Parse Redis dump.rdb files, Analyze Memory, and Export Data to JSON

Rdbtools is a parser for Redis' dump.rdb files. The parser generates events similar to an xml sax parser, and is very efficient memory wise.

In addition, rdbtools provides utilities to :

Generate a Memory Report of your data across all databases and keys
Convert dump files to JSON
Compare two dump files using standard diff tools

Rdbtools is written in Python, though there are similar projects in other languages. See FAQs for more information.

See https://rdbtools.com for a gui to administer redis, commercial support, and other enterprise features.

Installing rdbtools

Pre-Requisites :

python-lzf is optional but highly recommended to speed up parsing.
redis-py is optional and only needed to run test cases.

To install from PyPI (recommended) :

pip install rdbtools python-lzf

To install from source :

git clone https://github.com/sripathikrishnan/redis-rdb-tools
cd redis-rdb-tools
sudo python setup.py install

Command line usage examples

Every run of RDB Tool requires to specify a command to indicate what should be done with the parsed RDB data. Valid commands are: json, diff, justkeys, justkeyvals and protocol.

JSON from a two database dump:

> rdb --command json /var/redis/6379/dump.rdb

[{
"user003":{"fname":"Ron","sname":"Bumquist"},
"lizards":["Bush anole","Jackson's chameleon","Komodo dragon","Ground agama","Bearded dragon"],
"user001":{"fname":"Raoul","sname":"Duke"},
"user002":{"fname":"Gonzo","sname":"Dr"},
"user_list":["user003","user002","user001"]},{
"baloon":{"helium":"birthdays","medical":"angioplasty","weather":"meteorology"},
"armadillo":["chacoan naked-tailed","giant","Andean hairy","nine-banded","pink fairy"],
"aroma":{"pungent":"vinegar","putrid":"rotten eggs","floral":"roses"}}]

Filter parsed output

Only process keys that match the regex, and only print key and values:

> rdb --command justkeyvals --key "user.*" /var/redis/6379/dump.rdb

user003 fname Ron,sname Bumquist,
user001 fname Raoul,sname Duke,
user002 fname Gonzo,sname Dr,
user_list user003,user002,user001

Only process hashes starting with "a", in database 2:

> rdb -c json --db 2 --type hash --key "a.*" /var/redis/6379/dump.rdb

[{},{
"aroma":{"pungent":"vinegar","putrid":"rotten eggs","floral":"roses"}}]

Converting dump files to JSON

The json command output is UTF-8 encoded JSON. By default, the callback try to parse RDB data using UTF-8 and escape non 'ASCII printable' characters with the \U notation, or non UTF-8 parsable bytes with \x. Attempting to decode RDB data can lead to binary data curroption, this can be avoided by using the --escape raw option. Another option, is to use -e base64 for Base64 encoding of binary data.

Parse the dump file and print the JSON on standard output:

> rdb -c json /var/redis/6379/dump.rdb

[{
"Citat":["B\u00e4ttre sent \u00e4n aldrig","Bra karl reder sig sj\u00e4lv","Man ska inte k\u00f6pa grisen i s\u00e4cken"],
"bin_data":"\\xFE\u0000\u00e2\\xF2"}]

Parse the dump file to raw bytes and print the JSON on standard output:

> rdb -c json /var/redis/6379/dump.rdb --escape raw

[{
"Citat":["B\u00c3\u00a4ttre sent \u00c3\u00a4n aldrig","Bra karl reder sig sj\u00c3\u00a4lv","Man ska inte k\u00c3\u00b6pa grisen i s\u00c3\u00a4cken"],
"bin_data":"\u00fe\u0000\u00c3\u00a2\u00f2"}]

Generate Memory Report

Running with the -c memory generates a CSV report with the approximate memory used by that key. --bytes C and '--largest N can be used to limit output to keys larger than C bytes, or the N largest keys.

