Snowpipe Streaming — Java SDK

Stream synthetic e-commerce data (orders + order items) into Snowflake using the Java Snowpipe Streaming SDK (com.snowflake:snowpipe-streaming:1.1.2).

When to Use

• User wants to stream data into Snowflake using Java
• User mentions Java Snowpipe Streaming SDK, Maven streaming, or JVM-based ingestion
• User needs to set up RSA key-pair auth for Java streaming
• User wants to run single or parallel streaming instances with Java
• User needs to troubleshoot JWT, Arrow, or permission errors with the Java SDK

Tools Used

• bash — Build JAR (mvn clean package), run streaming, generate RSA keys
• snowflake_sql_execute — Create databases, schemas, tables (when CoCo connection matches target)
• ask_user_question — Confirm target objects, get approval at checkpoints
• read / write / edit — Configure profile.json and config.properties

Stopping Points

• ✋ Phase 0: User approves the workflow before any action
• ✋ Step 1: User confirms Snowflake target objects (database, schema, tables, role)
• ✋ Step 2: Java 21+ and Maven verified (or installed)
• ✋ Step 3: RSA public key registered with Snowflake user
• ✋ Step 5: profile.json and config.properties correctly configured
• ✋ Step 8: Confirm data landed in Snowflake

Phase 0: Briefing & Consent

Goal: Explain what this skill does and get explicit user approval before executing anything.

⚠️ STOP: This phase MUST be completed before ANY other action. Do not run any commands, read any files, or execute any tools until the user approves.

Present the following briefing to the user:

Snowpipe Streaming (Java) — What This Skill Does

Snowpipe Streaming is Snowflake's real-time data ingestion API. Instead of staging files and loading them in batches, it streams rows directly into tables with sub-second latency. This skill uses the Java SDK (com.snowflake:snowpipe-streaming:1.1.2) backed by a Rust engine.

What will happen (8 steps):

1. Gather target details — confirm which Snowflake account, database, schema, tables, and role to use
2. Check prerequisites — verify Java 21+ and Maven are installed
3. RSA key setup — generate a new key pair (or reuse existing *.p8 file) and register the public key
4. Create Snowflake objects — create the target database, schema, and tables (owned by your role)
5. Configure — fill in profile.json and config.properties with your Snowflake details
6. Build the JAR — run mvn clean package to produce a shaded uber-jar
7. Run streaming — stream synthetic e-commerce orders + order items into Snowflake
8. Verify data — query row counts after ingestion completes

What this skill will NOT do:

• Drop or alter any existing objects not created by this workflow
• Modify files outside the project directory
• Run anything without showing you the exact command first
• Continue past any checkpoint without your explicit approval

Authentication: Uses RSA key-pair auth (not password). A new key pair will be generated and the public key registered with your Snowflake user via ALTER USER.

JVM requirement: Java 21+ is required. The SDK uses Apache Arrow internally, which needs --add-opens=java.base/java.nio=ALL-UNNAMED (pre-configured in .mvn/jvm.config).

Ask the user: "Shall I proceed with Step 1 (Gather Target Details)?"

⚠️ STOP: Do NOT proceed until the user confirms.

Workflow

Step 1: Propose Target Objects

Confirm the Snowflake objects with the user. Propose defaults and let them override:

⚠️ STOP: Ask the user to confirm or change these before proceeding.

Table ownership: The streaming SDK's role must own the target database, schema, and tables. GRANT USAGE or GRANT ALL is not sufficient — the SDK's internal pipe-info API returns ERR_TABLE_DOES_NOT_EXIST_NOT_AUTHORIZED if the role doesn't own the objects.

Step 2: Prerequisites Check

1. Check Java 21+ is installed: java --version
2. Check Maven is installed: mvn --version
3. Look for existing project files — the Java code may already exist from a previous session. Check for pom.xml and src/main/java/ in the working directory.
4. If no project exists, create it from scratch (see Project Structure at the bottom).

All Java dependencies (streaming SDK, JDBC driver, Jackson, SLF4J) are declared in pom.xml and resolved automatically by mvn clean package.

⚠️ STOP: If Java < 21 or Maven is missing, help the user install them:

• Java: brew install openjdk@21 (macOS)
• Maven: brew install maven (macOS)

Step 3: RSA Key-Pair Setup

Skip if keys already exist (look for *.p8 files in the project).

openssl genrsa 2048 | openssl pkcs8 -topk8 -inform PEM -out streaming_key.p8 -nocrypt
openssl rsa -in streaming_key.p8 -pubout -out streaming_key.pub

Register in Snowflake:

ALTER USER <username> SET RSA_PUBLIC_KEY='<contents of .pub without header/footer>';

⚠️ STOP: Key must be registered before continuing.

