Shard Management
Shard Management
The Shard Management feature provides a visual, grid-based interface for managing Elasticsearch shard allocation and relocation.
Visual Shard Grid
Secan displays shards in an interactive grid where:
- Rows represent Elasticsearch nodes
- Columns represent indices (10 per page with pagination)
- Color coding indicates shard status (primary, replica, unallocated)
Grid Layout
The grid shows all nodes as rows and paginates indices across columns:
- Fixed Columns: Always displays 10 indices per page
- Pagination: Navigate between index pages using the pagination controls
- Node Column: Sticky left column showing node names (180px width)
- Unassigned Row: Special row at the top showing unassigned shards
Shard Status Colors
Shards are color-coded to show status:
- Blue: Primary shard (allocated)
- Green: Replica shard (allocated)
- Gray: Unallocated shard (waiting for assignment)
- Red: Unallocated primary (cluster health affected)
Shard Operations
Shard Relocation
Relocate shards between nodes using the interactive grid:
- Right-click on a shard to open the context menu
- Select "Select for Relocation"
- Destination nodes will be highlighted in the grid
- Click on a destination node to initiate the move
- Confirm the relocation in the modal dialog
The system validates:
- Destination node has the data role
- Destination doesn't already have a copy of this shard
- Sufficient disk space is available
Relocating Shard Indicators
Shards that are currently relocating show a special visual indicator:
- Yellow dot overlay on the shard cell
- Tooltip showing source and destination nodes
- Real-time updates as the relocation progresses
You can filter the view to show only relocating shards using the filter controls.
Shard Context Menu
Right-click any shard to access:
- View Stats: Open shard statistics modal
- Select for Relocation: Start the relocation workflow
- View Index Details: Navigate to the index details page
Shard Status
Shards are color-coded to show status:
- Blue: Primary shard (allocated)
- Green: Replica shard (allocated)
- Gray: Unallocated shard (waiting for assignment)
- Red: Unallocated primary (cluster health affected)
Use Cases
Rebalancing
After node additions or removals, rebalance shards across the cluster for optimal distribution.
Node Maintenance
Before taking a node offline, relocate its shards to other nodes using this interface.
Performance Optimization
Move shards based on query patterns or resource usage to optimize performance.
Important Notes
- Shard relocation consumes I/O resources; plan carefully
- Use during off-peak hours for large relocations
- Monitor cluster health during operations
- Elasticsearch throttles relocations to avoid overwhelming the cluster
Related Features
- Cluster Details - View cluster and node status
- Index Management - Manage indices
- REST Console - Execute shard allocation APIs directly