Managing failure behavior for computing nodes of provided computer networks

• US 8,488,446 B1
• Filed: 10/27/2010
• Issued: 07/16/2013
• Est. Priority Date: 10/27/2010
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Brandwine, Eric Jason
• Primary Examiner(s): ROSE, KERRI M
• Application Number:
  US12/913,501
• Time in Patent Office: 993 Days
• Field of Search: 370216-218, 370/225, 370/228, 370/252, 370/254, 370/400, 370/401, 370/409
• US Class Current: 370/225
• CPC Class Codes: H04L 41/064 involving time analysis
• CPC Class Codes: H04L 41/0654 using network fault recovery ring fault isolation or reconfiguration in loop networks without recovery actions by a network management system H04L12/437
• CPC Class Codes: H04L 41/0681 Configuration of triggering conditions
• CPC Class Codes: H04L 41/147 for predicting network behaviour
• CPC Class Codes: H04L 45/22 Alternate routing
• CPC Class Codes: H04L 45/28 using route fault recovery
• CPC Class Codes: H04L 69/40 for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection management of faults, events, alarms or notifications in data switching networks H04L41/06

Abstract

Techniques are described for providing managed computer networks. In some situations, the techniques include managing communications for computing nodes of a managed computer network by using one or more particular computing nodes of the managed computer network that are configured to operate as intermediate destinations to handle at least some communications that are sent by and/or directed to one or more other computing nodes of the managed computer network. In addition, the techniques may include managing the communications in accordance with configured failure behavior specified for one or more computing nodes of the computer network, such as specified failure behavior for a computing node configured to operate as an intermediate destination that indicates how communications that would otherwise be routed via the intermediate destination computing node are to be handled if the intermediate destination computing node fails or is otherwise unavailable (e.g., to block or allow such communications).

Claims

1. A computing system, comprising:

one or more processors; and

one or more of multiple manager modules that are each associated with one or more of multiple computing nodes of a first computer network overlaid on one or more distinct second networks, the one or more second networks interconnecting the multiple computing nodes via the multiple manager modules, each of the one or more manager modules being configured to, when executed by at least one of the one or more processors, manage multiple communications sent by the one or more computing nodes with which the manager module is associated in accordance with specified configuration information for the first computer network, the specified configuration information including a specified failure behavior for a first computing node of the multiple computing nodes that indicates how to handle any future communications to be sent using the first computing node as an intermediate destination if the first computing node is unavailable, the managing of the multiple communications by each of the one or more manager modules including;

determining for a first of the multiple communications to use the first computing node as an intermediate destination to receive the first communication before the first communication is forwarded to a distinct indicated final destination for the first communication, the determining to use the first computing node as the intermediate destination being based at least in part on the specified configuration information for the first computer network;

sending the first communication via the associated manager module to the one or more second networks for forwarding to the first computing node to enable the first computing node to manage further handling of the first communication before further forwarding of the first communication to the indicated final destination for the first communication; and

after the further handling of the first communication by the first computing node,

determining for a second of the multiple communications to use the first computing node as an intermediate destination to receive the second communication and to manage further handling of the second communication before the second communication is forwarded to a distinct indicated final destination for the second communication;

determining that the first computing node is unavailable for the further handling of the second communication; and

after determining an alternative manner of handling the second communication based on the specified failure behavior for the first computing node, performing the alternative manner of handling the second communication.

2. The computing system of claim 1 wherein the first computer network is a virtual computer network, wherein the specified configuration information indicates a specified network topology for the first virtual computer network that includes the first computing node being located between a first group of one or more of the multiple computing nodes and a second group of one or more other computing nodes, wherein the one or more second networks are a substrate network on which the first virtual computer network is overlaid without physically implementing the specified network topology for the first virtual computer network, wherein the first and second communications are each sent between the first and second groups, and wherein the determining to use the first computing node as the intermediate destination for the first communication and the second communication is performed based on the specified network topology and on the first and second communications each being sent between the first and second groups.

3. The computing system of claim 1 wherein the specified failure behavior for the first computing node indicates that a communication to be sent to the first computing node as an intermediate destination while the first computing node is unavailable is to be forwarded to an intended final destination of that communication despite the first computing node being unavailable, and wherein each of the one or more manager modules is further configured to determine the alternative manner of handling the second communication based on the specified failure behavior for the first computing node, the alternative manner of handling the second communication being to send the second communication to the one or more second networks for forwarding directly to the indicated final destination for the second communication without being forwarded to any other of the multiple computing nodes as an intermediate destination.

4. The computing system of claim 1 wherein the specified failure behavior for the first computing node indicates that a communication to be sent to the first computing node as an intermediate destination while the first computing node is unavailable is not to be forwarded to an intended final destination of that communication despite the first computing node being unavailable, and wherein each of the one or more manager modules is further configured to determine the alternative manner of handling the second communication based on the specified failure behavior for the first computing node, the alternative manner of handling the second communication being to block the second communication from being sent to the one or more second networks for forwarding to the indicated final destination for the second communication.

