DB Cassandra Module

Anca Vamanu

Boudewyn Ligthart

Edited by

Anca Vamanu


Table of Contents

1. Admin Guide
1. Overview
2. Dependencies
2.1. SIP Router Modules
2.2. External Libraries or Applications
3. Parameters
3.1. schema_path (string)
4. Functions
5. Installation
6. Table schema
7. Limitations

List of Examples

1.1. Set schema_path parameter

Chapter 1. Admin Guide

1. Overview

Db_cassandra is one of the Kamailio database modules. It does not export any functions executable from the configuration scripts, but it exports a subset of functions using the database API, and thus, other modules can use it as a database driver, instead of, for example, the Mysql module.

The storage backend is a Cassandra cluster and this module provides an SQL interface to be used by other modules for storing and retrieving data. Because Cassandra is a NoSQL distributed system, there are limitations on the operations that can be performed. The limitations concern the indexes on which queries are performed, as it is only possible to have simple conditions (equality comparison only) and only two indexing levels. These issues will be explained in an example below.

Cassandra DB is especially suited for storing large amounts of data or data that requires distribution, redundancy or replication. One usage example is a distributed location system in a platform that has a cluster of SIP Router servers, with several proxies and registration servers accessing the same location database. This was actually the main use case we had in mind when implementing this module. Please NOTE that it has only been tested with the usrloc, auth_db and domain modules.

You can find a configuration file example for this usage in the module - kamailio_cassa.cfg.

Because the module has to do the translation from SQL to Cassandra NoSQL queries, the schemas for the tables must be known by the module. You will find the schemas for location, subscriber and version tables in utils/kamctl/dbcassandra directory. You have to provide the path to the directory containing the table definitions by setting the module parameter schema_path.

There is no need to configure a table metadata in Cassandra cluster. You only need to define a keyspace with the name of the database and for each table a column family inside that keyspace with the name of the table. The comparator and validators should be either UTF8Type or ASCIIType. Example:

   ...
   create keyspace kamailio;
   use kamailio;
   create column family 'location' with comparator='UTF8Type' and 
default_validation_class='UTF8Type' and key_validation_class='UTF8Type';
   ...

Special attention was given to performance in Cassandra. Therefore, the implementation uses only the native row indexing in Cassandra and no secondary indexes, because they are costly. Instead, we simulate a secondary index by using the column names and putting information in them, which is very efficient. Also, for deleting expired records, we let Cassandra take care of this with its own mechanism (by setting the TTL for columns).

The module supports raw queries. However these queries must follow the CQL (Cassandra Query Language) syntax. The queries can be issued in the script by means of the AVPOPS module. Keep in mind that when passing back the results from the database only the first row is used to set the AVP variables. (default AVPOPS behaviour) The script lines below can be used as an example for issuing the query towards an cassandra instance.(This example will work once the column family `location` is configured correctly in the cassandra keyspace)

   ...
   $var(dballowed)="select * from location where key = 'userx' limit 1;";
   avp_db_query("$var(dballowed)");
   xlog("L_INFO","Got result here: [$avp(i:1)] [$avp(i:2)] [$avp(i:3)].\n");
   ...

2. Dependencies

2.1. SIP Router Modules

The following modules must be loaded before this module:

  • No dependencies on other SIP Router modules.

2.2. External Libraries or Applications

The following libraries or applications must be installed before running SIP Router with this module loaded:

  • Thrift library (tested with version 0.6.1 and version 0.7.0). You can download it from http://archive.apache.org/dist/thrift .

The implementation was tested with Cassandra version 1.0.1 and version 1.1.6. I used the sourced from DataStax Community Edition (http://www.datastax.com/download/community).

3. Parameters

3.1. schema_path (string)

The directory where the files with the table schemas are located. This directory has to contain the subdirectories corresponding to the database name (name of the directory = name of the database). These directories, in turn, contain the files with the table schemas. See the schemas in utils/kamctl/dbcassandra directory.

