Capturing Node Secures Crash Dump Files Without Power Fencing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current technologies face inefficiencies in capturing crash dump files in clustered systems, particularly due to the use of power fencing, which can result in data corruption and prolonged downtime during node crashes, as they require loading system dump images one at a time and are prone to errors.
Innovation Solution
A method where a capturing node becomes active before a totem token is declared lost, allowing for the secure capture of crash dump files without power fencing, by loading a secondary kernel that boots within 3 seconds and operates under the assumption that power fencing has occurred, preventing unsynchronized access to the storage system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If power fencing is used to prevent split brain in clustered systems, then cluster integrity is maintained, but crash dump files cannot be captured and data corruption occurs
Solution Approach 1:
The patent applies preliminary action by having the capturing node load the secondary kernel and prepare to capture crash dump files before the power fencing mechanism is activated. The capturing node becomes active and loads the secondary kernel within a predetermined time window before the totem token is declared lost, ensuring that crash dump files are captured before power fencing prevents further access to the storage system.
2Reliability
If power fencing is activated immediately upon node crash, then cluster safety is ensured, but downtime is prolonged and dump files are lost
Solution Approach 1:
The patent implements preliminary action by establishing a timeline where the capturing node loads the secondary kernel and prepares to capture dump files before the power fencing is activated. The capturing node must become active and load the secondary kernel within a predetermined time window before the totem token is declared lost, thereby capturing crash dump files before power fencing causes prolonged downtime.
Solution Approach 2:
The patent introduces an intermediary mechanism where a capturing node acts as a mediator between the crashing node and the storage system. The capturing node loads a secondary kernel that can access the storage system independently, serving as an intermediary that captures crash dump files before the power fencing mechanism takes effect and prevents further access.
3Loss of information
If crash dump files are captured using current technology, then some data may be recovered, but the process is inefficient and prone to errors
Solution Approach 1:
The patent introduces an intermediary capturing node that loads a secondary kernel to independently access the storage system and capture crash dump files. This intermediary mechanism eliminates the need for manual intervention and automated post-crash analysis, significantly improving capture efficiency and reducing errors associated with current technology.
Solution Approach 2:
The patent implements self-service by enabling the capturing node to automatically load the secondary kernel and capture crash dump files without requiring manual intervention or complex automated analysis processes. The system performs the capture operation autonomously within the predetermined time window before power fencing is activated, improving efficiency and reducing human error.
Data Source
AI summary
In a computer storage system, crash dump files are secured without power fencing in a cluster of a plurality of nodes connected to a storage system. Upon an occurrence of a panic of a crashing node and prior to receiving a panic message of the crashing node by a surviving node loading, in the cluster, a capturing node to become active, prior to a totem token being declared lost by the surviving node, for capturing the crash dump files of the crashing node, while manipulating the surviving node to continue to operate under the assumption the power fencing was performed on the crashing node.