Hazelcast provides two types of support: one for Community Edition users and one for Enterprise Edition customers.
Community support
Community support is available to every Hazelcast user. You can use the following channels to ask for help:
-
Hazelcast GitHub to report issues through our GitHub repository.
-
Hazelcast Community Slack workspace to ask for help from the Hazelcast user community.
Customer support
Hazelcast customer support is available to Enterprise Edition customers. See hazelcast.com for the support options.
A support subscription from Hazelcast provides rapid response times to technical support inquiries, access to critical software patches, and other services which will help you achieve increased productivity and quality. Learn more about Hazelcast support subscriptions: https://hazelcast.com/pricing/.
| You can create a support ticket using the Hazelcast Support Portal. |
The following section explains what information to collect and how to submit a ticket.
Before submitting a ticket
To help Hazelcast Support diagnose and resolve issues quickly, provide as much accurate and complete information as possible. This section explains how to prepare your environment and what information and logs to collect.
Hazelcast configuration
Much of the information required to understand a Hazelcast deployment is contained in the configuration, such as the number of members that are running, and details of the servers or virtualized environments they’re running on.
The Hazelcast configuration is supplied to a Hazelcast instance on startup in one of two ways:
-
Declaratively by supplying the Hazelcast member with an XML or YAML file; or
-
Programmatically by starting an embedded instance of Hazelcast with a
com.hazelcast.config.Configobject
You can provide the Hazelcast configuration to the Support team, either as an XML or YAML file, or the code blocks making up the Config object. Alternatively, you can obtain the configuration from each member in the cluster by visiting their member pages in the Members section of Management Center.
Client configuration
The Hazelcast clients also have their own configurations that determine their behavior and the features they utilize. Some clients, such as Java, .NET, and Node.js, can be configured declaratively or programmatically, while others, such as Go, Python, and C++, can only be configured programmatically.
You can provide the following for each Hazelcast client configuration:
| Client type | Declarative | Programmatic |
|---|---|---|
Java |
XML, YAML |
Yes |
.NET |
XML |
Yes |
Node.js |
JSON |
Yes |
Go |
- |
Yes |
Python |
- |
Yes |
C++ |
- |
Yes |
Hazelcast logs
The Hazelcast logs contain useful information when debugging an issue in a Hazelcast cluster. Each member has its own Hazelcast log, which typically includes:
-
Metrics about the Hazelcast member, such as HealthMonitor output.
-
Log messages indicating when Hazelcast clients connect and disconnect from the member.
-
Details about slow operations.
-
Stack traces.
-
Other log messages regarding cluster health and performance.
We recommend you configure Hazelcast to use log4j, log4j2 or logback via slf4j. For more information, see Configuring Logging.
| Review the configuration of your logging framework to ensure Hazelcast logs are being kept for an appropriate amount of time, and that the correct log level is set (INFO by default). |
Diagnostic logging
Hazelcast provides diagnostic logging capabilities that help with troubleshooting and monitoring.
An extended set of diagnostics plugins is provided for both Hazelcast members and clients. These can capture:
-
Information about the Hazelcast build version
-
Properties related to the Operating System, JVM, and Hazelcast configuration
-
Details about the invocations run on the member
-
State of the Hazelcast cluster
-
A log of system events, such as cluster version changes, member additions and removals, etc.
-
Map store metrics
-
Details about the type and volume of operations run on the member
-
WAN Replication diagnostic information
The Hazelcast configuration provides settings for the rotation and retention of these diagnostic log files. For more information, see Diagnostics.
| Do not truncate diagnostics logs. If you must truncate a log, ensure the startup section—containing environment and system details—is included. |
JVM garbage collection logs
The JVM’s garbage collection logs are often valuable in diagnosing performance problems in Hazelcast. Garbage collection (GC) logging is not enabled by default. We recommend you enable GC logging in production environments for performance analysis.
In Java 9 and above, the output file path is set using the file= parameter:
-Xlog:safepoint,gc+age=debug,gc*=debug:file=gc.log:uptime,level,tags:filesize=10m,filecount=10
For more information, see Garbage Collection Logs.
Submit a support ticket
Create a ticket using the Hazelcast Support Portal and include the following information:
-
Title and description
Write a clear, concise title summarising the problem. In the description, provide:
-
What you observed. Capture any relevant screenshots and or errors noted, and include the exact incident date and time
-
What you expected
-
Whether the issue is ongoing, intermittent, or resolved
-
-
Steps to reproduce
Document detailed steps so Support can attempt to reproduce the issue.
-
Optional reproducible test case
If available, include a minimal project or code sample that demonstrates the problem. Engineering may request this later if required.
-
Use case details
Explain how you use Hazelcast, including:
-
Data structures involved
-
Client types
-
Cluster size
-
Deployment environment (cloud, on-prem, containerised)
-
Any recent changes (upgrades, configuration changes, scaling events)
-
Set the ticket severity
Choose the correct severity to ensure the issue is routed appropriately.
| Severity | Description |
|---|---|
Level 1 |
Current production outage with no workaround |
Level 2 |
Current production outage with a temporary workaround (for example, serving static content) |
Level 3 |
Development issues, low-impact production issues, or analysis of an incident that has already ended |
Level 4 |
General questions, enhancement requests, or other non-critical issues |
After submitting your ticket, please respond promptly to any follow-up questions or requests for additional data as timely communication helps accelerate investigation and resolution.
Summary
To ensure the fastest possible resolution:
-
Keep diagnostics and logging enabled in all environments
-
Collect detailed observations before submitting a ticket
-
Provide complete, untruncated logs and diagnostics
-
Include Hazelcast member and client configuration files
-
Select the correct severity level
-
Communicate promptly with Hazelcast Support
Developer resources
-
Hazelcast source code can be found at GitHub/Hazelcast. This is also where you can contribute and report issues.
-
Hazelcast API can be found at hazelcast.org/docs/Javadoc.
-
Code samples can be downloaded from GitHub/hazelcast-code-samples.