Example 1.1. Set schema_path parameter

   ...
   modparam("db_cassandra", "schema_path",
               "/usr/local/kamailio/etc/kamctl/dbcassandra")
   ...

4. Functions

NONE

5. Installation

Because the db_cassandra module depends on an external library, it is not compiled and installed by default. You can use one of these options:

  • - edit the "Makefile" and remove "db_cassandra" from "excluded_modules" list. Then follow the standard procedure to install SIP Router: "make all; make install".

  • - from command line, run: 'make all include_modules="db_cassandra"; make install include_modules="db_cassandra"'.

6. Table schema

The module must know the table schema for the tables that will be used. You must configure the path to the schema directory by setting the schema_path parameter.

A table schema document has the following structure:

  • First row: the name and type of the columns in the form name(type) separated by spaces. The possible types are: string, int, double and timestamp.

    Thetimestamp type has a special meaning. Only one column of this type can be defined for a table, and it should contain the expiry time for that record. If defined this value will be used to compute the ttl for the columns and Cassandra will automatically delete the columns when they expire. Because we want the ttl to have meaning for the entire record, we must ensure that when the ttl is updated, it is updated for all columns for that record. In other words, to update the expiration time of a record, an insert operation must be performed from the point of view of the db_cassandra module ("insert" in Cassandra means "replace if exists or insert new record otherwise"). So, if you define a table with a timestamp column, the update operations on that table that also update the timestamp must update all columns. So, these update operations must in fact be insert operations.

  • Second row: the columns that form the row key separated by space.

  • Third row: the columns that form the secondary key separated by space.

Below you can see the schema for the location table (when use_domain not set):

   ...
   callid(string) cflags(int) contact(string) cseq(int) expires(timestamp) flags(int) last_modified(int) methods(int) path(string) q(double) received(string) socket(string) user_agent(string) username(string) ruid(string) instance(string) reg_id(int)
   username
   contact
   ...

Observe first that the row key is the username and the secondary index is the contact. We have also defined a timestamp column - expires.

If you need to use the domain part of the AOR also (you have set use_domain parameter for usrloc in the script), you should include the domain column in the list of columns and in the primary key. The schema will then look like this:

   ...
   callid(string) cflags(int) contact(string) cseq(int) domain(string) expires(timestamp) flags(int) last_modified(int) methods(int) path(string) q(double) received(string) socket(string) user_agent(string) username(string) ruid(string) instance(string) reg_id(int)
   username domain
   contact
   ...

Notice that a key (primary or secondary) can be composed from more columns, in which case you have to specify them separated by space.

To understand why the schema looks like this, we must first see which queries are performed on the location table. (The 'callid' condition was ignored as it doesn't really have a well defined role in the SIP RFC).

  • When Invite received, lookup location: select where username='..'.

  • When Register received, update registration: update where username='..' and contact='..'.

So, the relation between these keys is the following:

  • The unique key for a table is actually the combination of row key + secondary key.

  • A row defined by a row key will contain more records with different secondary keys.

The timestamp column that leaves the Cassandra cluster to deal with deleting the expired records. For this to work right we needed to modify a bit the behavior of usrloc module and replace update sql query performed at re-registration with an insert sql query (so that all columns are updated and the new timestamp is set for all columns). This behavior is enabled by setting a parameter in the usrloc module db_update_as_insert:

   ...
   modparam("usrloc", "db_update_as_insert", 1)
   ...

Also you should disable in usrloc module the timer routine that checks for expired records. You can do this by setting the timer interval to 0. timer_interval:

   ...
   modparam("usrloc", "timer_interval", 0)
   ...

The alternative would have been to define an index on the expire column and run a external job to periodically delete the expired records. However, obviously, this would be more costly.

7. Limitations

The module can be used only when the queries use only one index, which is also the unique key, or have two indexes that form the unique key like in the usrloc usage.