Taverna 2.3.0 "Taverna 2.3.0"

Milestone information

Code name:
Taverna 2.3.0
Stian Soiland-Reyes
Release registered:
No. Drivers cannot target bugs and blueprints to this milestone.  

Download RDF metadata


Assigned to you:
No blueprints or bugs assigned to you.
No users assigned to blueprints and bugs.
No blueprints are targeted to this milestone.
No bugs are targeted to this milestone.

Download files for this release

After you've downloaded a file, you can verify its authenticity using its MD5 sum or signature. (How do I verify a download?)

File Description Downloads
download icon taverna-commandline-2.3.0.zip (md5, sig) Taverna command line tool 2.3 200
last downloaded today
download icon taverna-workbench-2.3.0.tar.gz (md5, sig) Taverna workbench 2.3 Linux archive 1,220
last downloaded 3 days ago
download icon taverna-workbench-src-2.3.0.zip (md5, sig) Taverna workbench 2.3 source code 252
last downloaded 9 days ago
download icon taverna-workbench-2.3.0.dmg (md5, sig) Taverna workbench 2.3 OS X disk image 1,002
last downloaded 2 days ago
download icon taverna-workbench-2.3.0.exe (md5, sig) Taverna workbench 2.3 Windows self-extracting archive 1,107
last downloaded 2 days ago
download icon taverna-workbench-2.3.0-installer.exe (md5, sig) Taverna workbench 2.3 Windows installer 3,336
last downloaded 24 hours ago
Total downloads: 7,117

Release notes 

Taverna workbench 2.3.0

Released by myGrid, 2011-07-12
(c) Copyright 2005-2011 University of Manchester, UK

Taverna is licenced under the GNU Lesser General Public Licence. (LGPL) 2.1.
See the file LICENCE.txt or http://www.gnu.org/licenses/lgpl-2.1.html for

If the source code was not included in this download, you can download it from
http://www.taverna.org.uk/download/workbench/2-3/#download-source or

Taverna uses various third-party libraries that are included under compatible
open source licences such as the Apache Licence.

Running Taverna

If you used the installer you should have Taverna->Taverna workbench 2.3.0 in
your Start menu. If you used the self-extracting archive, double-click on
"taverna.exe" in the extracted folder.

If you used the installer, it should have checked that you have a recent
version of Java installed. See "Requirements" below.

Mac OS X
Running Taverna on a Mac requires Mac OS X 10.5 or later with the latest
updates installed for Java 6. 32-bit machines will need 10.6 or later.

Copy the "Taverna" application bundle to /Applications or a folder of your
preference. Eject the disk image, and run Taverna from the installed location.

You might also need to install Graphviz as it is no longer bundled with Taverna.
See "Requirements" below.

For Linux, either double-click on "taverna.sh", or start a Terminal and execute
"./taverna.sh" or "sh taverna.sh" from the folder where you extracted Taverna.

You will need to have Graphviz and Oracle's Java runtime installed. See
"Requirements" below.

See http://www.taverna.org.uk/documentation/taverna-2-x/ for
documentation and tutorials on using Taverna.

See http://www.mygrid.org.uk/dev/wiki/display/taverna23/User+Manual for the
Taverna 2.3 user manual.

See the file ''known-issues.txt'' for known issues with this release, and the
file ''release-notes.txt'' for improvements since the previous version of

Example Taverna 2.3 workflows can be found in the myExperiment starter pack at

You can also access these as the tab 'Starter pack' under the *myExperiment*
view inside Taverna.

You can share and find other workflows at http://www.myexperiment.org/

You can find and describe web services in life sciences at

See http://www.taverna.org.uk/about/contact-us/ for contact details.

You may email <support@mygrid.org.uk> for any questions on using Taverna
workbench. myGrid's support team should respond to your query within a

Mailing lists

We also encourage you to sign up to the public *taverna-users* mailing list,
where you may post about any problem or give us feedback on using Taverna.
myGrid developers are actively monitoring the list.

