Detection of an unresponsive application in a high availability system

• US 8,943,191 B2
• Filed: 04/02/2008
• Issued: 01/27/2015
• Est. Priority Date: 04/02/2008
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Coffey, Michael K., Muttur, Manjunath B.
• Primary Examiner(s): Vu, Viet
• Assistant Examiner(s): BELCHER, HERMAN A
• Application Number:
  US12/061,420
• Publication Number:
  US 20090252047A1
• Time in Patent Office: 2,491 Days
• Field of Search: 709/224
• US Class Current: 709/224
• CPC Class Codes: G06F 11/0748 in a remote unit communicating with a single-box computer node experiencing an error/fault remote testing G06F11/2294
• CPC Class Codes: G06F 11/0757 by exceeding a time limit, i.e. time-out, e.g. watchdogs
• CPC Class Codes: G06F 11/076 by exceeding a count or rate limit, e.g. word- or bit count limit
• CPC Class Codes: G06F 11/202 where processing functionality is redundant redundant communication control functionality G06F11/2005, redundant storage control functionality G06F11/2089
• CPC Class Codes: H04L 43/00 Arrangements for monitoring or testing data switching networks
• CPC Class Codes: H04L 43/16 Threshold monitoring
• CPC Class Codes: H04L 43/50 Testing arrangements

Abstract

A method, information processing system, and computer program storage product for determining if an application has become unresponsive. Network traffic at a receiver node is monitored. The receiver node is determined to have generated a window probe response indicating a predetermined window size. A threshold associated with the window size of the receiver node is set in response to the receiver node having generated the window probe response. The window size of the receiver is determined to have been at the predetermined size for at least one of a time period greater than the threshold and a count greater than the threshold. An application associated with the receiver node is classified as unresponsive in response to the window size of the receiver node having been the predetermined size for at least one of a timer period greater than the threshold and a count greater than the threshold.

Claims

1. A method, with an information processing system, for determining if an application has become unresponsive, the method comprising:

monitoring network traffic at a receiver node;

receiving, based on monitoring network traffic at a receiver node, a window probe response comprising a window size, the window size indicating a number of packets a receive buffer at the receiver node is currently able to hold;

comparing the window size of the window probe response to a predetermined window size;

determining, based on the comparing, that the window size of the window probe response matches the predetermined window size;

setting, based on determining that the window size of the window probe response matches the predetermined window size, at least one of a time threshold and a count threshold associated with the window size of the receiver node;

receiving, based on setting the at least one of the time threshold and the count threshold, a set of additional window probe responses, wherein each of the set of additional window probe responses comprises a window size update indicating a current window size of the receive buffer;

determining, based on receiving the set of additional window probe responses, at least one of

that the window size of the receiver node has been at the predetermined window size for a time period that is one of greater than and equal to the time threshold, and

that a number of window size updates in the set of additional window probe response, indicating that the window size of the receiver node remains at the predetermined window size, is one of greater than and equal to the count threshold; and

classifying, based on the determining, an application associated with the receiver node as unresponsive.

2. The method of claim 1, wherein the network traffic is Transmission Control Protocol/Internet Protocol traffic.

3. The method of claim 1, wherein the predetermined window size is 0.

4. The method of claim 1, wherein the time threshold is adjustable for different applications at the receiver node.

5. The method of claim 1, further comprising:

restarting, based on classifying the application associated with the receiver node as unresponsive, the application on one of the receiver node and a different node.

6. An information processing system for determining if an application has become unresponsive, the information processing system comprising:

a memory;

a processor communicatively coupled to the memory; and

an application monitor communicatively coupled to the memory and the processor, wherein the application monitor is configured to perform a method comprising;

monitoring network traffic at a receiver node;

receiving, based on monitoring network traffic a receiver node, a window probe response comprising a window size, the window size indicating a number of packets a receive buffer at the receiver node is currently able to hold;

comparing the window size of the window probe response to a predetermined window size;

determining, based on the comparing, that the window size of the window probe response matches the predetermined window size;

setting, based on determining that the window size of the window probe response matches the predetermined window size, at least one of a time threshold and a count threshold associated with the window size of the receiver node;

receiving, based on setting the at least one of the time threshold and the count threshold, a set of additional window probe responses, wherein each of the set of additional window probe responses comprises a window size update indicating a current window size of the receive buffer;

determining, based on receiving the set of additional window probe responses, at least one of

that the window size of the receiver node has been at the predetermined window size for a time period that is one of greater than and equal to the time threshold, and

that a number of window size updates in the set of additional window probe responses, indicating that the window size of the receiver node remains at the predetermined window size, is one of greater than and equal to the count threshold; and

classifying, based on the determining, an application associated with the receiver node as unresponsive.

7. The information processing system of claim 6, further comprising:

restarting, based on classifying the application associated with the receiver node as unresponsive, the application on one of the receiver node and a different node.

8. A tangible non-transitory computer program storage product for determining, with an information processing system, if an application has become unresponsive, the computer program storage product comprising instructions for:

monitoring network traffic at a receiver node;

receiving, based on monitoring network traffic a receiver node, a window probe response comprising a window size, the window size indicating a number of packets a receive buffer at the receiver node is currently able to hold;

comparing the window size of the window probe response to a predetermined window size;

determining, based on the comparing, that the window size of the window probe response matches the predetermined window size;

setting, based on determining that the window size of the window probe response matches the predetermined window size, at least one of a time threshold and a count threshold associated with the window size of the receiver node;

receiving, based on setting the at least one of the time threshold and the count threshold, a set of additional window probe responses, wherein each of the set of additional window probe responses comprises a window size update indicating a current window size of the receive buffer;

determining, based on receiving the set of additional window probe responses, at least one of

that the window size of the receiver node has been at the predetermined window size for a time period that is one of greater than and equal to the time threshold, and

that a number of window size updates in the set of additional window probe response, indicating that the window size of the receiver node remains at the predetermined window size, is one of greater than and equal to the count threshold; and

classifying, based on the determining, an application associated with the receiver node as unresponsive.

9. The tangible non-transitory computer program storage product of claim 8, wherein the time threshold is adjustable for different applications at the receiver node.

10. The tangible non-transitory computer program storage product of claim 8, further comprising instructions for:

restarting, based on classifying the application associated with the receiver node as unresponsive, the application on one of the receiver node and a different node.