Data service tasks are executable integration processes that
contain a service published from Talend Studio.
Before you begin
Data service tasks can be executed
only on remote engines that have a microservice or
Talend Runtime (OSGi) run profile
configured.
About this task
You can deploy and undeploy data
services manually through the user
interface. Data services cannot
be scheduled or added to plans; this type of service runs until it is explicitly
undeployed by the operator.
Procedure
-
Open the Management page.
-
Go to the Tasks tab.
-
Select the environment and the workspace using the faceted
search at the top of the page.
-
Click Add task.
-
Select the Data service
artifact type.
-
Select the artifact you want to deploy from the drop-down
list.
The details and description of the artifact
appears.
-
Select the version
of the artifact to be used in the task.
You can select
Always use the latest available artifact
version if you want your task to be
updated automatically each time a new version of the
artifact is published. You can also select a specific
version and update it manually as needed.
The
details and description of the artifact are refreshed
automatically.
-
Select the
workspace for the task.
-
Give a name to the
new task.
By default, the
task name is the name of the artifact from which it is created.
Note: Tasks must have a unique name within a
workspace.
- Optional:
Add a tag to the
task.
-
Select an existing tag from the drop-down list.
-
Create a new tag by typing in the name and clicking on
it in the drop-down list.
- Optional:
Enter the description of the task.
-
Click Continue.
-
In the Parameters step, enter the value for the parameters of
the data service.
In this step, when the Show order as
in Studio toggle is disabled, you can
only set the values to parameters you have defined with the
parameter_ prefix in
Studio.
Note: Use \\ instead of \ when the
value could be interpreted as a special character, for example,
\t.
- Optional:
In the Advanced parameters step, enter the values for the
parameters you have defined in Studio without prefixes, and
connection_ parameters that are not defined in the data
service. This
section is available when the Show order
as in Studio toggle is disabled.
Otherwise, all parameters are in the
Parameters step.
Note: Use \\ instead of \ when the
value could be interpreted as a special character, for example,
\t.
- Optional:
In the Connections step, select the appropriate connection
from the drop-down list or create a new connection based on the parameters already
defined in the artifact.
- Optional:
In the Resources step, select the appropriate
resource to use in the task.
-
In the Engine step, select the execution runtime on which to
deploy your task from the Engine drop-down list.
Option |
Description |
Remote Engine |
Data services and Routes can only be executed on Remote Engines that are
compatible with the used Studio version. For the compatibility matrix, see
Artifact compatibility for Cloud Engine and Talend Remote
Engine. Depending on the type of the task, the selected Remote
Engine must have the corresponding run profile set either when the engine is
created, or later in the Edit engine wizard
accessible via the path.
|
Remote Engine cluster |
All Remote Engines assigned to the cluster must be of the same version
(v2.8.4 or above). Depending on the type of the task, the Remote Engines in
the cluster must have the appropriate Microservice or Talend Runtime (OSGi)
run profile set. The engines must also be compatible with the used
artifact's version.
|
- Optional:
When deploying a microservice on a Remote Engine or cluster,
you can select a run profile.
You can choose a previously created run profile for the
selected Remote Engine or cluster to set the JVM
parameters for the task run.
This feature is supported from Remote Engine 2.12.4
onwards.
-
In the Metric Path field, enter the metric path to be used at
deployment time to retrieve metric data, for example, the number of requests at
runtime.
The default value
is the artifact name, and we recommend to keep this value to see the metric data
displayed on the
Task Details page. If this field is empty, no
metric information will be displayed on the page.
Tip: You can add the project name to the artifact name, if artifacts
with the same name exist in more than one project and are configured to run
on the same Talend Remote Engine/Talend Runtime.
- Optional:
When deploying a microservice on a Remote Engine or cluster,
you can run your task as another user by entering their
username in the Run as impersonated
user field.
This feature is supported only if the Remote Engine is
installed on Unix or similar environments.
This feature is supported from Remote Engine 2.12.4
onwards.
-
When deploying a
microservice, set the port assignment.
Option |
Description |
Automatic |
The port on which the microservice instance(s) will run
is defined by the system. |
Custom |
You can set a static port number on which to run all
instances of the microservice, making external load balancing
easier. |
-
If you have chosen
to deploy the task on a cluster, select the deployment strategy.
Option |
Description |
Parallel |
Deploy the tasks simultaneously to all the Remote
Engines of the cluster. |
Rolling |
Deploy the tasks one by one to the Remote Engines of
the cluster. |
-
Save your changes.
You can deploy the task from the Task Details page by
clicking Deploy.
Results
Your task is ready to run on
Talend Cloud Management Console.
Note:
Exponential back-off retry pattern has
been adopted to provide not only faster first and second
retries (1 minute and 3 minutes respectively after the
initially failed attempt), but also better global behaviour
on peak workload situation.
For example, when you try to
execute a task that is already running, the execution fails.
Talend Cloud Management Console retries
the execution with incrementally increased intervals, that is to
say, the first retry takes place at the end of 1 minute only, the
second one 2 minutes, then subsequently 4 minutes and 8 minutes.
If the fourth retry fails, you will get the error message
Exceeded the limit of deployment attempts: another
instance of the same task was running on the engine
at same time
. Try to run the task
later.