File and Object Storage

 View Only

A Deep Dive into Remote Mount with IBM Storage Scale Ansible Toolkit releases – 5.1.9.0

By Sujeet Jha posted Tue January 30, 2024 06:21 AM

  

Introduction:
In the ever-evolving landscape of storage solutions, the IBM Storage Scale installation toolkit has taken a significant leap forward with its 5.1.9.0 release. In this blog post, I will delve into the enhanced remote mount support offered by the Ansible-based automation tool and explore how it has streamlined the process of mounting filesystems from a storage cluster to a client cluster. The blog will walk you through the history, phases, limitations, and workflow, providing a comprehensive guide for users seeking to leverage this powerful feature.

History:
Before the 5.1.9.0 release, the IBM Storage Scale installation toolkit faced limitations in supporting remote filesystem mounts from storage to client clusters. Users found themselves navigating a cumbersome process with manual steps, making it a challenging feature to utilize effectively.

Phases:
The 5.1.9.0 release brings a game-changing feature to the forefront – the Ansible-toolkit's ability to remotely mount filesystems from a storage cluster to a client cluster. Additionally, the toolkit introduces a feature allowing users to revoke previously configured remote mount filesystems. These advancements mark a significant stride towards a more user-friendly and efficient storage solution.

Limitations:
While the toolkit excels in its capabilities, it's essential to note that remote mount support is currently available exclusively on RHEL and SLES operating systems. Understanding these limitations ensures a smooth integration into environments that align with the supported platforms.

Workflow:
The workflow for implementing remote mount configurations using the IBM Storage Scale installation toolkit is well-defined and user-friendly. The blog provides a step-by-step guide, starting from setting up the installer node, issuing relevant commands, and configuring remote mount user settings. The inclusion of screenshots and command-line examples enhances the clarity of the process.

Remote Mount Configuration Steps:
The blog outlines the necessary steps to configure remote mounts, from downloading and extracting the IBM Spectrum Scale package to setting up the installer node and defining user configurations. Users are guided through the spectrumscale setup command and instructed on using the Ansible-based install toolkit to configure remote mount settings effectively.

# Download and extract the IBM Spectrum Scale self-extracting package on the installer node.
root@test-21:~# /root/Storage_Scale_Advanced-5.1.9.0-x86_64-Linux-install –silent

# Go to the IBM Spectrum Scale installation toolkit directory. 
root@test-21:~# cd /usr/lpp/mmfs/5.1.9.0/ansible-toolkit/

# Issue the spectrumscale setup command to set up the new installer node.
root@test-21:/usr/lpp/mmfs/5.1.9.0/ansible-toolkit# ./spectrumscale setup -s 10.0.100.95

Pasted Graphic 2.png

# Use the Ansible based install toolkit cli config to define remote mount user config. so the toolkit knows what to act upon.

test-31:/usr/lpp/mmfs/5.1.9.0/ansible-toolkit # ./spectrumscale remote_mount config -client-gui-username admiin -client-gui-password -client-gui-hostname mount-22 -storage-gui-username admiin -storage-gui-password -storage-gui-hostname mount-21 -storage-filesystem fs1 -remotemount-path /ibm/fs1 -client-filesystem fs1
[ INFO  ] Setting client_gui_username to admiin

[ INFO  ] Setting client_gui_username to admiin
Enter gui client password:
Repeat  gui client password:
Enter gui storage password:
Repeat  gui storage password:

Enter the secret encryption key:

Repeat the secret encryption key: 

[ INFO  ] Setting client_gui_hostname to mount-22

[ INFO  ] Setting storage_gui_username to admiin

[ INFO  ] Setting storage_gui_hostname to mount-21

[ INFO  ] Setting storage_filesystem to fs1

[ INFO  ] Setting client_filesystem to fs1

[ INFO  ] Setting remotemount_path to /ibm/fs1

Steps to Execute the Configured Remote Mount Filesystem:
The blog offers detailed instructions on running a remote mount precheck to ensure the environment is ready. Users are then guided through the process of executing the configured remote mount filesystem using the spectrumscale remote_mount grant command. Furthermore, the blog covers the essential steps to revoke a remote mount when needed.

