describe how to use scylladb as a storage backend

README.md

@@ -85,7 +85,7 @@ Jaeger can be used with a growing a number of storage backends:
 * It natively supports two popular open source NoSQL databases as trace storage backends: Cassandra and Elasticsearch.
 * It integrates via a gRPC API with other well known databases that have been certified to be Jaeger compliant: [TimescaleDB via Promscale](https://github.com/timescale/promscale), [ClickHouse](https://github.com/jaegertracing/jaeger-clickhouse).
 * There is embedded database support using [Badger](https://github.com/dgraph-io/badger) and simple in-memory storage for testing setups.
-* ScyllaDB can be used as a drop-in replacement for Cassandra since it uses the same data model and query language.
+* ScyllaDB [can be used](https://github.com/jaegertracing/jaeger/blob/main/plugin/storage/scylladb/README.md) as a drop-in replacement for Cassandra since it uses the same data model and query language.
 * There are ongoing community experiments using other databases, such as InfluxDB, Amazon DynamoDB, YugabyteDB(YCQL).
 
 ### Modern Web UI

plugin/storage/scylladb/README.md

@@ -0,0 +1,40 @@
+### ScyllaDB as a storage backend
+Jaeger could be configured to use ScyllaDB as a storage backend. This is an experimental feature and this is not an officially supported backend, meaning that Jaeger team will not proactively address any issues that may arise from incompatibilities between the ScyllaDB and Cassandra databases (the team may still accept PRs).
+
+### Configuration
+
+Setup Jaeger server to use Cassandra database and just replace conn string to ScyllaDB cluster. No additional configuration is required.
+
+### Known issues
+
+#### Protocol version
+
+Jaeger server detects Cassandra protocol version automatically. At the date of the demo with specified versions server detects that it connected via protocol version 3 while it is actually 4. This leads to warn log in cassandra-schema container:
+```
+WARN: DESCRIBE|DESC was moved to server side in Cassandra 4.0. As a consequence DESRIBE|DESC will not work in cqlsh '6.0.0' connected to Cassandra '3.0.8', the version that you are connected to. DESCRIBE does not exist server side prior Cassandra 4.0.
+Cassandra version detected: 3
+```
+
+Otherwise, it should be fully compatible.
+
+### Demo
+
+Docker compose file consists of Jaeger server, Jaeger Cassandra schema writer, Jaeger UI, Jaeger Demo App `HotRod` and a ScyllaDB cluster.
+
+There is known issues with docker compose network configuration and containers connectivity on Apple Silicone. That's why before `upping` the docker compose you need to manually create network which later be used by docker compose:
+```shell
+docker network create --driver bridge jaeger-scylladb
+```
+
+Create ScyllaDB cluster with 3 nodes and give it some time to start(around 1 minute):
+```shell
+docker compose up -d scylladb scylladb2 scylladb3
+```
+
+Start Jaeger server, Jaeger UI, Jaeger Demo App `HotRod` and Jaeger Cassandra schema writer:
+```shell
+docker compose up -d
+```
+Open Demo app in your browser: http://localhost:8080 and make some clicks 
+
+Open Jaeger UI in your browser: http://localhost:16686 and check traces

plugin/storage/scylladb/docker-compose.yml

@@ -0,0 +1,87 @@
+networks:
+  jaeger-scylladb:
+    external: true
+    driver: bridge
+
+services:
+  collector:
+    image: jaegertracing/jaeger-collector:1.43
+    environment:
+      SPAN_STORAGE_TYPE: cassandra
+      CASSANDRA_SERVERS: scylladb
+      CASSANDRA_KEYSPACE: jaeger_v1_test
+    networks:
+      - jaeger-scylladb
+    depends_on:
+      - scylladb
+
+  web:
+    image: jaegertracing/jaeger-query:1.43
+    restart: unless-stopped
+    ports:
+      - 16686:16686
+      - 16687:16687
+    environment:
+      SPAN_STORAGE_TYPE: cassandra
+      CASSANDRA_SERVERS: scylladb
+      CASSANDRA_KEYSPACE: jaeger_v1_test
+    networks:
+      - jaeger-scylladb
+    depends_on:
+      - scylladb
+
+  cassandra-schema:
+    image: jaegertracing/jaeger-cassandra-schema:1.43
+    environment:
+      CQLSH_HOST: scylladb
+      DATACENTER: test
+      MODE: test
+    networks:
+      - jaeger-scylladb
+    depends_on:
+      - scylladb
+
+  scylladb:
+    restart: always
+    image: scylladb/scylla:5.1.7
+    ports:
+      - 9042:9042
+    volumes:
+      - .docker/scylladb/1:/var/lib/scylla
+    networks:
+      - jaeger-scylladb
+    healthcheck:
+      test: ["CMD", "./scylla-healthcheck.sh"]
+      interval: 2s
+      retries: 120
+      timeout: 1s
+
+  scylladb2:
+    restart: always
+    image: scylladb/scylla:5.1.7
+    command: --seeds=scylladb
+    volumes:
+      - .docker/scylladb/2:/var/lib/scylla
+    networks:
+      - jaeger-scylladb
+
+  scylladb3:
+    restart: always
+    image: scylladb/scylla:5.1.7
+    command: --seeds=scylladb
+    volumes:
+      - .docker/scylladb/3:/var/lib/scylla
+    networks:
+      - jaeger-scylladb
+
+  hotrod:
+    image: jaegertracing/example-hotrod:latest
+    ports:
+      - 8080:8080
+    command: [ "all" ]
+    environment:
+      - OTEL_EXPORTER_JAEGER_ENDPOINT=http://collector:14268/api/traces
+    networks:
+      - jaeger-scylladb
+    depends_on:
+      - collector

plugin/storage/scylladb/scylladb_healthcheck.sh

@@ -0,0 +1,15 @@
+#!/bin/bash
+
+# Define the expected number of nodes
+EXPECTED_NODES=3
+SCYLLADB_CONTAINER_NAME=jaeger-scylladb-1
+
+# Get the actual number nodes which status is UN (Up and Normal)
+ACTUAL_NODES=$(docker exec -it $(docker ps -a | grep $SCYLLADB_CONTAINER_NAME | cut -d' ' -f1) nodetool status | grep '^UN' | wc -l)
+
+# Compare the actual number of nodes with the expected number
+if test "$ACTUAL_NODES" -ne "$EXPECTED_NODES"; then
+  exit 1
+fi
+
+exit 0