End-to-end monitoring of overlay networks providing virtualized network services

• US 9,787,559 B1
• Filed: 03/28/2014
• Issued: 10/10/2017
• Est. Priority Date: 03/28/2014
• Status: Active Grant

• Current Assignee: Juniper Networks Incorporated
• Original Assignee: Juniper Networks Incorporated
• Inventors: Schroeder, Hartmut
• Primary Examiner(s): Grey, Christopher
• Application Number:
  US14/228,874
• Time in Patent Office: 1,292 Days
• CPC Class Codes: H04L 43/067 using time frame reporting
• CPC Class Codes: H04L 43/08 Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
• CPC Class Codes: H04L 43/10 Active monitoring, e.g. heartbeat, ping or trace-route
• CPC Class Codes: H04L 43/106 using time related information in packets, e.g. by adding timestamps
• CPC Class Codes: H04L 43/12 Network monitoring probes
• CPC Class Codes: H04L 43/20 the monitoring system or the monitored elements being virtualised, abstracted or software-defined entities, e.g. SDN or NFV
• CPC Class Codes: H04L 43/55 Testing of service level quality, e.g. simulating service usage

Abstract

In one example, a network device external to a services complex injects a plurality of probe packets along service chains provided by the services complex, wherein each of the plurality of probe packets includes a first timestamp indicating a time at which the network device sent the respective probe packet. Each of a plurality of service nodes in the services complex modifies each of the plurality of probe packets by inserting a respective second timestamp indicating a respective time at which the respective service node processed the respective one of the plurality of probe packets. An analytics device aggregates probe report information received from each of the plurality of service nodes to determine one or more path monitoring metrics.

Claims

1. A method comprising:

injecting, by a network device external to a services complex of a data center having a plurality of service nodes, a first plurality of probe packets along service chains provided by the services complex,

wherein each of the service chains comprises a different ordered combination of network services applied to packets received by the service nodes, and wherein each of the first plurality of probe packets includes a first timestamp indicating a time at which the network device sent the respective probe packet, and overlay encapsulation with header information relating to a transport tunnel and tenant identification associated with a tenant enterprise network or a tenant public network configured for forwarding packets of a particular customer of the data center;

by one or more of the plurality of service nodes in the services complex, modifying each of the first plurality of probe packets by inserting a respective second timestamp indicating a respective time at which the respective service node processed the respective one of the first plurality of probe packets and applying the network services to the modified probe packets according to the respective service chains;

aggregating, by an analytics device, separate from the plurality of service nodes, probe report information received from each of the plurality of service nodes to determine one or more path monitoring metrics associated with service level agreements of customers of the data center;

identifying, by the analytics device and based on the probe report information, a component in a service node of the plurality of service nodes that is not receiving probe packets along one of the service chains;

in response to identifying the component, configuring a probe packet injector at a virtual router component of the service node having the component; and

injecting, by the probe packet injector of the virtual router component, a second plurality of probe packets for forwarding, by the service node having the component, along a remaining portion of the service chain that includes the component.

2. The method of claim 1, wherein the one or more path monitoring metrics comprise one or more of metrics regarding healthchecks, performance loss, jitter, or latency within a network.

3. The method of claim 1, further comprising:

in response to receiving the probe packets and by each of the plurality of service nodes, reporting the probe report information to the analytics device, wherein the probe report information comprises at least information relating to the first timestamp and the respective second timestamp.

4. The method of claim 1, wherein each packet of the first plurality of probe packets and the second plurality of probe packets comprises an Internet Control Message Protocol (ICMP) packet, and wherein modifying the probe packets comprises adding the timestamp entry to the timestamp field in place of padding data.

5. The method of claim 1, wherein the analytics device comprises a central network controller device.

6. The method of claim 1, wherein the network device comprises a data center border router, and wherein the analytics device comprises an analytics function on the data center border router that originated the packet.

7. The method of claim 1, wherein modifying the probe packets comprises modifying the probe packets by one or more of a physical switch, a virtual router, or a virtual switch of a service node of the plurality of service nodes.

8. The method of claim 1, wherein the network device comprises a data center border router positioned at a border of the service complex and a service provider network external to the service complex.

9. The method of claim 1, wherein the network device comprises a customer edge router positioned at a border of an access network and a service provider network external to the service complex.

10. The method of claim 1, further comprising generating the first plurality of probe packets such that at least some of the plurality of probe packets have different source Internet Protocol (IP) addresses to direct the probe packets to different service virtual machines along different service chains within the service complex.

11. The method of claim 1, wherein each packet of the first plurality of probe packets and the second plurality of probe packets comprises:

a Uniform Datagram Protocol (UDP) packet having the first timestamp in a UDP options field, or a plurality of Transmission Control Protocol (TCP) packet having the first timestamp in a TCP options field.

12. The method of claim 1,

wherein injecting the first plurality of probe packets comprises injecting the plurality of probe packets using a virtual routing and forwarding instance (VRF) other than a VRF configured for forwarding packets of the particular customer.

13. A system comprising:

a services complex of a data center, the services complex comprising a plurality of service nodes;

a border router external to the services complex, wherein the border router is configured to inject a first plurality of probe packets along service chains provided by the service complex, wherein each of the service chains comprises a different ordered combination of network services applied to packets received by the service nodes, wherein each of the first plurality of probe packets includes a first timestamp indicating a time at which the border router sent the respective probe packet and overlay encapsulation with header information relating to a transport tunnel and tenant identification associated with a tenant enterprise network or a tenant public network configured for forwarding packets of a particular customer of the data center;

wherein one or more of the plurality of service nodes is configured to modify each of the first plurality of probe packets by inserting a respective second timestamp indicating a respective time at which the respective service node processed the respective one of the first plurality of probe packets, and apply the network services to the modified probe packets according to the respective service chains; and

a central controller device configured to aggregate probe report information received from each of the plurality of service nodes to determine one or more path monitoring metrics associated with service level agreements of customers of the data center, identify, based on the probe report information, a component in a service node of the plurality of service nodes that is not receiving probe packets along one of the service chains, and, in response to identifying the component, configure a probe packet injector at a virtual router component of the service node having the component,

wherein the probe packet injector of the virtual router component injects a second plurality of probe packets for forwarding, by the service node having the component, along a remaining portion of the service chain that includes the component.