Method and apparatus for detecting failures in network components

• US 7,167,912 B1
• Filed: 08/09/2002
• Issued: 01/23/2007
• Est. Priority Date: 08/09/2002
• Status: Expired due to Fees

• Current Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Original Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Inventors: Dhingra, Anurag
• Primary Examiner(s): VU, THONG H
• Application Number:
  US10/215,636
• Time in Patent Office: 1,628 Days
• Field of Search: 709/229, 709/238, 709/246, 709/201, 709223-225, 709/207, 709/202, 709/217, 370/252, 370/401, 370/339, 370/352, 714/43, 340/5.8
• US Class Current: 709/223
• CPC Class Codes: H04L 43/0817 by checking functioning
• CPC Class Codes: H04L 43/10 Active monitoring, e.g. heartbeat, ping or trace-route

Abstract

A monitoring system includes a monitoring component that compares a number of application-specific request packets sent with a number of response packets received pursuant to the sent request packets during a designated monitoring period to detect failure in a monitored component without using any pinging packets. The application-specific request and response packets contain data to perform tasks specific to an application that is not associated with operational status monitoring information and the request and response packets are not specifically ping packets. A method for monitoring includes counting a number of application-specific request packets sent with a number of response packets received during a monitoring period without any pinging. If one or more request packets are sent but no response packets are received during a designated monitoring period, the monitored component is identified as failed. If some, but not all, response packets are received, the monitored component is identified as operational. Alternatively, if no response packets are received after a first monitoring period, the monitored component is identified as potentially failed. If no response packets are received after the end of a second monitoring period, the monitored component is identified as failed.

Claims

1. A monitoring system, comprising:

a monitoring component that compares a number of application specific request packets sent by an application with a number of response packets received pursuant to the sent request packets during a designated monitoring period to detect failure in a monitored component without using any pinging packets, wherein

the request and response packets contain data used by the monitored component to perform an application that is not associated with component operation status monitoring, and wherein the request and response packets are not specifically ping packets;

the monitoring component identifies failure in the monitored component if one or more request packets are sent and no corresponding response packet is received by the monitoring component during a designated monitoring period.

2. The system according to claim 1 wherein the monitoring component identifies potential failure in the monitored component when at least one request packet is sent during a first monitoring period and no response packets are received back during the first monitoring period.

3. A The system according to claim 2 wherein the monitoring component identifies failure in the monitored component when no response packets are received back during either the first or a second monitoring period.

4. The system according to claim 1, wherein the application includes a web browser application, the monitored component includes a web server, and the request and response packets contain data used for exchanging web browser application-specific information with the web server that is not associated with operational status information of the web server.

5. The system according to claim 1, wherein the application includes an authentication application, the monitored component includes an authentication server, and the authentication request and response packets contain data used for exchanging authentication application-specific information with the authentication server that is not associated with operational status information of the authentication server and the authentication request and response packets are not specifically ping packets.

6. The system according to claim 5, wherein the authentication application comprises a Remote Access Dial-In User Service (RADIUS) application.

7. A monitoring system, comprising:

a network access server (NAS) for a Remote Access Dial-In User Service (RADIUS) application to transmit an access request packet to a primary RADIUS server and to receive a corresponding response packet from the primary RADIUS server,

a component to detect failure in the primary RADIUS server by comparing a number of RADIUS request packets with a number of RADIUS response packets that are already transmitted between the NAS and the primary RADIUS server without exchanging any ping specific packets, wherein the primary RADIUS server is considered fail if one or more access request packet is transmitted and no corresponding response packet is received by the network access server during a designated monitoring period.

8. The system according to claim 7 including at least one standby RADIUS server, the component redirecting the access requests to the standby RADIUS server when the component identifies failure in the primary RADIUS server.

9. The monitoring system of claim 7, the access request packet contains a user name, an encrypted password, a NAS IP address and port information.

10. The monitoring system of claim 7, the response packet is an access reject or an access accept.

11. The monitoring system of claim 10, the access accept contains a list of attribute-value pairs that describes the parameters to be used for a session, including service type, protocol type, IP address to assign the user, access list to apply or a static route to install in the NAS routing table.

