primary process is not operating or is over utilizing CPU time, then the primary process is restarted. The software watchdog process may also check and correct configuration and data files before restarting the primary process. Alternatively, rather than using operating system calls, the software watchdog process and primary process may communication through a loop back TCP/IP address for monitoring purposes.

Claims

1. A method for monitoring a software process in a computer system having an operating system, the method comprising the steps of:

issuing a operating system call for information regarding processes being executed by the computer system;

determining whether the software process is executing properly based upon the response to the operating system call;

restarting the software process when a determination is made that the software process is not executing properly.

2. The method for monitoring a software process according to claim 1, wherein the determining step includes the steps of:

determining whether the software process is identified in the response to the operating system call as an executing process; and

determining whether the software process has a proper CPU usage based upon the response to the operating system call.

3. The method for monitoring a software process according to claim 1, wherein the restarting step includes the step of terminating operation of the software process if it is not executing properly.

4. The method for monitoring a software process according to claim 1, wherein the restarting step includes the step of correcting information used by the software process prior to restarting the software process.

5. The method for monitoring a software process according to claim 4, wherein the step of correcting information used by the software process includes the step of verifying dependencies of the software process on system resources.

6. The method for monitoring a software process according to claim 1, further comprising the step of providing a user notification when a determination is made that the software process is not executing properly.

7. The method for monitoring a software process according to claim 1, further comprising the step of storing information regarding operation of the software process.

8. A method for monitoring a software process comprising the steps of:

opening a communication channel with the software process;

sending a message to the software process on the communication channel, wherein the software process responds to the message;

receiving a response to the message; and

restarting the software process if a response to the message is not received.

9. The method for monitoring a software process according to claim 8, wherein the restarting step includes the step of correcting information used by the software process prior to restarting the software process.

10. A fault tolerant computer system comprising:

a central processing unit for executing a plurality of processes;

an operating system for controlling the execution of the plurality of processes on the central processing unit and providing information regarding execution of the plurality of processes in response to an operating system call;

a first process to be executed on the central processing unit;

a second process to be executed on the central processing unit, the second process including;

means for issuing an operating system call;

means for determining whether the first process is executing properly based upon a response to the operating system call;

means for restarting the first process when a determination is made that the first process is not executing properly.

11. The fault tolerant computer system according to claim 10, wherein the means for determining includes:

means for determining whether the first process is identified in the response to the operating system call as an executing process; and

means for determining whether the first process has a proper CPU usage based upon the response to the operating system call.

12. The fault tolerant computer system according to claim 10, wherein the second process further includes means for correcting information used by the first process prior to restarting the first process.

13. The fault tolerant computer system according to claim 12, wherein the second process further includes means for checking dependencies of the first process on system resources prior to restarting the first process.

14. The fault tolerant computer system according to claim 10, wherein the second process includes means for storing information regarding operation of the first process.

15. The fault tolerant computer system according to claim 10, wherein the first process includes:

means for issuing an operating system call;

means for determining whether the second process is executing properly based upon a response to the operating system call;

means for restarting the second process when a determination is made that the first process is not executing properly.

16. A fault tolerant computer system comprising:

a central processing unit for executing a plurality of processes;

a communication link allowing communication between two of the plurality of processes executing on the central processing unit;

a first process executing on the central processing unit, the first process including means for responding on the communication link to a message received on the communication link when the first process is executing;

a second process executing on the central processing unit, the second process including;

means for transmitting a message to the first process on the communication link;

means for receiving a response on the communication link from the first process;

means for restarting the first process when a response is not received.