How to stop a GlusterFS geo-replication session

Stopping a GlusterFS geo-replication session is useful for maintenance windows, planned cutovers, or preventing accidental changes from propagating to a disaster-recovery site. A stopped session keeps the relationship defined but pauses the data synchronization pipeline so the secondary volume stops receiving updates from the primary.

Geo-replication runs asynchronous replication by spawning worker processes (commonly gsyncd) that scan the primary volume and ship changes over SSH to the secondary volume. Running the gluster volume geo-replication … stop command terminates those workers for the selected session while leaving the session configuration in place, enabling a later restart without recreating the pairing.

After a stop, the secondary side becomes progressively out of date as new writes accumulate on the primary, so record the stop time and the last synchronized timestamp for recovery planning. Ensure the session being stopped matches the intended secondary host and volume (user@host::volume) before proceeding, especially when multiple sessions exist for the same primary volume.

1. Stop the geo-replication session on the primary cluster.

   $ sudo gluster volume geo-replication gvol-primary geoaccount@snode1.example.com::gvol-secondary stop
   geo-replication: stop: success

   Stopping replication pauses updates to the secondary volume.

2. Verify the session reports a Stopped state.

   $ sudo gluster volume geo-replication gvol-primary geoaccount@snode1.example.com::gvol-secondary status
   PRIMARY NODE    PRIMARY VOL    PRIMARY BRICK    SECONDARY USER    SECONDARY               SECONDARY NODE    STATUS    CRAWL STATUS       LAST_SYNCED
   mnode1          gvol-primary    /bricks/b1      geoaccount        snode1.example.com::gvol-secondary  snode1            Stopped   N/A                2025-01-10 12:05:14

   Related: How to check GlusterFS geo-replication status