Supported platforms

Still need help?

The Atlassian Community is here for you.

Ask the community

This page lists the supported platforms for Fisheye 4.8 and its bug-fix releases. We cannot ensure the stability of Fisheye used with a platform or a versions that is not mentioned in this page.

Key(tick) = Supported  (warning) = Deprecated  (error) = Not Supported


Oracle Java

(tick) 1.8

Fisheye requires the Java Runtime (JDK or JRE), version as noted. Pre-release/early access versions of Java are not supported. We recommend that you use the Oracle JDK or Adopt OpenJDK as these are implementations we test Fisheye against.

Please note:

  • Ensure that you've set your max heap size (--Xmx) to a reasonable value, considering the memory requirements of your system.
  • You'll need the JDK for the Fisheye JSP source download.
  • For OpenJDK, you'll need the DejaVu font package installed. Installation instructions can be found here:
  • For OpenJDK on Windows, you may need to instal the 'Microsoft Visual C++ Redistributable 2013' package in order to have avatars rendered correctly (details).


(tick) 1.8

Operating systems

Microsoft Windows


  • Fisheye is a pure Java application and should run on any platform provided the requirements for the JRE or JDK are satisfied.

Although Fisheye can be run in virtualized environments, Atlassian is not yet able to provide technical support for performance-related problems in a virtualized environment. If you do choose to run Fisheye in a VM, please ensure that you choose a VM with good IO throughput.



Apple Mac OS X



HSQLDB(tick) Bundled; for evaluation use only

The Fisheye built-in database, running HSQLDB, is somewhat susceptible to data loss during system crashes. We recommend that you do not use HSQLDB for production systems.

External databases (such as MySQL) are generally more resistant to data loss during a system crash.

See the  Fisheye Database documentation  for further details.


(tick) 5.6, 5.7 Enterprise Server
(tick) 5.6, 5.7 Community Server

(error) MariaDB, Percona


(tick) 9.6 - 14

(warning) 9.4, 9.5

PostgreSQL 12 is supported since Fisheye 4.8.5.

PostgreSQL 13&14 are supported since Fisheye 4.8.9.


(tick) 19c

(tick) 12c R1, R2

Oracle 19c is supported since Fisheye 4.8.5.

SQL Server

(tick) 2019

(tick) 2017

(tick) 2016

(warning) 2012, 2014

SQL Server 2019 is supported since Fisheye 4.8.5.

Web browsers

Internet Explorer

(warning) 11.0

Microsoft Edge(tick) Latest stable version supportedLegacy and Chromium.

Mozilla Firefox

(tick) Latest stable version supported


(tick) Latest stable version supported


(tick) Latest stable version supported

Version control systems


(tick) 1.8, 1.9, 1.10

(warning) 1.7

(tick) SVNKit (bundled & the default)

(tick) Native JavaHL 1.8, 1.9, 1.10

(warning) Native JavaHL 1.7


(tick) All versions


(tick) 2018.x - 2023.x

(warning) 2007.3 - 2017.2

The Server must support the ztag tagged protocol.

(error) Perforce Streams are not supported (FE-3886).


(tick) 2.39.2

(tick) 2.38.4

(tick) 2.37.6

(tick) 2.36.5

(tick) 2.35.7

(tick) 2.34.7

(tick) 2.33.7

(tick) 2.32.6

(tick) 2.31.7

(tick) 2.30.8

(warning) 1.9.5-2.29

These are the versions of Git that we currently test Fisheye against.

Git for Windows is the only supported distribution when running Fisheye on Windows. Cygwin Git is not supported.


(tick) 3.7.3-4.5.3
till Fisheye 4.8.8

(tick) 3.7.3-5.9.3
since Fisheye 4.8.9

(warning) 2.8.2, 2.9.1, 3.0.2, 3.1.2, 3.2.4, 3.3.3, 3.4.2, 3.5.2, 3.6.3

These are the versions of Mercurial that we currently test Fisheye against.

