A GlusterFS replicate volume can hide split-brain until an application reads a conflicted path or a heal attempt stalls. Checking the split-brain list before choosing a winner tells the operator which paths or GFIDs need manual resolution and which bricks reported them.
GlusterFS reports the split-brain subset through gluster volume heal <VOLNAME> info split-brain. The broader heal info output can also tag entries with Is in split-brain, but the split-brain-only form keeps the check focused on conflicts that automatic self-heal cannot safely choose.
Run the check from a node in the trusted pool against the volume that owns the replica set. A clean result shows zero split-brain entries for each brick; a nonzero result should be recorded and resolved only after the correct replica copy is identified.
Steps to check for split-brain in GlusterFS:
- List available volumes to confirm the target volume name.
$ sudo gluster volume list volume1
- Confirm the target volume is a started replicate volume.
$ sudo gluster volume info volume1 Volume Name: volume1 Type: Replicate Status: Started Number of Bricks: 1 x 2 = 2 Transport-type: tcp Bricks: Brick1: node1:/srv/gluster/brick1 Brick2: node2:/srv/gluster/brick2 Options Reconfigured: cluster.self-heal-daemon: enable
Split-brain checks apply to replicate, distributed-replicate, and arbiter layouts where Automatic File Replication (AFR) tracks replica state.
- List split-brain entries for the target volume.
$ sudo gluster volume heal volume1 info split-brain Brick node1:/srv/gluster/brick1 /dir/file1 /dir/file4 Number of entries in split-brain: 2 Brick node2:/srv/gluster/brick2 /dir/file1 /dir/file4 Number of entries in split-brain: 2
No split-brain entries are indicated by Number of entries in split-brain: 0 on each brick.
- Record the listed paths, GFID strings, parent directories, and reporting bricks before attempting resolution.
GFID split-brain can appear as a gfid: entry or as the parent directory of the affected file. Resolution selects one replica copy as authoritative and can permanently overwrite the other copies.
Mohd Shakir Zakaria is a cloud architect with deep roots in software development and open-source advocacy. Certified in AWS, Red Hat, VMware, ITIL, and Linux, he specializes in designing and managing robust cloud and on-premises infrastructures.