# Use the Ansible based install toolkit to run a remote mount precheck and ensure the environment is ready for remote mount.

root@test-21:/usr/lpp/mmfs/5.1.9.0/ansible-toolkit# ./spectrumscale remote_mount precheck 

# Run the remote mount process by using the ./spectrumscale remote_mount grant command. 

root@test-21:/usr/lpp/mmfs/5.1.9.0/ansible-toolkit# ./spectrumscale remote_mount grant 

# Run the remote mount revoke process by using the ./spectrumscale remote_mount revoke command. 
root@test-21:/usr/lpp/mmfs/5.1.9.0/ansible-toolkit# ./spectrumscale remote_mount revoke

Conclusion:
In conclusion, the IBM Storage Scale installation toolkit's extended remote mount support in the 5.1.9.0 release has revolutionized the way filesystems are mounted between clusters. This blog aims to empower users with a comprehensive understanding of the history, phases, limitations, and workflow, enabling them to leverage this feature effectively in their storage environments. Embrace the future of seamless remote mounts with IBM Storage Scale 5.1.9.0.


#softwaredefinedstorage
#IBMStorageScale#SpectrumScale#filesystem

4 comments
35 views

Permalink

Comments

Sujeet Jha
Mon February 19, 2024 11:32 PM

Hello Renar,
Thank you so much for bringing that mistake to my attention! I truly appreciate your feedback. I've now corrected the error in the blog post. Your input helps me improve the quality of my content, and I'm grateful for it. Thanks again for taking the time to let me know.

renar Grunenberg
Fri February 16, 2024 11:07 AM

Hallo Sujeet,
can you go over the cmd-parameter, i want to use your cmds but had no luck because of some wrong name like -client-username should -client-gui-username or storage-username should storage-gui-username. I think it would be help.
I think this are happening because i use 5.1.9.2 here.Thanks

Sujeet Jha
Mon February 05, 2024 03:54 AM

Hi Nils,
Thank you for your valuable feedback and questions! I appreciate your engagement with the content.

Regarding the installation toolkit commands, I'd like to clarify that the commands can be executed on any node where the Spectrum Scale (SE) package is extracted. Whether it's a client or storage node, as long as the necessary components are available, you can run these commands to configure your environment. I'll make sure to emphasize this point in the blog to provide a more accurate representation.

Additionally, I'll enhance the description of the numbers (1-3) in the picture depicting the storage and client clusters to explain that these represent the sequence of steps or actions during the configuration process.

Thank you again for your feedback, and I hope these clarifications address your queries. If you have further questions or suggestions, please feel free to let me know.

Thanks

Nils Haustein
Wed January 31, 2024 04:34 AM

Hi Sujeet, thanks for the explanation, interesting topic. 

I'm a bit confused where the installation toolkit commands are executed, on the client cluster or storage cluster. Especially, the step below, which is executed on test-31, whereas the previous spectrumscale commands are executed on test-21. Is this command executed on the storage cluster or on the client cluster:

# Use the Ansible based install toolkit cli config to define remote mount user config. so the toolkit knows what to act upon.

test-31:/usr/lpp/mmfs/5.1.9.0/ansible-toolkit # ./spectrumscale remote_mount config -client-username admiin -client-gui-hostname mount-22 -storage-username admiin -storage-gui-hostname mount-21 -storage-filesystem fs1 -remotemount-path /ibm/fs1 -client-filesystem fs1

Can you please add some further descriptions where commands are executed?

Furthermore, in the picture showing the storage and client clusters, there are numbers (1-3). What do these numbers mean? Can you elaborate on this in your article?

Thanks