 * http://lists.sourceforge.net/lists/listinfo/taverna-users
 * http://taverna-users.markmail.org/search/?q=

If you are a developer, writing plugins for Taverna, dealing with the code
behind Taverna or integrating Taverna with other software, you might find it
interesting to also sign up for the public *taverna-hackers* mailing list,
where you can also track the latest developments of Taverna.

  * http://lists.sourceforge.net/lists/listinfo/taverna-hackers
  * http://taverna-hackers.markmail.org/search/?q=

Taverna might ask if you would like to register as a Taverna user. The myGrid
team would appreciate if you do, as it would:

  * Allow us to support you better; future plans will be directed towards
    solutions Taverna users require

  * Help sustain Taverna development

By registering, you will *not* be giving us access to your data or service
usage. We will *not* be monitoring your usage of Taverna, we will only record
the information you provide in the registration form.

For full terms and condition of our registration, see


The external tool Graphviz ( http://www.graphviz.org/ ) is used for generating
the workflow diagram in Taverna. A version of Graphviz is bundled in the
Windows install, while Linux and Mac OS X users will need to install Graphviz

For OS X, see http://www.graphviz.org/Download_macos.php where you can download
the *current stable release* for your operating system.

For Linux, most distributions have a way to install Graphviz. For Ubuntu, try:
  sudo apt-get install graphviz

For other distributions, see http://www.graphviz.org/Download.php for
instructions. Please note that you do *not* need the Grappa java library.

In most cases Taverna should be able to pick up the location of the Graphviz
binary "dot" from the PATH, if this is not the case, see
http://www.mygrid.org.uk/dev/wiki/display/taverna/Workbench+preferences on how
to configure the location of "dot".

Taverna requires the Java Runtime Environment (JRE) version 6 or later from
Oracle. No other versions of Java are officially tested with Taverna.

Windows users might need to download Java from http://java.com/ - but in most
cases the Taverna installer would already have checked if Java is installed.

Linux users have different options to install Java depending on their Linux
distribution. Some distributions, such as Ubuntu, might come with alternative
open source implementations of Java, like Gnu GCJ and OpenJDK. We've identified
some issues with these implementations, and recommend using the official Java
implementation from Sun/Oracle.

To download Oracle Java 6 for Ubuntu, start a Terminal, and type the following:
  sudo apt-get install sun-java6-jre

and follow the instructions. You might also need to change the default Java
implementation by running:
  sudo update-alternatives --config java

To check your version of Java on the command line, try:

  $ java -version
  java version "1.6.0_22"
  Java(TM) SE Runtime Environment (build 1.6.0_22-b04)
  Java HotSpot(TM) 64-Bit Server VM (build 17.1-b03, mixed mode)

Read http://www.taverna.org.uk/download/workbench/system-requirements/
for more requirement details.

Secure web services
If you need to invoke secure services or access secured data from your
workflows (e.g. if you need to provide username and password to gain access to
your service/data or your service's URL starts with HTTPS), then you need to
allow Taverna to communicate securely to such resources using the strongest
possible cryptography.

To do so, you have to install the 'Unlimited Strength Java Cryptography
Extension' policy instead of the default restrictive policy that is
shipped with Java; the default policy will for export reasons only
allows the use of "weak" cryptography (e.g. short passwords and keys).

Java by default only comes with limited-strength cryptography support
because of import control restrictions in some countries, where
unlimited-strength cryptography is classified as a "weapon technology".
Thus, you must determine whether your country's laws allow you to
install such software and you are responsible for doing so.

The policy files on your system are located in:


where <java-home> is the jre directory of the Java Development Kit (JDK) or the
top-level directory of the JRE (Java Runtime Environment) on your system.

If eligible, you may download the unlimited strength versions of the above
policy files and replace the ones that were installed by default. This will
effectively "turn on" the use of unlimited-strength security in Java on your
system. Not installing these files will most probably cause invoking of secure
services from Taverna to fail. Note that if you switch to another Java version
you will have to install the policy files again in the appropriate directory of
your new Java installation.

For Java 6, you can download the unlimited cryptography policy jar files from:
  http://tinyurl.com/java6sec -or-



If you are using a Windows operating system and want to see Taverna's
logging information, then run "taverna-debug.bat" from the installed
Taverna folder.