5. The computing system of claim 1 further comprising the first computing node, the first computing node being configured to receive multiple communications that are intended for other indicated final destinations and to handle each of at least some of the received multiple communications by performing specified processing on the communication, the first communication being one of the at least some communications.

6. The computing system of claim 1 wherein the one or more manager modules each includes software instructions for execution by the at least one processors of the computing system.

7. The computing system of claim 1 wherein the one or more manager modules each consists of a means for managing multiple communications for the first computer network in accordance with specified configuration information for the first computer network by:

determining for a first of the multiple communications to use the first computing node as an intermediate destination to receive the first communication before the first communication is forwarded to a distinct indicated final destination for the first communication, the determining to use the first computing node as the intermediate destination being based at least in part on the specified configuration information for the first computer network;

sending the first communication via the associated manager module to the one or more second networks for forwarding to the first computing node to enable the first computing node to manage further handling of the first communication before further forwarding of the first communication to the indicated final destination for the first communication; and

after the further handling of the first communication by the first computing node,

determining for a second of the multiple communications to use the first computing node as an intermediate destination to receive the second communication and to manage further handling of the second communication before the second communication is forwarded to a distinct indicated final destination for the second communication;

determining that the first computing node is unavailable for the further handling of the second communication; and

after determining an alternative manner of handling the second communication based on the specified failure behavior for the first computing node, performing the alternative manner of handling the second communication.

8. A computer-implemented method comprising:

managing, by one or more configured computing systems, multiple communications sent by one or more of multiple computing nodes of a first computer network in accordance with specified configuration information for the first computer network, the first computer network being overlaid on one or more distinct second networks that interconnect the multiple computing nodes, the specified configuration information including a specified failure behavior for a first computing node of the multiple computing nodes that indicates how to handle future communications to be sent using the first computing node as an intermediate destination if the first computing node is unavailable, the managing of the multiple communications including;

determining, by the one or more configured computing systems and for a first of the multiple communications, to use the first computing node as an intermediate destination to receive the first communication before the first communication is forwarded to an indicated final destination, the determining to use the first computing node as the intermediate destination being based at least in part on the specified configuration information for the first computer network;

sending, by the one or more configured computing systems, the first communication to the one or more second networks for forwarding to the first computing node, to enable the first computing node to manage further handling of the first communication before further forwarding the first communication to the indicated final destination;

after the further handling of the first communication by the first computing node, determining, by the one or more configured computing systems and for a second of the multiple communications, that the first computing node is an intermediate destination to receive the second communication based at least in part on the specified configuration information and that the first computing node is unavailable for handling of the second communication; and

after determining an alternative manner of handling the second communication based on the specified failure behavior for the first computing node, initiating performing, by the one or more configured computing systems, the alternative manner of handling the second communication.

9. The method of claim 8 wherein the first computer network is a virtual computer network, and wherein the one or more second networks include a substrate network on which the virtual computer network is overlaid.

10. The method of claim 9 wherein the specified configuration information indicates a specified network topology for the virtual computer network that includes the first computing node being located between a first group of one or more of the multiple computing nodes and a second group of one or more other computing nodes, and wherein the virtual computer network is overlaid on the substrate network without physically implementing the specified network topology for the virtual computer network.

11. The method of claim 10 wherein the first and second communications are each sent between the first and second groups, and wherein the determining to use the first computing node as the intermediate destination for the first communication and for the second communication is performed based on the specified network topology and on the first and second communications being sent between the first and second groups.

12. The method of claim 8 wherein the specified failure behavior for the first computing node indicates, for a communication to be sent to the first computing node as an intermediate destination, to forward the communication to an intended final destination of that communication if the first computing node is unavailable, and wherein the initiating performing of the alternative manner of handling the second communication includes sending the second communication to the one or more second networks for forwarding directly to a final destination for the second communication.

13. The method of claim 12 wherein the specified configuration information further indicates a second specified failure behavior for a second computing node of the multiple computing nodes that indicates how to handle future communications to be sent using the second computing node as an intermediate destination if the second computing node is unavailable, the second specified failure behavior indicating, for a communication to be sent to the second computing node as an intermediate destination, to not forward the communication to an intended final destination of that communication if the second computing node is unavailable, and wherein the method further comprises:

determining that the second computing node is unavailable; and

blocking one or more additional communications to be sent to the second computing node as an intermediate destination.

14. The method of claim 8 wherein the specified failure behavior for the first computing node indicates, for a communication to be sent to the first computing node as an intermediate destination, to not forward the communication to an intended final destination of that communication if the first computing node is unavailable, and wherein the initiating performing of the alternative manner of handling the second communication includes blocking the second communication from being sent to a final destination for the second communication.

