Rollover strategies

Rollover strategies control manual rollover of indices or data streams. Rollover creates a new backing index when conditions are met.

IRolloverStrategy

		public interface IRolloverStrategy
{
    Task<bool> RolloverAsync(RolloverContext context, CancellationToken ctx = default);
    bool Rollover(RolloverContext context);
}
		
	

Built-in strategies

NoRolloverStrategy

No-op. Always returns true without performing any rollover.

ManualRolloverStrategy

Calls POST {target}/_rollover with optional conditions. To use it, compose a full IngestStrategy<T> that includes the rollover strategy:

		var strategy = new IngestStrategy<LogEntry>(
    LoggingContext.LogEntry.Context,
    BootstrapStrategies.DataStream(),
    new DataStreamIngestStrategy<LogEntry>("logs-myapp-production", "/_bulk"),
    new AlwaysCreateProvisioning(),
    new NoAliasStrategy(),
    new ManualRolloverStrategy()
);
var options = new IngestChannelOptions<LogEntry>(transport, strategy,
    LoggingContext.LogEntry.Context);
using var channel = new IngestChannel<LogEntry>(options);

// Rollover with conditions
await channel.RolloverAsync(maxAge: "7d", maxSize: "50gb");

// Rollover with max document count
await channel.RolloverAsync(maxDocs: 10_000_000);

// Unconditional rollover
await channel.RolloverAsync();
		
	

Rollover conditions

Condition	Description
`maxAge`	Maximum age of the index (e.g. "7d", "30d")
`maxSize`	Maximum primary shard size (e.g. "50gb")
`maxDocs`	Maximum number of documents

When no conditions are specified, rollover is unconditional.

When to use

Manual rollover is useful when:

ILM is not available (e.g. self-managed clusters without ILM)
You need programmatic control over when rollover happens
You want to trigger rollover based on application-level events