> rdb -c memory /var/redis/6379/dump.rdb --bytes 128 -f memory.csv
> cat memory.csv

database,type,key,size_in_bytes,encoding,num_elements,len_largest_element
0,list,lizards,241,quicklist,5,19
0,list,user_list,190,quicklist,3,7
2,hash,baloon,138,ziplist,3,11
2,list,armadillo,231,quicklist,5,20
2,hash,aroma,129,ziplist,3,11

The generated CSV has the following columns - Database Number, Data Type, Key, Memory Used in bytes and RDB Encoding type. Memory usage includes the key, the value and any other overheads.

Note that the memory usage is approximate. In general, the actual memory used will be slightly higher than what is reported.

You can filter the report on keys or database number or data type.

The memory report should help you detect memory leaks caused by your application logic. It will also help you optimize Redis memory usage.

Find Memory used by a Single Key

Sometimes you just want to find the memory used by a particular key, and running the entire memory report on the dump file is time consuming.

In such cases, you can use the redis-memory-for-key command:

> redis-memory-for-key person:1

> redis-memory-for-key -s localhost -p 6379 -a mypassword person:1

Key             person:1
Bytes               111
Type                hash
Encoding            ziplist
Number of Elements      2
Length of Largest Element   8

NOTE :

This was added to redis-rdb-tools version 0.1.3
This command depends redis-py package

Comparing RDB files

First, use the --command diff option, and pipe the output to standard sort utility

> rdb --command diff /var/redis/6379/dump1.rdb | sort > dump1.txt
> rdb --command diff /var/redis/6379/dump2.rdb | sort > dump2.txt

Then, run your favourite diff program

> kdiff3 dump1.txt dump2.txt

To limit the size of the files, you can filter on keys using the --key option

Emitting Redis Protocol

You can convert RDB file into a stream of redis protocol using the protocol command.

> rdb -c protocol /var/redis/6379/dump.rdb

*4
$4
HSET
$9
users:123
$9
firstname
$8
Sripathi

You can pipe the output to netcat and re-import a subset of the data. For example, if you want to shard your data into two redis instances, you can use the --key flag to select a subset of data, and then pipe the output to a running redis instance to load that data. Read Redis Mass Insert for more information on this.

When printing protocol output, the --escape option can be used with printable or utf8 to avoid non printable/control characters.

By default, expire times are emitted verbatim if they are present in the rdb file, causing all keys that expire in the past to be removed. If this behaviour is unwanted the -x/--no-expire option will ignore all key expiry commands.

Otherwise you may want to set an expiry time in the future with -a/--amend-expire option which adds an integer number of seconds to the expiry time of each key which is already set to expire. This will not change keys that do not already have an expiry set.

Using the Parser

from rdbtools import RdbParser, RdbCallback
from rdbtools.encodehelpers import bytes_to_unicode

class MyCallback(RdbCallback):
    ''' Simple example to show how callback works.
        See RdbCallback for all available callback methods.
        See JsonCallback for a concrete example
    '''

    def __init__(self):
        super(MyCallback, self).__init__(string_escape=None)

    def encode_key(self, key):
        return bytes_to_unicode(key, self._escape, skip_printable=True)

    def encode_value(self, val):
        return bytes_to_unicode(val, self._escape)

    def set(self, key, value, expiry, info):
        print('%s = %s' % (self.encode_key(key), self.encode_value(value)))

    def hset(self, key, field, value):
        print('%s.%s = %s' % (self.encode_key(key), self.encode_key(field), self.encode_value(value)))

    def sadd(self, key, member):
        print('%s has {%s}' % (self.encode_key(key), self.encode_value(member)))

    def rpush(self, key, value):
        print('%s has [%s]' % (self.encode_key(key), self.encode_value(value)))

    def zadd(self, key, score, member):
        print('%s has {%s : %s}' % (str(key), str(member), str(score)))

callback = MyCallback()
parser = RdbParser(callback)
parser.parse('/var/redis/6379/dump.rdb')

Other Pages

Frequently Asked Questions
Redis Dump File Specification
Redis Dump File Version History - this also has notes on converting a dump file to an older version.

sripathikrishnan / redis-rdb-tools

readme