cassandra

storage-cassandra

This CQL-based Cassandra 2.2+ storage component includes a GuavaSpanStore and GuavaSpanConsumer. GuavaSpanStore.getDependencies() returns pre-aggregated dependency links (ex via zipkin-dependencies-spark).

The implementation uses the Datastax Java Driver 3.x.

zipkin.storage.cassandra.CassandraStorage.Builder includes defaults that will operate against a local Cassandra installation.

Logging

Queries are logged to the category "com.datastax.driver.core.QueryLogger" when debug or trace is enabled via SLF4J. Trace level includes bound values.

See Logging Query Latencies for more details.

Testing

This module conditionally runs integration tests against a local Cassandra instance.

Tests are configured to automatically access Cassandra started with its defaults. To ensure tests execute, download a Cassandra 2.2-3.4 distribution, extract it, and run bin/cassandra.

If you run tests via Maven or otherwise when Cassandra is not running, you'll notice tests are silently skipped.

Results :

Tests run: 62, Failures: 0, Errors: 0, Skipped: 48

This behaviour is intentional: We don't want to burden developers with installing and running all storage options to test unrelated change. That said, all integration tests run on pull request via Travis.

Tuning

This component is tuned to help reduce the size of indexes needed to perform query operations. The most important aspects are described below. See CassandraStorage for details.

Service and span name indexing

Redundant requests to store service or span names are ignored for an hour to reduce load.

Trace indexing

Indexing of traces are optimized by default. This reduces writes to Cassandra at the cost of memory needed to cache state. This cache is tunable based on your typical trace duration and span count.

Core annotations, ex "sr", are not written to annotations_index, as they aren't intended for use in user queries. Also, binary annotation values longer than 256 characters are not indexed. These optimizations significantly limit writes per trace.

Over-fetching on Trace indexes

User-supplied query limits are over-fetched according to a configured index fetch multiplier in attempts to mitigate redundant data returned from index queries.

Cassandra 2.1

While not supported, here are some notes if you are running the original schema on Cassandra 2.1.

GET /api/v1/traces without the serviceName query will fail

The zipkin-ui always specifies a service name, but the http api permits leaving it out. This won't work in Cassandra 2.1

Disable CASSANDRA_ENSURE_SCHEMA

Upgrades assume features that are present in Cassandra 2.2+. Do not have zipkin upgrade your schema if you are running Cassandra 2.1.

If using zipkin-server, set CASSANDRA_ENSURE_SCHEMA=false.

Manually running below will add default ttls to the original schema.

ALTER TABLE zipkin.traces WITH default_time_to_live = 604800;
ALTER TABLE zipkin.service_span_name_index WITH default_time_to_live = 259200;
ALTER TABLE zipkin.service_name_index WITH default_time_to_live = 259200;
ALTER TABLE zipkin.span_names WITH default_time_to_live = 259200;
ALTER TABLE zipkin.annotations_index WITH default_time_to_live = 259200;
ALTER TABLE zipkin.dependencies WITH default_time_to_live = 259200;
ALTER TABLE zipkin.service_names WITH default_time_to_live = 259200;