Online file system consistency check for container data on a clustered filesystem

Patent metadata

Field	Value
Publication number	US-11650975-B2
Application number	US-202017119009-A
Country	US
Kind code	B2
Filing date	Dec 11, 2020
Priority date	Dec 11, 2020
Publication date	May 16, 2023
Grant date	May 16, 2023

Abstract

Online file system consistency check for container data on a clustered file system is provided via identifying inodes (index nodes) of a group of files in a clustered file system based on a cyber-resiliency for the clustered file system; grouping the inodes based on a buffer size allocated to a FSCK (File System Consistency Check) operation; passing the inodes of to the FSCK operation in a single iteration when a total size of the inodes is less than the buffer size; or when the total size of the inodes is greater than the buffer size, identifying inodes that belong to a first container and that belong to a second container; passing the inodes that belong to the first container to the FSCK operation in a first iteration; and passing, after the first iteration completes, the inodes that belong to the second container to the FSCK operation in a second iteration.

First claim

What is claimed is:

1. A method comprising:
identifying inodes of a first group of files in a clustered file system based on a cyber-resiliency for the clustered file system;
grouping the inodes based on a buffer size allocated to a FSCK (File System Consistency Check) operation for the clustered file system;
in response to the buffer size being greater than a total size of the inodes of the first group of files, passing the inodes of the first group of files to the FSCK operation in a single iteration;
in response to the buffer size being less than a total size of the inodes of a second group of files:
identifying the inodes of the second group of files that belong to a first container and that belong to a second container;
passing the inodes of the second group of files that belong to the first container to the FSCK operation in a first iteration; and
passing, after the first iteration completes, the inodes of the second group of files that belong to the second container to the FSCK operation in a second iteration.

2. The method of claim 1 , further comprising:
identifying the second group of files in the clustered file system based on the cyber-resiliency for the clustered file system;
separating inodes of the second group of files; and
grouping the inodes of the second group of files based on the buffer size allocated to the FSCK operation for the clustered file system.

3. The method of claim 1 , wherein identifying the first group of files in the clustered file system identifies the files that are important to the cyber-resiliency of the clustered file system further comprises:
monitoring file operations captured;
analyzing frequency patterns for how often the files are generated or updated by a given container;
analyzing access patterns of how many other containers, besides the given container, access the files; and
analyzing operational patterns for what types of operations are performed on the files by the other containers.

4. The method of claim 1 , wherein identifying the first group of files in the clustered file system identifies the files that are important to the cyber-resiliency of the clustered file system further comprises:
analyzing a time of last access for a given file; and
marking as the given file as one of passive or active based on whether the time of last access is within a threshold.

5. The method of claim 1 , further comprising:
selecting a number of Erasure Code (EC) fragments to validate for each of the files passed to the FSCK operation; and
scanning the files based on priority levels of the files.

6. The method of claim 5 , wherein the number of EC fragments are dynamically selected based on system load and resources allocated to the FSCK operation.

7. The method of claim 1 , wherein the first group of files consists of files assigned a priority level above a priority threshold for cyber-resiliency for the clustered file system.

8. A system, comprising:
a processor; and
a memory storing instructions that when executed by the processor enable performance of an operation that includes:
identifying inodes of a first group of files in a clustered file system based on a cyber-resiliency for the clustered file system;
grouping the inodes based on a buffer size allocated to a FSCK (File System Consistency Check) operation for the clustered file system;

in response to the buffer size being greater than a total size of the inodes of the first group of files, passing the inodes of the first group of files to the FSCK operation in a single iteration;
in response to the buffer size being less than a total size of the inodes of a second group of files:
identifying the inodes of the second group of files that belong to a first container and that belong to a second container;
passing the inodes of the second group of files that belong to the first container to the FSCK operation in a first iteration; and
passing, after the first iteration completes, the inodes of the second group of files that belong to the second container to the FSCK operation in a second iteration.

9. The system of claim 8 , wherein the operation further includes:
identifying the second group of files in the clustered file system based on the cyber-resiliency for the clustered file system;
separating inodes of the second group of files; and
grouping the inodes of the second group of files based on the buffer size allocated to the FSCK operation for the clustered file system.

10. The system of claim 8 , wherein identifying the first group of files in the clustered file system identifies the files that are important to the cyber-resiliency of the clustered file system further comprises:
monitoring file operations captured;
analyzing frequency patterns for how often the files are generated or updated by a given container;
analyzing access patterns of how many other containers, besides the given container, access the files; and
analyzing operational patterns for what types of operations are performed on the files by the other containers.

11. The system of claim 8 , wherein identifying the first group of files in the clustered file system identifies the files that are important to the cyber-resiliency of the clustered file system further comprises:
analyzing a time of last access for a given file; and
marking as the given file as one of passive or active based on whether the time of last access is within a threshold.

12. The system of claim 8 , further comprising:
selecting a number of Erasure Code (EC) fragments to validate for each of the files passed to the FSCK operation; and
scanning the files based on priority levels of the files.

13. The system of claim 12 , wherein the number of EC fragments are dynamically selected based on system load and resources allocated to the FSCK operation.

14. The system of claim 8 , wherein the first group of files consists of files assigned a priority level above a priority threshold for cyber-resiliency for the clustered file system.

15. A computer-readable storage medium including instructions that when executed by a processor enable performance of operations including:
identifying inodes of a first group of files in a clustered file system based on a cyber-resiliency for the clustered file system;
grouping the inodes based on a buffer size allocated to a FSCK (File System Consistency Check) operation for the clustered file system; and
in response to the buffer size being greater than a total size of the inodes of the first group of files, passing the inodes of the first group of files to the FSCK operation in a single iteration;
in response to the buffer size being less than a total size of the inodes of a second group of files:
identifying the inodes of the second group of files that belong to a first container and that belong to a second container;
passing the inodes of the second group of files that belong to the first container to the FSCK operation in a first iteration; and
passing, after the first iteration completes, the inodes of the second group of files that belong to the second container to the FSCK operation in a second iteration.

16. The computer-readable storage medium of claim 15 , wherein the operations further include:
identifying the second group of files in the clustered file system based on the cyber-resiliency for the clustered file system;
separating inodes of the second group of files; and
grouping the inodes of the second group of files based on the buffer size allocated to the FSCK operation for the clustered file system.

17. The computer-readable storag

Assignees

• IBM

Inventors

• Ali Asmahan
• Lara Christina
• Jain Abhishek
• Eda Sasikanth

Classifications

• G06F16/182 Primary

  Distributed file systems · CPC title

• G06F16/2365 Primary

  Ensuring data consistency and integrity · CPC title