Various methods and apparatuses to restart a server

• US 9,183,068 B1
• Filed: 11/18/2005
• Issued: 11/10/2015
• Est. Priority Date: 11/18/2005
• Status: Active Grant

• Current Assignee: Sun Microsystems Incorporated (Oracle Corporation)
• Original Assignee: Oracle America, Inc. (Oracle Corporation)
• Inventors: Shum, Han, Newman, Chris
• Primary Examiner(s): Bui, Jonathan
• Application Number:
  US11/282,256
• Time in Patent Office: 3,644 Days
• Field of Search: 709223-224
• US Class Current: 1/1
• CPC Class Codes: G06F 11/0703 Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
• CPC Class Codes: G06F 11/0715 in a system implementing multitasking multitasking per se G06F9/46
• CPC Class Codes: G06F 11/0757 by exceeding a time limit, i.e. time-out, e.g. watchdogs
• CPC Class Codes: G06F 11/0793 Remedial or corrective actions recovery from an exception in an instruction pipeline G06F9/3861; by retry G06F11/1402; for recovering from a failure of a protocol instance or entity H04L69/40
• CPC Class Codes: G06F 11/14 Error detection or correction of the data by redundancy in operation G06F11/16 takes precedence
• CPC Class Codes: G06F 11/3003 Monitoring arrangements specially adapted to the computing system or computing system component being monitored
• CPC Class Codes: G06F 11/3055 Monitoring arrangements for monitoring the status of the computing system or of the computing system component, e.g. monitoring if the computing system is on, off, available, not available error or fault processing without redundancy G06F11/0703; error detection or correction by redundancy in data representation G06F11/08; error detection or correction by redundancy in operation G06F11/14; error detection or correction by redundancy in hardware G06F11/16
• CPC Class Codes: G06F 2201/81 Threshold
• CPC Class Codes: G06F 2201/865 Monitoring of software

Abstract

Various methods and apparatuses are described for a server that includes at least a server starter program, a monitor program, and a service program. The server starter program runs resident on a computer readable storage medium. The server starter program contains code scripted to start and stop server applications. The monitor program also runs resident on the computer readable storage medium. The monitor program contains code scripted to detect a failure associated with a first server application and to communicate to the server starter program to re-start one or more server applications based upon the detected failure associated with the first server application. The service program registers as an operating system service with the operating system. The service program controls requests and communicates to the operating system on behalf of the first server application in the server.

Claims

1. A non-transitory computer readable storage medium having a computer readable program embodied therein, said computer readable program for execution on a server, the non-transitory computer readable storage medium comprising:

service program code to run resident on the server and register as a service of an operating system for the server, the service program code including instructions to control requests, to communicate with the operating system on behalf of a plurality of server applications, and to invoke a server starter program code to start particular server applications of said plurality of server applications;

the server starter program code to run resident on the server with root-level access privileges, the server starter program code including instructions scripted to start and stop the particular server applications of the plurality of server applications in a logical order according to dependencies of the particular server applications and to initiate a monitor program code before starting any of said particular server applications;

a probing program code to run resident on the server without the root-level access privileges, the probing program code containing instructions scripted to detect a failure associated with a first server application of the plurality of server applications and communicate the failure to the monitor program code; and

the monitor program code to run resident on the server with the root-level access privileges, the monitor program code including instructions scripted to communicate with the server starter program code to re-start one or more of the particular server applications based upon the detected failure associated with the first server application.

2. The non-transitory computer readable storage medium of claim 1, wherein the probing program code contains instructions scripted to connect, via a respective protocol used by a respective one of the server applications, to each server application, and then communicate via the respective protocol used by that server application to determine if that server application is functioning correctly.

3. The non-transitory computer readable storage medium of claim 2, wherein the server starter program code, the monitor program code, and the probing program code are scripted as three separate program codes.

4. The non-transitory computer readable storage medium of claim 1, wherein the probing program code detects the failure associated with the first server application at least in part by detecting a corrupted file associated with a database.

5. The non-transitory computer readable storage medium of claim 1, wherein a first one of the server applications connects to the monitor program code over an internal socket, and the monitor program code calls the server starter program code to restart one or more of the server applications on either a detection of a failure with the internal socket for the first server application or a notification of another failure associated with the first server application.

6. The non-transitory computer readable storage medium of claim 1, wherein the monitor program code contains instructions scripted to have instruction calls to use a shared library for Operating System system-calls but no other calls to use the shared library.

7. The non-transitory computer readable storage medium of claim 1, wherein the service program code contains instructions scripted to register as a Windows service with the operating system to control requests and to communicate to a service control manager of the operating system on behalf of all of the server applications.

8. The non-transitory computer readable storage medium of claim 1, wherein the server starter program code and the monitor program code cooperate together to cumulatively contain instructions scripted to detect failures with one or more server applications in a single node and then re-start Previously Presented instances of a failed server application based upon a detected failure of the failed server application.

