An rdb file begins with "REDIS0003" encoded as a bytestring. [9 bytes]
For each database in the dump
- The first byte is the SELECT_DB opcode (0xfe) [1 byte]
- The second byte is the length-encode database number (See below) [1-4 byte(s)]
- For each key-value pair in the database:
- If the object has an expire time:
- The first byte is the EXPIRETIME opcode (0xfd) or, in Redis 2.6+, EXPIRETIME_MS opcode (0xfc)
- What follows is the expire time in seconds or, in Redis 2.6+, in ms (See below)
- The first byte is the object type (See below for list of object type opcodes)
- Then, the string-encoded key for the object (See below)
- Then, the object itself, specially encoded for its type (See below)
- If the object has an expire time:
An rdb file ends with the EOF opcode (0xff) [1 byte]
======
The first two most-significant bits hold the encoding type
- Less than 64 (00)
- Less than 16,384 (01)
- 32-bit integer (10)
- Special encoding (11)
The remaining bits hold the length (or, in the last case, encoding type) to be encoded (marked below as X's)
- (00) | XXXXXX [1 byte total]
- (01) | XXXXXX XXXXXXXX [2 bytes total]
- (10) | 000000 XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX [5 bytes total]
- (11) | XXXXXX [1 byte total]
The first 1-5 bytes of a string object contains the length of the string object and whether or not it is an encoded string, as described in the length-encoding section above.
If the string is an encoded type with either a 8, 16, or 32-bit encoding type, the string is loaded as an integer, as described above in the integer-loading section, and converted to a string.
If the string is encoded with both bits set, it is loaded as an lzf compressed string, as described below.
If the string is not an encoded type, it is saved as the next [length] bytes.
If a length-encoded byte has its most-significant two bits set, the remaining 6 bytes will contain either a 0,1,2 or 3.
- A zero means the value following the length is encoded as an 8-bit integer.
- A one means the value following the length is encoded as an 16-bit integer.
- A two means the value following the length is encoded as an 32-bit integer.
- A three means the value following the length is encoded as an lzf compressed string.
[13 00 00 00] -> Little-endian 32-bit length (in bytes) of ziplist
[0e 00 00 00] -> little-endian 32-bit offset (in bytes) to the end of the last entry in the list
[02 00] -> little-endian 16-bit number of list entries
[00] -> number of bytes of previous entry
[c0] -> value encoding
[01 00] -> 16-bit value
[04] -> number of bytes of previous entry
[c0] -> value encoding
[01 00] -> 16-bit value
[ff] -> end of ziplist
[02 00 00 00] -> Little-endian 32-bit length encoding, in bytes, of members of the intset
[01 00 00 00] -> Little-endian 32-bit number of elements in the intset
[01 00] -> One little-endian member, two-bytes long, as specified in the encoding above