You should restart Fisheye after upgrading Mercurial.

Use Mercurial compiled with Python 2.7, Fisheye is incompatible with Mercurial built with Python 3.

Mercurial on Windows does not provide support for non-ASCII characters in file names and paths. Read more about it in: FE-7094 - Getting issue details... STATUS
Atlassian applications


(tick) Crowd 2.4.x+

(tick) Crowd client library: 2.4.1

Jira(tick) Jira 6.4+

Fisheye to Jira communication requires Jira 6.4.x or later. Communication the other way, from Jira to Fisheye, depends on the Jira Fisheye Plugin.

Note that the Jira Fisheye Plugin is bundled with Jira. If you are using a version of Jira earlier than Jira 6.4 you can upgrade the plugin in Jira to get support for Fisheye.


You can use official images to deploy FishEye in a Docker container, or customize a Docker deployment on your own.

We support the Atlassian Docker templates and can help with FishEye related problems. We do not provide support for Docker itself or problems with any Docker environment.

Hardware requirements

Fisheye should ideally run on a dedicated server. The most important aspect for a large-repository deployment will be I/O speed. You definitely want a fast local HDD for Fisheye's cache. Note that NFS and SAN are not supported.




1.8GHz or higher, a single core is sufficient. More cores or higher GHz will result in better load-handling ability.


1GB minimum, 2GB will provide performance "headroom". Your Java heap should be sized at 512MB with the FISHEYE_OPTS environment variable, adjustable up to 1024MB depending on performance.


Fisheye's input/output is an important element of its overall performance. If Fisheye accesses your repository remotely, make sure that the throughput is maximum and the latency minimum (ideally the servers are located in the same LAN, running at 100Mbps or faster).

MonitorMinimum screen resolution of 1024x768. Recommended screen resolution of 1280x768 or above.

While some of our customers run Fisheye on SPARC-based hardware, Atlassian only officially supports Fisheye running on x86 hardware and 64-bit derivatives of x86 hardware.

Disk space requirement estimates

Disk space requirements for Fisheye may vary due to a number of variables such as the repository implementation, file sizes, content types, the size of diffs and comments being stored. The following table contains some real-world examples of Fisheye disk space consumption.

Repository technology


Codebase size (HEAD of trunk)

Fisheye index size



466 MB in 12151 files

647 MB



115 MB in 11433 files

220 MB

(warning) These disk space estimates are to be used as a guideline only. We recommend you monitor the disk space that your Fisheye instance uses over time, as needs for your specific environment may vary. It may be necessary to allocate more space than indicated here. Additionally, you can reduce disk space consumption by turning off diff storage in Fisheye.

Deployment notes for version control systems

Subversion (server)

Fisheye can communicate with any repository running Subversion 1.1 or later.

Subversion (client)

Fisheye now bundles the SVNkit client, which becomes the default Subversion interface. An alternative is to use the native subversion client, using JavaHL bindings. Please see Subversion Client Setup for more information.

Perforce (client)

Fisheye needs access to the p4 client executable. Due to some problems with earlier versions of the client, we recommend version 2007.3 or later.


If you are using CVS, Fisheye needs read-access to your CVS repository via the file system. It does not support protocols such as pserver at the moment.

Support for other version control systems is planned for future releases. Let us know what SCM system you would like to see supported by creating a Jira issue or adding your vote to an issue, if the request already exists.

WAR deployment

Fisheye/Crucible is a standalone Java program. It cannot be deployed to web application servers such as WebSphere, Weblogic or Tomcat.

Single sign on with Atlassian Crowd

From version 2.8.x, Fisheye bundles the Crowd 2.4.1 client library, and supports the Crowd 2.4.x server, and later versions.
Last modified on Nov 8, 2023

Was this helpful?

Provide feedback about this article

In this section

Powered by Confluence and Scroll Viewport.