15. The method of claim 8 further comprising configuring the first computing node to receive multiple communications that are intended for other indicated final destinations and to handle each of at least some of the received multiple communications by performing specified processing on the communication, the first communication being one of the at least some communications.

16. The method of claim 8 further comprising, after the further handling of the first communication by the first computing node:

receiving updated specified failure behavior for the first computing node that indicates, while the first computing node is unavailable, a specified alternative manner for handling future communications to be sent to the first computing node as an intermediate destination; and

implementing the updated specified failure behavior for one or more additional communications to be sent to the first computing node as an intermediate destination after the receiving of the updated specified failure behavior.

17. The method of claim 16 wherein the specified alternative manner for the updated specified failure behavior involves taking one or more specified actions that do not include forwarding the additional communications to intended final destinations of the additional communications and that do not include blocking the additional communications from being forwarded to the intended final destinations of the additional communications.

18. The method of claim 8 wherein the specified configuration information further indicates that the specified failure behavior for the first computing node is to be provided temporarily after the first computing node becomes unavailable until a new computing node of the first computer network is available to replace the first computing node, and wherein the method further comprises, after the determining that the first computing node is unavailable for handling of the second communication, initiating, by the one or more configured computing systems, providing of the new computing node for the first computer network to replace the unavailable first computing node.

19. The method of claim 8 further comprising, after the determining that the first computing node is unavailable, performing the specified failure behavior temporarily until a later time at which the first computing node again becomes available.

20. The method of claim 8 wherein the specified configuration information further indicates that communications are to be sent to the first computing node as an intermediate destination if those communications are between computing nodes of indicated first and second groups, the first and second groups each including two or more of the multiple computing nodes, and wherein the method further comprises, before the first computing node is unavailable, forwarding multiple additional communications over the one or more second network between the computing nodes of the first group without sending the additional communications to the first computing node.

21. The method of claim 8 further comprising, after the further handling of the first communication by the first computing node, and if the first computing node determines to initiate further forwarding of the first communication to the indicated final destination, initiating further forwarding of the first communication over the one or more second networks from the first computing node to the indicated final destination.

22. The method of claim 8 wherein the further handling of the first communication by the first computing node includes providing firewall capabilities, intrusion detection capabilities, content analysis capabilities, network acceleration capabilities, or virtual private network (“

VPN”

) connection capabilities.

23. The method of claim 8 further comprising, under control of the first computing node:

receiving the forwarded first communication; and

performing the further handling of the first communication by determining not to further forward the first communication to the indicated final destination.

24. The method of claim 8 further comprising monitoring availability of the first computing node, and wherein the determining that the first computing node is unavailable is based on the first computing node failing and is identified based at least in part on the monitoring.

25. The method of claim 8 wherein the alternative manner of handling the second communication includes forwarding the second communication toward a final destination computing node for the second communication by bypassing the first computing node.

26. The method of claim 8 wherein the one or more configured computing systems are part of a configurable network service that provides the first computer network to a first client and that provides multiple other computer networks to multiple other clients, wherein the specified configuration information further indicates that communications are to be sent to the first computing node as an intermediate destination if those communications are between computing nodes of indicated first and second groups, wherein the first computing node includes a first logical network interface associated with the first group and a second logical network interface associated with the second group, and wherein the specified failure behavior for the first computing node further indicates to logically interconnect the first and second logical network interfaces while the first computing node is unavailable.

27. A non-transitory computer-readable medium having stored contents that configure a computing system to:

manage, by the configured computing system, multiple communications sent by one or more of multiple computing nodes of a first computer network in accordance with specified configuration information for the first computer network that indicates to use a first computing node of the multiple computing nodes as an intermediate destination for the multiple communications, the first computer network being overlaid on one or more distinct second networks that interconnect the multiple computing nodes, the specified configuration information including a specified failure behavior for the first computing node that indicates how to handle the multiple communications if the first computing node is unavailable, the managing of the multiple communications including;

determining, for a first of the multiple communications, to use the first computing node as an intermediate destination based at least in part on the specified configuration information for the first computer network;

sending the first communication to the one or more second networks for forwarding to the first computing node, to enable the first computing node to manage further handling of the first communication before further forwarding the first communication to an indicated final destination for the first communication;

after the further handling of the first communication by the first computing node, determining that the first computing node is unavailable for handling of a second of the multiple communications; and

after determining an alternative manner of handling the second communication based on the specified failure behavior for the first computing node, initiating performing the alternative manner of handling the second communication.

28. The non-transitory computer-readable medium of claim 27 wherein the specified failure behavior for the first computing node indicates, if the first computing node is unavailable for one of the multiple communications, to forward the one communication directly to an intended final destination of the one communication.

29. The non-transitory computer-readable medium of claim 27 wherein the specified failure behavior for the first computing node indicates, if the first computing node is unavailable for one of the multiple communications, to block forwarding of the one communication to an intended final destination of the one communication.