Skip to Main Content
IBM System Storage Ideas Portal


This portal is to open public enhancement requests against IBM System Storage products. To view all of your ideas submitted to IBM, create and manage groups of Ideas, or create an idea explicitly set to be either visible by all (public) or visible only to you and IBM (private), use the IBM Unified Ideas Portal (https://ideas.ibm.com).


Shape the future of IBM!

We invite you to shape the future of IBM, including product roadmaps, by submitting ideas that matter to you the most. Here's how it works:

Search existing ideas

Start by searching and reviewing ideas and requests to enhance a product or service. Take a look at ideas others have posted, and add a comment, vote, or subscribe to updates on them if they matter to you. If you can't find what you are looking for,

Post your ideas
  1. Post an idea.

  2. Get feedback from the IBM team and other customers to refine your idea.

  3. Follow the idea through the IBM Ideas process.


Specific links you will want to bookmark for future use

Welcome to the IBM Ideas Portal (https://www.ibm.com/ideas) - Use this site to find out additional information and details about the IBM Ideas process and statuses.

IBM Unified Ideas Portal (https://ideas.ibm.com) - Use this site to view all of your ideas, create new ideas for any IBM product, or search for ideas across all of IBM.

ideasibm@us.ibm.com - Use this email to suggest enhancements to the Ideas process or request help from IBM for submitting your Ideas.

Status Delivered
Created by Guest
Created on Jul 18, 2014

Improve ability to detect BROKEN files

Currently it is not possible to easily identify BROKEN files in a GPFS FPO cluster following the failure of a disk. This can arise when there is a single copy of the data (which is on the failed disk). IBM advise running the following command:

mmfileid <file_system> -d:BROKEN > mmfileid.out 2>&1

However this can take an extremely long time and requires scanning the whole file system. Please can you inmprove this functionality. For example, can this command run as a lower priority process all the time effectively scanning the system ? Or can some mechanism allow only the files which belonged to that failed disk to be scanned ?

thanks

Chris

Idea priority Medium
  • Guest
    Reply
    |
    Sep 30, 2020

    Due to processing by IBM, this request was reassigned to have the following updated attributes:
    Brand - Servers and Systems Software
    Product family - IBM Spectrum Scale
    Product - Spectrum Scale (formerly known as GPFS) - Private RFEs
    Component - Product functionality

    For recording keeping, the previous attributes were:
    Brand - Servers and Systems Software
    Product family - IBM Spectrum Scale
    Product - Spectrum Scale (formerly known as GPFS) - Private RFEs
    Component - V3 Product functionality

  • Guest
    Reply
    |
    Oct 14, 2014

    The basic task of finding a BROKEN address in a file system cannot really be done in many different ways. One has to examine all disk address containers (i.e. inodes and indirect blocks) in the file system, which is exactly what mmfileid does. This operation is known as a metadata scan. If metadata reads are slow, and the file system has a large number of files, naturally a metadata scan takes a long time. However, trying to address this issue specifically for mmfileid would have very limited utility, since other disk management tasks (e.g. mmdeldisk or mmrestripefs) have to perform the same time of scan, and would exhibit similar performance issues. The best way to speed up metadata scans is to improve the capability of disks holding metadata.

    Note that mmfileid is not a command that requires an outage window to run. It can be run in online mode. Of course, if the throughput of the disks holding metadata (and their servers, in the NSD serving scenario) are a performance bottleneck, the extra metadata IO load generated by mmfileid may have an impact on other workloads. As with other long-running mm commands, this can be controlled, with coarse granularity, by varying the set of participating worker nodes via the -N parameter (the fewer nodes are participating, the less impact on the overall cluster). In the future, we may be able to provide ways to control the amount of IO bandwidth consumed by admin commands with finer granularity.

    Since BROKEN disk addresses can only be introduced as the result of mmfsck and mmdeldisk -p, it would make sense to identify the affected files at that time, as opposed to doing a separate metadata scan later. A work item to implement this is currently under consideration. If there's mutual agreement, we can use this RFE to track it.

  • Guest
    Reply
    |
    Sep 19, 2014

    Due to processing by IBM, this request was reassigned to have the following updated attributes:
    Brand - Servers and Systems Software
    Product family - General Parallel File System
    Product - GPFS
    Component - V3 Product functionality
    Operating system - Linux

    For recording keeping, the previous attributes were:
    Brand - Servers and Systems Software
    Product family - General Parallel File System
    Product - GPFS
    Component - V3
    Operating system - Linux

  • Guest
    Reply
    |
    Jul 19, 2014

    Creating a new RFE based on Community RFE #56292 in product GPFS.