Sometimes you may wish to collect logging information for use in 3rd party reporting and monitoring systems, e.g. Splunk.
The following query gives the essential information for each task and log message:
Log Extract SQL
select
lf.name as task_name,
lf.start_dtm as task_start_dtm,
lf.end_dtm as task_end_dtm,
u.username,
concat(concat(lu.first_name,' '), lu.last_name) as local_user_name,
concat(concat(eu.first_name,' '), eu.last_name) as domain_user_name,
lf.complete_boo as task_complete_boo,
lm.message_type,
lm.message_sequence,
lm.message_dtm,
lm.message
from
log_file lf
left join log_message lm on lf.log_file_id = lm.log_file_id
left join cv_user u on u.user_id = lf.user_id
left join local_user lu on lu.user_id = lf.user_id
left join external_user eu on eu.user_id = lf.user_id
To run this query you will need read access to the following tables in the PhixFlow schema:
- log_file
- log_message
- local_user
- external_user
The following message types (m.message_type) are found in PhixFlow log files:
- WARNING
- DEBUG
- ERROR
- INFO
The task_name field in the results will be one of:
- A task plan name, if a task plan was run directly or under a schedule
- A stream name, if a Table was run directly (for example, from a model view)
- An action name, if an Table-Action triggered the task plan
This is simply what appears in the Name field of the Active Tasks and Completed Tasks windows in the System Console.