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
Post an idea.
Get feedback from the IBM team and other customers to refine your idea.
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.
Delivered in 5.1.1
IIRC this is implemented now in the latest 5.1.0 PTF. For older version you can achieve the same like this:
1. add this snippet to the service that needs to wait for gpfs
$ cat /etc/systemd/system/ctdb.service.d/gpfs.conf
[Unit]
Wants=gpfs-wait-local-mounts-sc.service
After=gpfs-wait-local-mounts-sc.service
2. add the waiter service:
$ Cat /etc/systemd/system/gpfs-wait-local-mounts-sc.service
[Unit]
After=gpfs.service
PartOf=gpfs.service
Wants=gpfs.service
[Service]
ExecStart=/usr/local/sc-gpfs/sbin/wait-for-mounts.sh all_local
TimeoutStartSec=200
Type=oneshot
RemainAfterExit=yes
[Install]
WantedBy=gpfs.service
3. write script that waits for the given FS(s), e.g. using findmnt
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) - Public 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) - Public RFEs
Component - Technical Foundation
We have GPFS and Spectrum Archive deployed on ships with no regular IT personnel so automation is very important for us and as it stands now, both systems must be started manually after a reboot.
We also use a systemd setup (in our Ubuntu based cluster) with GPFS and SLURM.
We don't see the problem very often, probably because we have only about 70 nodes, so
GPFS is normally mounted when it comes to SLURM.
But of course one should be able to rely on systemd service configuration, and
not only having it working by chance.
Best regards, Ralf
This is from a ticket I had with IBM:
This is not about RedHat. It is about systemd, which is used by many Linux distributions. And you deliver a systemd service with your product. It is just that this very service only does half the job. Its job is to start the service (which it does reliably) _and_ to return once the service is up and useable (which it does not). I have no way of configuring any other systemd service that requires a fully functional Spectrum Scale by basic systemd configuration (e.g. adding a After statement to a service). Just because it returns too early.
You should design your system stuff similar to the NetworkManager setup: There's a service that starts up the interfaces. And there's a second service which is only there to be used for dependency configurations. It is called NetworkManager-wait-online.service which waits until all interfaces are fully usable (have an IP).
For Spectrum Scale I think several additional services should be done: One that can be waited on if you just want to have the basic system running (without automatically mounted filesystems), one that waits for all local file systems to be mounted, one that also includes remote file systems and one mount service for each file system (just like current RedHat systems already offer (but they are not working for Spectrum Scale file systems): net-gpfsm-fs1.mount) so that you can wait for the mount of a a specific filesystem.
For systemd there's a mechanism called sd_notify. I image this could be used by a suitably designed callback from Spectrum Scale to signal the systemd that Spectrum Scale ist up.
This should work both with automount enabled and disabled.