In OS X, start "/Applications/Utillities/Console".

In Linux run "taverna.sh" from a Terminal.

You can also find detailed logs in Taverna's home directory. You can
view this directory from within Taverna by choosing
Advanced->Show Log Folder in the menu.


    Windows 7/Vista: C:\Users\JohnDoe\AppData\Roaming\taverna-2.3.0\logs

    Windows XP: C:\Documents and settings\JohnDoe\
                   Application Data\taverna-2.3.0\logs

    OS X: /Users/JohnDoe/Library/Application support/taverna-2.3.0/logs

    Linux: /home/johndoe/.taverna-2.3.0/logs

Note that "Application data" and "AppData" are hidden folders in Windows.

To modify the log levels, edit "conf/log4j.properties" in the Taverna
installation folder.

Default services and plugins

You can edit the default service list for the Taverna installation by
editing "conf/default_service_providers.xml". You can generate this
service list from within Taverna by right-clicking on "Available
Services" and selecting "Export Services To File".

Similarly you can replace the installation "plugins/plugins.xml" with
"plugins.xml" from the Taverna home directory to force installation of a
plugin by default. Note that in this case it is also recommended to copy
the full content of "repository" from the Taverna home directory to
append the installation directory's "repository".

Memory usage
For OS X and Linux, Taverna will use a maximum of 400 MB. In Windows,
"taverna-debug.bat" will also use 400 MB, while "taverna.exe" will use up to
400 MB or, if higher, 15% of available memory at the point of launching

If Taverna happens to run out of memory while running a workflow with
large data (or large number of data), you can set Taverna to store data
in a database by going to Preferences->Data and Provenance and remove
the tick for "In-memory storage". This will reduce execution speed
slightly, but should consume less memory. Also remember to click
'Remove' on old runs you are no longer interested in.

has more details on the data settings.

If you need to increase the available memory, edit "taverna.sh" or
"taverna-debug.bat" and replace "-Xmx400m" with say "-Xmx600m" to use
600 MiB.

For OS X you would need to right click on "Taverna.app" and go inside
the application bundle to edit Contents/Info.plist and change the same
parameter using the Property List Editor.

has more details on memory allocation.


View the full changelog

Changes in Taverna 2.3.0

The major changes are the addition of REST, XPath and Tool services,
the addition of the Service Catalogue perspective, and the ability to
export a workflow run as OPM or Janus.

For a complete list of issues that have been solved for Taverna 2.3.0
(including those for 2.2.1 and 2.2.2) see

General user interface
Control-click now works on Macs (T2-237)

User-friendly way of reporting bugs (T2-549)

The open workflow dialog has been slightly improved (T2-731)

"Save diagram" has been renamed to "Export diagram" (T2-738)

"Open location" icon is now available in the toolbar (T2-752)

The top level icons have been ordered to agree with the corresponding
menu items (T2-753)

The Credential Manager checks for previous versions of Taverna
and asks the user to import old Keystore/Truststore (T2-944)

Empty service providers can now be removed (T2-1151)

Important notice mechanism added (T2-1282)

Reduction in the size of the registration dialog (T2-1302)

The "Save as" dialog now suggests a name (T2-1318)

Help is now shown in a standard browser, fixing many issues (T2-1518
and T2-1706)

A problem with service configuration dialog text has been fixed

Users can now see the "Application data" directory (T2-1612)

Taverna now uses the standard SwingWorker (T2-1683)

The service panel now allows services with identical names (T2-1714)

Taverna now handles up-to-date dot (T2-1874)

A possible ConcurrentModificationException in the service description
registry has been removed (T2-1894)

Improved icons are used for most dialogs (T2-1914)

Workflow engine
Workflow runs are now removed from memory when deleted
(T2-1531 and T2-1596)

Job queues are cleared by the dispatch stack (T2-1597) and cross
product, dot product, strategy nodes and the crystalizer no longer cache (T2-1598,
T2-1599, T2-1601, T2-1602)

The memory footprint of references has been greatly reduced (T2-1600)

Some mySQL provenance problems have been fixed (T2-1605)