Critical:

• Private key must be PKCS8 unencrypted (BEGIN PRIVATE KEY, not BEGIN RSA PRIVATE KEY)
• Role must be specified in the profile — the SDK does not default to any role

Cross-account key registration: If the target is a different account from the CoCo connection, either:

1. Add a key-pair auth connection to the target account and use it
2. Have the user run ALTER USER directly in Snowsight for the target account

Step 4: Create Snowflake Objects

CoCo connection vs streaming profile: The Cortex Code SQL connection may route to a different account or user than the streaming target. If so, you cannot create tables via the CoCo SQL tool.

If CoCo connection matches the streaming target account + role, use SQL directly:

CREATE DATABASE IF NOT EXISTS <DATABASE>;
CREATE SCHEMA IF NOT EXISTS <DATABASE>.<SCHEMA>;
CREATE TABLE IF NOT EXISTS <DATABASE>.<SCHEMA>.ORDERS (
    ORDER_ID VARCHAR(36), CUSTOMER_ID NUMBER(38,0), ORDER_DATE TIMESTAMP_NTZ,
    ORDER_STATUS VARCHAR(20), TOTAL_AMOUNT FLOAT, DISCOUNT_PERCENT FLOAT, SHIPPING_COST FLOAT);
CREATE TABLE IF NOT EXISTS <DATABASE>.<SCHEMA>.ORDER_ITEMS (
    ORDER_ITEM_ID VARCHAR(36), ORDER_ID VARCHAR(36), PRODUCT_ID NUMBER,
    PRODUCT_NAME VARCHAR, PRODUCT_CATEGORY VARCHAR, QUANTITY NUMBER,
    UNIT_PRICE FLOAT, LINE_TOTAL FLOAT);

If CoCo connection differs, write and run a temporary Python setup script:

import snowflake.connector
from cryptography.hazmat.primitives import serialization
from cryptography.hazmat.backends import default_backend

with open("streaming_key.p8", "rb") as f:
    p_key = serialization.load_pem_private_key(f.read(), password=None, backend=default_backend())
pkb = p_key.private_bytes(encoding=serialization.Encoding.DER, format=serialization.PrivateFormat.PKCS8, encryption_algorithm=serialization.NoEncryption())

conn = snowflake.connector.connect(user="<USER>", account="<ACCOUNT>", private_key=pkb, role="<ROLE>", warehouse="<WAREHOUSE>")
cursor = conn.cursor()
for stmt in [
    "CREATE DATABASE IF NOT EXISTS <DATABASE>",
    "CREATE SCHEMA IF NOT EXISTS <DATABASE>.<SCHEMA>",
    "CREATE TABLE IF NOT EXISTS <DATABASE>.<SCHEMA>.ORDERS (ORDER_ID VARCHAR(36), CUSTOMER_ID NUMBER(38,0), ORDER_DATE TIMESTAMP_NTZ, ORDER_STATUS VARCHAR(20), TOTAL_AMOUNT FLOAT, DISCOUNT_PERCENT FLOAT, SHIPPING_COST FLOAT)",
    "CREATE TABLE IF NOT EXISTS <DATABASE>.<SCHEMA>.ORDER_ITEMS (ORDER_ITEM_ID VARCHAR(36), ORDER_ID VARCHAR(36), PRODUCT_ID NUMBER, PRODUCT_NAME VARCHAR, PRODUCT_CATEGORY VARCHAR, QUANTITY NUMBER, UNIT_PRICE FLOAT, LINE_TOTAL FLOAT)",
]:
    cursor.execute(stmt)
    print(cursor.fetchone()[0])
cursor.close(); conn.close()

Streaming pipes are auto-created by Snowflake with the naming convention <TABLE_NAME>-STREAMING.

Step 5: Configure Profile and Properties

Create profile.json with the user's confirmed values from Step 1:

{
  "user": "<USER>",
  "account": "<ACCOUNT>",
  "url": "https://<ACCOUNT>.snowflakecomputing.com:443",
  "role": "<ROLE>",
  "private_key_file": "../streaming_key.p8",
  "database": "<DATABASE>",
  "schema": "<SCHEMA>",
  "warehouse": "<WAREHOUSE>"
}

Create config.properties:

channel.orders.name=orders_channel
channel.order_items.name=order_items_channel
pipe.orders.name=<ORDERS_TABLE>-STREAMING
pipe.order_items.name=<ORDER_ITEMS_TABLE>-STREAMING
max.client.lag=60
batch.size=1000
orders.batch.size=10000
num.orders.per.batch=100
max.retries=3
retry.delay.ms=1000

Profile supports two key formats:

• "private_key_file": Path to .p8 file (relative to the java/ directory). Preferred.
• "private_key": Inline PEM string with \n newlines.

⚠️ STOP: Confirm profile and config files are created and populated before continuing.

Step 6: Build the JAR

mvn clean package -q

Produces: target/snowpipe-streaming-app-1.0.0.jar (shaded uber-jar).

CRITICAL JVM flag for Arrow memory compatibility:

--add-opens=java.base/java.nio=ALL-UNNAMED

Pre-configured in .mvn/jvm.config. Must be added manually when running java -cp directly.

