# Directory Leasing Feature - Detailed Design Document

## 1. Overview

Directory leasing extends the SMB3 lease concept to directories, enabling client-side caching of directory metadata and change notifications. This significantly improves performance for applications that frequently enumerate directories or monitor directory changes.

## 2. Protocol Specification Reference

- **MS-SMB2 Section 2.2.13.2.12**: SMB2_CREATE_REQUEST_LEASE_V2 for directories

   * specific length.
   *
   * <p>For example, if you need 1024-bit hash codes, you could join two {@link Hashing#sha512} hash
   * functions together: {@code Hashing.concatenating(Hashing.sha512(), Hashing.sha512())}.
   *
   * @since 19.0
   */
  public static HashFunction concatenating(
      HashFunction first, HashFunction second, HashFunction... rest) {

const (
	xMinIOHealing = ReservedMetadataPrefix + "healing"
	xMinIODataMov = ReservedMetadataPrefix + "data-mov"
)

// SetHealing marks object (version) as being healed.
// Note: this is to be used only from healObject
func (fi *FileInfo) SetHealing() {
	if fi.Metadata == nil {
		fi.Metadata = make(map[string]string)
	}
	fi.Metadata[xMinIOHealing] = "true"
}

            assertEquals(0x1, Smb2Constants.SMB2_GLOBAL_CAP_DFS, "DFS capability must be 0x1");
        }

        @Test
        @DisplayName("Leasing capability should be 0x2")
        void testGlobalCapLeasing() {
            assertEquals(0x2, Smb2Constants.SMB2_GLOBAL_CAP_LEASING, "Leasing capability must be 0x2");
        }

        @Test
        @DisplayName("Large MTU capability should be 0x4")

     * @return lease key
     */
    public Smb2LeaseKey requestDirectoryLease(String directoryPath, int requestedState, DirectoryCacheScope scope) {
        // Directory leasing requires SMB 3.0 or higher
        // MS-SMB2: Level 2 leasing (which includes directory leasing) is only supported in SMB 3.0+
        // We'll validate this when we actually need to use the session

        // Request base lease

In GitHub, the template will guide you to write the right question so that you can more easily get a good answer, or even solve the problem yourself even before asking. And in GitHub I can make sure I always answer everything, even if it takes some time. I can't personally do that with the chat systems. 😅

// inspection to understand the root cause. E.g, this could be due to
// backend filesystem corruption.

// listOnlineDisks - returns
// - a slice of disks where disk having 'older' xl.meta (or nothing)
// are set to nil.
// - latest (in time) of the maximally occurring modTime(s), which has at least quorum occurrences.

Klaus Post <******@****.***> 1756435188 +0200

 * <p>The primary way to provide the data that your hash function should act on is via a {@link
 * Hasher}. Obtain a new hasher from the hash function using {@link #newHasher}, "push" the relevant
 * data into it using methods like {@link Hasher#putBytes(byte[])}, and finally ask for the {@code
 * HashCode} when finished using {@link Hasher#hash}. (See an {@linkplain #newHasher example} of
 * this.)
 *

 * <p>The primary way to provide the data that your hash function should act on is via a {@link
 * Hasher}. Obtain a new hasher from the hash function using {@link #newHasher}, "push" the relevant
 * data into it using methods like {@link Hasher#putBytes(byte[])}, and finally ask for the {@code
 * HashCode} when finished using {@link Hasher#hash}. (See an {@linkplain #newHasher example} of
 * this.)
 *