Make sure to keep some snapshots around even if they stop coming.

common/nfs-services-standby.nix

@@ -42,14 +42,24 @@ in
     # Cleanup old snapshots on standby (keep last 4 hours for HA failover)
     systemd.services.cleanup-services-standby-snapshots = {
       description = "Cleanup old btrfs snapshots in services-standby";
-      path = [ pkgs.btrfs-progs pkgs.findutils ];
+      path = [ pkgs.btrfs-progs pkgs.findutils pkgs.coreutils ];
-
       script = ''
         set -euo pipefail
-        # Keep last 4 hours of snapshots (48 snapshots at 5min intervals)
-        find /persist/services-standby -maxdepth 1 -name 'services@*' -mmin +240 -exec btrfs subvolume delete {} \; || true
-      '';
 
+        # Keep at least 2 hours of snapshots (24 snapshots at 5min intervals)
+        MIN_KEEP=24
+
+        # Count existing snapshots
+        count=$(find /persist/services-standby -maxdepth 1 -name 'services@*' -type d | wc -l)
+
+        # Only delete old snapshots if we have more than the minimum
+        if [ $count -gt $MIN_KEEP ]; then
+          # Delete snapshots older than 4 hours
+          find /persist/services-standby -maxdepth 1 -name 'services@*' -mmin +240 -exec btrfs subvolume delete {} \; || true
+        else
+          echo "Only $count snapshots found, keeping all (minimum: $MIN_KEEP)"
+        fi
+      '';
       serviceConfig = {
         Type = "oneshot";
         User = "root";