If error occurs:

• Compilation error → check Java version is 21+ (java --version)
• Dependency download failure → check network connectivity and Maven settings
• shade plugin error → run mvn clean package (not just mvn package)

Step 7: Run Streaming

Customer IDs are generated synthetically (1-10000). No database lookup or CUSTOMERS table needed.

Single instance (1K-50K orders):

java --add-opens=java.base/java.nio=ALL-UNNAMED \
  -cp target/snowpipe-streaming-app-1.0.0.jar \
  com.snowflake.demo.StreamingApp <num_orders> <config_file> <profile_file>

Parallel (10K+ orders, recommended):

java --add-opens=java.base/java.nio=ALL-UNNAMED \
  -cp target/snowpipe-streaming-app-1.0.0.jar \
  com.snowflake.demo.ParallelStreamingOrchestrator <total_orders> <num_instances> [config_file] [profile_file]

Examples:

# 10K orders, 5 instances
java --add-opens=java.base/java.nio=ALL-UNNAMED \
  -cp target/snowpipe-streaming-app-1.0.0.jar \
  com.snowflake.demo.ParallelStreamingOrchestrator 10000 5 config.properties profile.json

Use run_in_background=true in the Bash tool for long runs.

If error occurs:

• InaccessibleObjectException → missing --add-opens JVM flag
• ClassNotFoundException → use -cp (not -jar) with full class name
• ERR_TABLE_DOES_NOT_EXIST_NOT_AUTHORIZED → role doesn't own tables (see Troubleshooting)
• JWT / auth errors → check profile.json account format and key registration

Parallel details:

• Each instance gets unique channels: orders_channel_instance_0, etc.
• Customer ID range (1-10000) is partitioned across instances
• Config/profile are CLI args 3 and 4

Step 8: Verify Data

After the run completes, the app waits 65 seconds for flush then runs reconciliation.

Data may take 1-3+ minutes to appear in queries after completion. If COUNT returns 0, wait 2-3 minutes and retry.

SELECT COUNT(*) FROM <DATABASE>.<SCHEMA>.ORDERS;
SELECT COUNT(*) FROM <DATABASE>.<SCHEMA>.ORDER_ITEMS;

Expected ratio: ~3-4 order items per order.

Known issue: Reconciliation uses JDBC which may fail with org-account format. This is non-critical — data is already ingested. Just verify counts manually.

⚠️ STOP: Confirm data landed successfully.

Troubleshooting

First-Time / Trial Account Setup

Skip this section if the user already has Cortex Code and a Snowflake connection configured.

Install Cortex Code

brew install Snowflake-Labs/cortex-code/cortex
# or: npm install -g @snowflake-labs/cortex-code

Set Up Snowflake Connection

If the streaming target is a different account from the user's current Cortex Code session, use key-pair auth (not externalbrowser).

[streaming]
account = "orgname-accountname"
user = "YOUR_USERNAME"
authenticator = "SNOWFLAKE_JWT"
private_key_file = "/absolute/path/to/key.p8"
host = "orgname-accountname.snowflakecomputing.com"
database = "<DATABASE>"
schema = "<SCHEMA>"
warehouse = "<WAREHOUSE>"
role = "<ROLE>"

Finding your account identifier:

• Snowsight → click your name (bottom-left) → Account → copy the orgname-accountname value
• Uses hyphens not dots or underscores

Activate: /connection streaming

Architecture Notes

• Customer IDs are synthetic (1-10000) — no database lookup. Matches Python implementation.
• One SDK client per pipe/table — separate clients for orders and order items.
• Streaming SDK uses private_key as PEM string. JDBC (reconciliation) uses parsed PrivateKey object — different formats.
• Rust SDK backend produces very verbose logs (cyclone_shared). Use filters when checking output.
• Offset tokens are strings: order_{uuid}, item_{uuid}.

Project Structure

If creating from scratch, the project needs:

<project_root>/
├── java/
│   ├── pom.xml                                    # Maven build (Java 21, shade plugin)
│   ├── .mvn/jvm.config                            # --add-opens flag
│   ├── config.properties                          # Default config
│   ├── profile.json                               # Default credentials
│   └── src/main/java/com/snowflake/demo/
│       ├── StreamingApp.java                     # Single-instance entry point
│       ├── ParallelStreamingOrchestrator.java     # Multi-instance entry point
│       ├── SnowpipeStreamingManager.java          # SDK client/channel management
│       ├── ReconciliationManager.java             # Post-ingestion cleanup (JDBC)
│       ├── ConfigManager.java                     # Config + profile loader
│       ├── DataGenerator.java                     # Synthetic data generation
│       ├── Order.java                             # Order model with toMap()
│       ├── OrderItem.java                         # OrderItem model with toMap()
│       └── Customer.java                          # Customer model (unused for streaming)
├── streaming_key.p8                               # RSA private key
└── streaming_key.pub                              # RSA public key

Output

• Java project directory with compiled uber-JAR, keys, and profiles
• Snowflake database/schema/tables created and owned by the specified role
• Synthetic e-commerce data (orders + order items) streamed into target tables
• Verified row counts confirming successful ingestion