Method and system for non-disruptive data capture in networks

• US 7,525,910 B2
• Filed: 07/12/2004
• Issued: 04/28/2009
• Est. Priority Date: 07/16/2003
• Status: Active Grant

• Current Assignee: Marvell Asia Pte Limited (Marvell Technology Group Limited)
• Original Assignee: QLogic Corporation (Marvell Technology Group Limited)
• Inventors: Wen, William J.
• Primary Examiner(s): Rao; Seema S
• Assistant Examiner(s): WONG, XAVIER S
• Application Number:
  US10/889,635
• Publication Number:
  US 20050015518A1
• Time in Patent Office: 1,751 Days
• Field of Search: 370/452, 370/258, 370/424, 370/454, 370/460, 370/906, 370/230, 370/235, 370/389, 370/392, 370403-406, 370/488, 370/497, 710/15, 710/17, 710/62, 710/1
• US Class Current: 370/230
• CPC Class Codes: H04L 49/357 Fibre channel switches
• CPC Class Codes: H04L 49/555 Error detection
• CPC Class Codes: H04L 63/0227 Filtering policies mail message filtering H04L51/212

Abstract

A method and fiber channel switch element for capturing data in a fiber channel system is provided. The method includes, detecting a condition for enabling data capture; setting a control bit to enable data capture; and capturing data based on the set control bit. The switch element includes, a diagnostic port that can configured as a traditional communication port and/or a data capture port depending on incoming frame data. The diagnostic port filters incoming frames, which is used to enable data capture.

Claims

1. A method for capturing data in a communication port of a fibre channel switch element in a computer network, the method comprising:

a processor in the switch element detecting a condition for enabling data capture;

the processor, in response to the detected condition, setting a control bit to enable the communication port to capture data; and

capturing data in the communication port based on the set control bit, and regardless of whether the communication port has available buffer-to-buffer credit;

wherein data capture comprises collecting information from the computer network; and

wherein buffer-to-buffer credit limits a number of frames that the communication port may receive before a source port stops sending frames, the source port being a communication port in the computer network that sends frames.

2. The method of claim 1, wherein incoming data is sampled to determine what level of data capture may be required.

3. The method of claim 2, wherein incoming data is sampled to determine if the data is from a private or public loop.

8. The method of claim 3, wherein the processor is configured to use a frame source address field (S_ID) to determine if the loop is private or public.

4. A fibre channel switch element in a computer network, comprising:

a diagnostic port in the switch element configured to act as both a traditional communication port and a data capture port depending on the incoming frame data; and

a processor in the switch element;

wherein the processor is configured to detect a condition for enabling data capture by the diagnostic port to capture data; and

wherein the diagnostic port is configured to capture data based on the set control bit, and regardless of whether the diagnostic port has available buffer-to-buffer credit;

wherein data capture comprises collecting information from computer network; and

wherein buffer-to-buffer credit limits a number of frames that the diagnostic port may receive before a source port in another network device stops sending frames, the source port being a communication port in the computer network that sends frames.

5. The switch element of claim 4, wherein the processor is configured to filter incoming frames to enable data capture.

6. The switch element of claim 5, wherein the processor is configured to use frame filtering to determine if a frame is from a public or private loop.

7. The switch element of claim 6, wherein the processor is configured to use a frame source address field (S_ID) to determine if the loop is private or public.