Method and apparatus for outage measurement

• US 7,149,917 B2
• Filed: 07/30/2002
• Issued: 12/12/2006
• Est. Priority Date: 07/30/2002
• Status: Active Grant

• Current Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Original Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Inventors: Sejun, Song, Huang, Jiandong, Marathe, Madhav
• Primary Examiner(s): Beausoliel, Robert
• Assistant Examiner(s): McCarthy, Christopher S.
• Application Number:
  US10/209,845
• Publication Number:
  US 20040024865A1
• Time in Patent Office: 1,596 Days
• Field of Search: 714/43, 714/47, 714/4
• US Class Current: 714/4.3
• CPC Class Codes: H04L 41/0213 Standardised network management protocols, e.g. simple network management protocol [SNMP]
• CPC Class Codes: H04L 41/044 comprising hierarchical management structures
• CPC Class Codes: H04L 41/0604 using filtering, e.g. reduction of information by using priority, element types, position or time
• CPC Class Codes: H04L 41/064 involving time analysis
• CPC Class Codes: H04L 43/0811 by checking connectivity
• 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

Abstract

An Outage Measurement System (OMS) monitors and measures outage data at a network processing device. The outage data can be stored in the device and transferred to a Network Management System (NMS) or other correlation tool for deriving outage information. The OMS automates the outage measurement process and is more accurate, efficient and cost effective than pervious outage measurement systems.

Claims

1. A method for measuring outages in a network, comprising:

identifying single point of failure devices in the network;

selectively locating outage measurement systems in the identified single point of failure devices;

automatically testing for outages locally in the selected single point of failure devices using the outage measurement systems;

polling for layer-2 outages using the outage measurement systems and polling for layer-3 outages according to results of the layer-2 polling; and

using a combination of the polled layer-2 and layer-3 outages to distinguish between local outages local to the identified devices and remote outages associated with links or devices connected to the identified devices.

2. A method for identifying outages, comprising:

polling for local outages of local objects at a router, peer router, router processor or switch;

polling with the router, peer router, router processor or switch for remote outages of remote objects connected to the router, peer router, router processor or switch; and

comparing the local outages with the remote outages to distinguish between outages caused by the local objects and outages caused by the remote objects.

deriving an Accumulated Outage Time (AOT) for the local objects;

deriving an Accumulated Outage Time (AOT) for the remote objects; and

distinguishing local object outages from remote object outages by comparing the AOT for the local objects with the AOT for the remote objects.

3. A method for network outage monitoring, comprising:

monitoring with a router, peer router, router processor or switch for outages associated with a configuration of devices connected to the router, peer router, router processor or switch;

deriving by the router peer router, router processor or switch accumulated outage time or failure count per object based on the outages;

automatically discovering a new configuration of devices connected to the router, peer router, router processor or switch; and

dynamically updating outage monitoring according to the new configuration of devices.

4. A method according to claim 3 including using a Cisco Discovery Protocol (CDP), Address Resolution Protocol (ARP) protocol, or Internet Control Message Protocol (ICMP) for automatically discovering the new configuration of devices.

5. A method according to claim 3 including:

storing a configuration table to identify devices connected to the router, peer router, router processor or switch; and

automatically updating the configuration table with the new configuration of devices.

6. A method for measuring outages, comprising:

locally monitoring for outages at a network processing device;

filtering the monitored outages with the local network processing device;

sending the filtered outages to an outage correlation system that is physically separate from the network processing device;

identifying a plurality of outage events associated with the network processing device; and

sending the plurality of outage events to the outage correlation system in a same communication.

7. A method according to claim 6 including calculating an Accumulated Outage Time (AOT) outage parameter or a Number of Accumulated Failures (NAF) outage parameter from the monitored outages and sending the AOT or NAF to the correlation system.

8. A method according to claim 7 including deriving a Mean Time Between Failures or a Mean Time To Repair value from the AOT or NAF parameters with the correlation system.

9. A method according to claim 6 including filtering the monitored outages from a system log file.

10. A method for measuring outages, comprising:

locally monitoring for outages at a network processing device;

filtering the monitored outages with the local network processing device;

sending the filtered outages to an outage correlation system;

identifying different outages associated with a same line card;

filtering the different line card outages into a single line card outage indication; and

sending the single line card outage indication to the outage correlation system.

11. A method for measuring an outage for a network processing device, comprising:

generating with a router, peer router, router processor or switch a time stamp for the network processing device;

storing the time stamp locally in the network processing device;

periodically updating with a router, peer router, router processor or switch the stored time stamp with a most recent time stamp for the network processing device;

retaining the most recent stored time stamp during a network processing device outage;

using the retained most recent stored time stamp to determine an outage time for the network processing device;

identifying a system up time when the network processing device has recovered from the outage; and

comparing the stored time stamp with the system up time to determine the outage time.

12. A method according to claim 11 including periodically storing the most recent time stamp in a persistent memory.

13. A method according to claim 11 wherein the outage comprises a power loss to the network processing device.

14. A method according to claim 11 wherein the time stamp stored in the network processing device is updated about once every minute by the router, peer router, router processor or switch.

15. A method for measuring outages, comprising:

locally monitoring for outages at a network processing device;

filtering the monitored outages with the local network processing device;

sending the filtered outages to an outage correlation system that is physically separate from the network processing device;

identifying different outages associated with a same line card; and

filtering the different line card outages into a single line card outage indication.

16. A method for measuring outages, comprising:

locally monitoring for outages at a network processing device;

filtering the monitored outages with the local network processing device;

sending the filtered outages to an outage correlation system that is physically separate from the network processing device;

identifying a plurality of outage events with a same line card; and

sending the plurality of outage events to the outage correlation system in a same communication.

17. A method for measuring outages, comprising:

locally monitoring for outages at a network processing device;

filtering the monitored outages with the local network processing device;

sending the filtered outages to an outage correlation system that is physically separate from the network processing device;

identifying different outages associated with the network processing device; and

filtering the different outages into a single outage indication.