pycassaShell is an interactive Cassandra python shell. It is useful for exploring Cassandra, especially for those who are just beginning.


Python 2.6 or later is required.

Make sure you have pycassa installed as shown in Installing.

It is strongly recommended that you have IPython, an enhanced interactive python shell, installed. This gives you tab completion, colors, and working arrow keys!

On Debian based systems, this can be installed by:

apt-get install ipython

Alternatively, if easy_install is available:

easy_install ipython


pycassaShell -k KEYSPACE [OPTIONS]

The available options are:

  • -H, --host - The hostname to connect to. Defaults to ‘localhost’
  • -p, --port - The port to connect to. Defaults to 9160.
  • -u, --user - If authentication or authorization are enabled, this username is used.
  • -P, --passwd - If authentication or authorization are enabled, this password is used.
  • -S, --streaming - Use a streaming transport. Works with Cassandra 0.6.x and below.
  • -F, --framed - Use a framed transport. Works with Cassandra 0.7.x. This is the default.

When pycassaShell starts, it creates a ColumnFamily for every existing column family and prints the names of the objects. You can use these to easily insert and retrieve data from Cassandra.

>>> STANDARD1.insert('key', {'colname': 'val'})
>>> STANDARD1.get('key')
{'colname': 'val'}

If you are interested in the keyspace and column family definitions, pycassa provides several methods that can be used with SYSTEM_MANAGER:

Example usage:

>>> describe_keyspace('Keyspace1')

Name:                                Keyspace1

Replication Strategy:                SimpleStrategy
Replication Factor:                  1

Column Families:

>>> describe_column_family('Keyspace1', 'Indexed1')

Name:                                Indexed1
Column Type:                         Standard

Comparator Type:                     BytesType
Default Validation Class:            BytesType

Cache Sizes
  Row Cache:                         Disabled
  Key Cache:                         200000 keys

Read Repair Chance:                  100.0%

GC Grace Seconds:                    864000

Compaction Thresholds
  Min:                               4
  Max:                               32

Memtable Flush After Thresholds
  Throughput:                        63 MiB
  Operations:                        295312 operations
  Time:                              60 minutes

Cache Save Periods
  Row Cache:                         Disabled
  Key Cache:                         3600 seconds

Column Metadata
  - Name:                            birthdate
    Value Type:                      LongType
    Index Type:                      KEYS
    Index Name:                      None

>>> SYSTEM_MANAGER.create_keyspace('Keyspace1', strategy_options={"replication_factor": "1"})
>>> SYSTEM_MANAGER.create_column_family('Keyspace1', 'Users', comparator_type=INT_TYPE)
>>> SYSTEM_MANAGER.alter_column_family('Keyspace1', 'Users', key_cache_size=100)
>>> SYSTEM_MANAGER.create_index('Keyspace1', 'Users', 'birthdate', LONG_TYPE, index_name='bday_index')
>>> SYSTEM_MANAGER.drop_keyspace('Keyspace1')

Table Of Contents

Previous topic

Version 1 UUIDs (TimeUUIDType)

Next topic

Using pycassa with Other Tools

This Page