The remote health checker tries a GET if the HEAD fails (T2-1679)

Out-of-date annotations are removed when a workflow is serialized

A thread deadlock when running nested workflows in parallel has been
fixed (T2-1713)

The run-deletion thread no longer hangs (T2-1759)

The timeouts for the plugin manager etc. have been reduced (T2-1783
and T2-1784)

Workflow design
The links within the workflow explorer are now ordered (T2-238)

The name of the current workflow is shown on the diagram border
(T2-473); this makes editing a nested workflow less confusing (T2-791)

Undo now unselects a removed processor/port/merge (T2-517)

Validation errors are given if the tested output port involved in
making a service loop is not connected within the workflow. (T2-641)

The iteration strategy editing has been improved (T2-813)

Can now drag a link to a merge (T2-845)

The details panel attempts to show the same type of information as
previously shown (T2-1489)

Nested workflow labels are now shown corrected (T2-1654)

Opening a workflow that uses HTTPS no longer hangs the UI (T2-1659)

An empty workflow is no longer indicated in the workflow explorer as
having problems (T2-1717)

Selection of the next annotation field has been improved (T2-1730)

The zoom-related actions now correctly change with the selection of a
workflow (T2-1813)

An exception is now shown when addition of a service to the workflow
fails (T2-1850)

Larger example values can now be specified (T2-1872)

Import workflow no longer throws null pointer exceptions (T2-1915)

Running workflows
Merging of lists to feed a service that expects a single input value
now works (T2-1423)

Pausing a workflow now changes the icon to paused (T2-1530)

Workflow run progress monitor no longer stores a node for each service
invocation (T2-1603)

Suggested solutions for failed HTTPS client authentication have been
added (T2-1738 and T2-1740)

The credential manager now delays asking for passwords until it is
needed (T2-1743)

Users can specify the character encoding and data type of input data

When the first attempt to give a password for the credential manager
fails, it now accepts a correct password (T2-1907)

You can now edit the executed workflow that caused a workflow run

Deleting a workflow run now moves to the next entry (T2-1323)

The save data dialog has been improved (T2-1498)

The ordering of data for previous runs is correct (T2-1514)

Expansion triangles are shown for top-level data and invocations

The split pane jumps around less (T2-1516)

Result renderers
Tick box for wrapping result text (T2-633)

Save as OPM and Save as Janus are now available for a workflow run

The hex renderer is now read-only (T2-1569)

The rendering of error documents containing HTML has been improved

A PDF and an improved HTML (open in browser) renderer have been added

Virtual machine objects can now be rendered as their string
representation (T2-1859)

Beanshell/RShell scripts
The scripts now accept Alt-keystrokes on the Mac (T2-425)

The Beanshell interpreter now clears its variable bindings (T2-1480)

Existing scripts are shown when loading and saving (T2-1499)

The text of the check script dialog has been improved (T2-1664)

R expressions can now be passed (T2-1695)

R services now use the credential manager to handle authorization

BioMart services
A NullPointerException is no longer thrown when datasets referenced by
filters are missing not when attributes are missing (T2-1513, T2-1913)

BioMoby services
The BioMoby port names are now coped with better in provenance
(T2-602 and T2-1529)

BioMoby no longer pops up an error message at startup (T2-1496)

The BioMart service provider now loads its services in chunks (T2-1785)

Local services
Select and Choose LocalWorkers now throw an exception for an empty
list of values (T2-298)

Some local services were changed to close streams (T2-1684)

"Get image from URL" no longer reads contents twice (T2-1685)

A new "Transform XML with parameters" service has been addded

"Read text file" can now take an encoding (T2-1750)

SoapLab services
The URL dialog now copes with the lack of a trailing / (T2-1787)

WSDL services
Operations that Taverna could not run are no longer shown in the
service panel (T2-1116)

XML attributes of top-level SOAP elements are no longer stripped out

0 blueprints and 0 bugs targeted

There are no feature specifications or bug tasks targeted to this milestone. The project's maintainer, driver, or bug supervisor can target specifications and bug tasks to this milestone to track the things that are expected to be completed for the release.

This milestone contains Public information
Everyone can see this information.