GlusterFS

KB Product Documentation Data Storage Container Mount Protocols GlusterFS Viewed: 7234
0 0

GlusterFS

GlusterFS is the rockstar of cloud storage network file systems! It’s known for being highly reliable, handling hiccups with ease, and easily scaling to meet your growing needs. What’s really neat is that it operates using this free software interface called FUSE, so you say goodbye to the hassle of setting up extra servers just for metadata – unlike some other clunky systems out there.

NFS is a solid choice, but GlusterFS is like the heavyweight champion of reliability! GlusterFS works across multiple servers, making it perfect for situations where you need tons of users accessing files at once (high concurrency), super-fast write speeds, and automatic recovery if a server failover. If you’re curious about how it all works under the hood, the official GlusterFS documentation has all the details:

Starting from the 6.1 release, the platform offers support for the Gluster Native client, but only on the Shared Storage Cluster.

All containers, except for the alpine-based nodes, are equipped with the Gluster Native client.

Additionally, the Shared Storage Cluster supports both the Gluster Native client and server functionalities.

Here’s a tip: Gluster Native is the way to go if you prioritize data safety and redundancy through backups and replication. However, if performance is your main focus, especially for topologies where speed is key, the NFS protocol might be a better fit.

1. The platform uses a special type of GlusterFS volume called a “replicated volume.” This basically means your data gets copied and stored on multiple servers (like having backups on different computers).

So, even if one server has a meltdown, your data is safe and sound because you’ve got copies on the other servers. You can explore further into how GlusterFS replication operates within the platform by analyzing the corresponding JPS manifest file, which you can find at this link: GlusterFS Replication Logic JPS Manifest

Note: It’s important not to mix up GlusterFS volumes with Docker container volumes available on the platform.

2. The default GlusterFS settings are automatically configured and should not be changed manually. These settings include the volume name (“data”) and the IP addresses of the relevant servers.

3. When sharing data with GlusterFS, keep in mind the following limitations:

  • Shares do not function with sleeping environments and during live migration.
  • Additionally, while custom volumes can be created via SSH, it’s not recommended due to limited functionality, such as scaling automation.

4. If you encounter any issues with GlusterFS, you can check the Tasks Panel in the dashboard for updates. Additionally, you can refer to the following logs for additional information:

/var/log/glusterfs on the GlusterFS server and run the journalctl command on the Shared Storage, custom Docker, or VPS containers.

Refer to the guides on mount points or exporting data to effectively share files using the GlusterFS protocol.