12. A method for monitoring a component, comprising:

counting a number of application-specific request packets sent during a monitoring period with a number of response packets received pursuant to the sent request packets wherein

the request and response packets contain information that is used by a monitored component to perform an application that is not associated with component operation status monitoring, and wherein the request and response packets are not specifically ping packets;

identifying failure in a monitored component if one or more request packets are sent and no corresponding response packet is received by a monitoring component during a designated monitoring period; and

using the request packets and response packets as part of an authentication process for gaining access to a system.

13. The method according to claim 12 including counting the request packets and response packets for a Remote Access Dial-In User Service (RADIUS) request and response authentication protocol.

14. The method according to claim 12 including:

identifying failure in the monitored component when one or more request packets are sent during the designated monitoring period and some, but not all, response packets are received during the designated monitoring period.

15. A The method according to claim 12 including:

identifying failure in the monitored component when the request packets are sent during a first monitoring period and no response packets are received during the first monitoring period;

monitoring for response packets during a second monitoring period;

identifying failure in the monitored component as dead when no response packets are received during the first or second monitoring period; and

removing the potentially failure identification from the monitored component when response packets are received during the second monitoring period.

16. A The method according to claim 12 including selecting the designated monitoring period to be at least as long as a time required for the monitored component to receive a request packet and send back a response packet.

17. A The method according to claim 12 including:

sending the request packets from a network access server to a primary authentication server; and

redirecting the requests to a standby authentication server when the primary authentication server is identified fail.

18. An electronic storage medium containing computer executable code for monitoring a component, the computer executable code comprising:

code for counting a number of application-specific request packets sent during a monitoring period with a number of response packets received pursuant to the sent request packets;

wherein

the application-specific request and response packets contain information that is used by a monitored component to perform an application that is not associated with

component operation status monitoring, and wherein the request and response packets are not specifically ping packets;

code for identifying failure in a monitored component if one or more request packets are sent and no corresponding response packets are received by a monitoring component during a designated monitoring period.

19. The electronic storage medium according to claim 18 including code for counting the request packets and response packets for a Remote Access Dial-In User Service (RADIUS) request and response authentication protocol.

20. The electronic storage medium according to claim 18 including:

code for identifying failure in the monitored component when one or more request packets are sent during the designated monitoring period and some, but not all, response packets are received during the designated monitoring period.

21. The electronic storage medium according to claim 18 including:

code for identifying failure in the monitored component when the request packets are sent during a first monitoring period and no response packets are received during the first monitoring period;

code for monitoring for response packets during a second monitoring period;

code for identifying failure in the monitored component when no response packets are received during the first or second monitoring period; and

code for removing the potentially failure identification from the monitored component when response packets are received during the second monitoring period.

22. The electronic storage medium according to claim 18 including:

code for sending the request packets from a network access server to a primary authentication server; and

code for redirecting the requests to a standby authentication server when the primary authentication server is identified fail.

23. A system for monitoring a component, comprising:

means for counting a number of application specific request packets sent during a monitoring period with a number of response packets received pursuant to the sent request packets wherein the request and response packets contain information that is used by a monitored component to perform an application that is not associated with component operation status monitoring and wherein the request and response packets are not specifically ping packets;

means for identifying failure in a monitored component if one or more request packet is sent and no corresponding response packet is received by a monitoring component during a designated monitoring period.

24. The system according to claim 23 including means for counting the request packets and response packets for a Remote Access Dial-In User Service (RADIUS) request and response authentication protocol.

25. The system according to claim 23 including:

means for identifying failure in the monitored component when one or more request packets are sent during the designated monitoring period and some, but not all, response packets are received during the designated monitoring period.

26. The system according to claim 23 including:

means for identifying failure in the monitored component when the request packets are sent during a first monitoring period and no response packets are received during the first monitoring period;

means for monitoring for response packets during a second monitoring period;

means for identifying failure in the monitored component when no response packets are received during the first or second monitoring period; and

means for removing the potentially failure identification from the monitored component when response packets are received during the second monitoring period.

27. The system according to claim 23 including:

means for sending the request packets from a network access server to a primary authentication server; and

means for redirecting the requests to a standby authentication server when the primary authentication server is identified fail.