9. The non-transitory computer readable storage medium of claim 1, wherein the server starter program code contains instructions scripted to re-start merely one of the server applications based upon a detected failure associated with the one server application.

19. The non-transitory computer readable storage medium of claim 1, wherein the probing program code detects the failure associated with the first server application at least in part by detecting a particular amount of remaining usable disk space.

20. The non-transitory computer readable storage medium of claim 1, wherein the probing program code detects the failure associated with the first server application at least in part by detecting a particular amount of remaining usable disk space.

10. A system, comprising:

a messaging server that includes a non-transitory computer readable storage medium having a computer readable program embodied therein for execution on the messaging server, the non-transitory computer readable storage medium having;

a message store program code to store, retrieve and manipulate messages;

a service program code to run resident on the messaging server and register as an operating system service for the messaging server, the service program code including instructions to control requests, to communicate with the operating system on behalf of one or more server applications, and to invoke a server starter program code to start others of said server applications;

the server starter program code to run resident on the messaging server, the server starter program code including instructions scripted to start and stop the server applications in a logical order according to dependencies of the server applications and to initiate a monitor program code before starting any others of said server applications;

the monitor program code to run resident on the messaging server, the monitor program code including instructions scripted to detect a failure associated with a first of the server applications and to communicate with the server starter program code to re-start one or more of the server applications based upon the detected failure associated with the first server application; and

a probing program code to run resident on the messaging server, wherein the probing program code contains instructions scripted to detect a failure on a database associated with the message store program code based on remaining usable disk space and then connect to the monitor program code to communicate to the monitor program code to initiate a restart communication to the server starter program code, wherein the probing program code contains instructions scripted to connect via a protocol used by the first server application to the first server application and communicates via the protocol used by the first server application to determine if the first server application is functioning correctly, and wherein the probing program determines if the first server application is functioning correctly by measuring a response time of the first server application.

11. The system of claim 10, wherein the monitor program code contains instructions scripted to give the monitor program code permanent root level access to files on the messaging server.

12. The system of claim 10, wherein the probing program code contains instructions to run resident when prompted by a scheduler program code on the messaging server.

18. The system of claim 10, wherein in response to the probing program code ascertaining that the measured response time of the first server application exceeds a warning threshold or a specified timeout period:

an alarm is sent;

or

the first server application is restarted.

13. A method, comprising:

registering a service program as an operating system service with an operating system of a server;

invoking a single instance of a server starter program code to start other server applications in the server;

inhibiting invocation of additional instances of the server starter program code while the single instance of the server starter program code runs;

initiating a monitor program before starting any of the server applications and then starting others of the server applications in a logical order according to dependencies of the server applications;

upon detecting a failure of a first one of the server applications, communicating to the server starter program code to re-start merely the first server application; and

communicating to the operating system on behalf of all of the server applications in the server.

14. The method of claim 13, further comprising:

connecting to the first server application via a specific protocol used by the first server application; and

initiating two-way communications with the first server application via the specific protocol used by the first server application to determine if the first server application is functioning correctly.

15. A non-transitory computer readable storage medium having a computer readable program embodied therein, said computer readable program for execution on a server computer, the non-transitory computer readable storage medium comprising:

a service program code to run resident on a server computer and register as an operating system service with an operating system of the server computer, the service program code including instructions to control requests, to communicate with the operating system on behalf of one or more server applications;

an application program code for invoking a server starter program code to start other server applications in the server computer;

the server starter program code to run resident on the server computer, the server starter program code including instructions scripted to start and stop the server applications in a logical order according to dependencies of the server applications and to initiate a monitoring program code before starting any others of said server applications;

the monitoring program code to run resident on the server computer for communicating to the server starter program code to re-start merely a first one of the server applications based upon a detected failure associated with the first server application; and

a communication program code for communicating to the operating system on behalf of all of the server applications in the server computer,

wherein the server starter program code and the monitor program code cooperate together to cumulatively contain instructions scripted to detect failures with a plurality of server applications in a single node and then restart Previously Presented instances of a failed plurality of server applications based upon detected failures of the plurality of failed server applications, wherein the monitoring program code includes instructions to determine a total number of failures of each of the server applications within a specified time period.

16. The non-transitory computer readable storage medium of claim 15, further comprising:

a first probing program code to run resident on the server computer for connecting to the first server application via a specific protocol used by the first server application; and

a second probing program code for initiating two-way communications with the first server application via the specific protocol used by the first server application to determine if the first server application is functioning correctly.

17. The non-transitory computer readable storage medium of claim 15, wherein the monitoring program code includes instructions to:

determine, for each server application whether the total number of failures is greater than a threshold number of failures; and

inhibit communications to the server starter program code to restart a particular one of the server applications when it is determined that its total number of failures is greater than the threshold number of failures.