Options

(1) To prevent jobs started by the Input Agent from running unnoticed over a long period of time, you can initiate a notification by email. The SMTP server to be used as well as the email address of the sender of the error email is specified in entry enableErrorMail in the configuration file ./etc/startup.xml. If the field has a value greater than 0, an email is sent if the calculated start time of the next job plus the value of the input field is less than the current time.

(2) If an absolute runtime is set here, the start time of the current job is used in (1).

(3) If the checkbox is set, an error is generated if a job is started and no data is available. See details in section Error Handling.

(4) In the case of a connection error, an error email is sent. See details in section Error Handling.

(5) If the Input Agent has started at least one job, a subsequent (active) profile can be triggered after the last job. This profile is also triggered if no job has gone through successfully. Important note: The subsequent profile is triggered once by this profile. The triggered profile must have a time-driven Input Agent (triggered by another profile), profiles with Input Agent Message cannot be used. Variables with prefix MSG_CALL_ cannot be accessed in triggered profiles (variables are not yet initialised in the triggering profile in phase 1), unless this profile here is also triggered by another profile. Its MSG_CALL_ variables can then be passed through. See also sections Profile Chains and Variables.

Note: If the option Activate parallel processing (item 10) is set, the selected profile is called when the last file to be processed has been transferred to the system. Thus, it can no longer be guaranteed that all data to be processed has already been processed before the start of the selected profile.

(6) If an error occurred during the fetching or processing of the data, you can specify a profile that should then be called. Example: A profile is supposed to fetch data from a customer via FTP. If the customer's FTP server is not reachable during the profile run, an error is generated. You could specify a subsequent profile for the error case to query an alternative server of the customer. Important note: The subsequent profile is triggered once by this profile. The triggered profile must have a time-driven Input Agent (triggered by another profile), profiles with Input Agent Message cannot be used. Variables with prefix MSG_CALL_ cannot be accessed in triggered profiles (variables are not yet initialised in the triggering profile in phase 1), unless this profile here is also triggered by another profile. Its MSG_CALL_ variables can then be passed through. See also sections Profile Chains and Variables.

(7) Timestamp of the last profile run. The default value before the first run is 01.01.70. If you click Request, the actual timestamp of the last profile run is entered. You can manually change and save the value. In the query, you can access this timestamp with placeholder @0:t@. This allows you to generate a condition like ...WHERE act_time > @0:t@ in an SQL statement. The timestamp is updated with each run of the profile, but only if the placeholder is used in the query. The timestamp will also be updated if the cron job ran, but no job was generated. Only if there was an error in phase 1, the timestamp will not be changed. Note: Alternatively, @0:d@ can be used for the date of the last run and @0:s#<template>@ stands for the formatted string of the date. Example: @0:s#yyyyMMdd@ returns 20150310 at runtime if the cron job ran last at 10.3.2015.

(8) For data sources File, FTP (not with option via DMZ), SCP and WebDAV you can select a file function class, which, for example, can check for the presence of another file belonging to the source file. If the check fails, no job is created. Note: A documentation of a selected class can be obtained via the question mark icon.