I wonder if there is a way to retrieve the full list of executed jobs by SQL?
Are those details (the same offered by the APIs) saved on a system table, taht I could be able to query?
My purpose is to create a report of some of the details from jobs (for example execution time) using SQL: is that possible in some way? I was not able to find a way to do it without relying on HTTP APIs…
Currently we can only retrieve the job details for a particular job via REST API using the job id. You should be able to search for a table_name ,SQL match on the jobs search via UI. This capability is not yet there through REST API
Are you looking for a system table like v$sql in Oracle that shows all the SQL run based of a SQL ID?
HI Bali thanks for the reply.
The idea to search by table_name is useful via API, but I’d like to have a similar approach available by SQL views/tables.
I mean: it could be very useful to have a dedicated system space (for example enabled by a configuration flag), in which all the job details were automatically saved. This way a user could query dremio jobs themselves using the same tools, and for example it could be possible to have report on those…
It should be possibile to implment this feature, because it’s something we can already do now by hand, I suppose retrieving the results before exposing them on REST API and save them in an internal space should be very easy to accomplish.
I agree, we would like to be able to run some reports showing: the most used/least used VDS we have, the most active users, the busiest time of day, the worst performing queries, the most hit reflections, the largest spaces/folders, the most used personal spaces, regular or reoccurring queries of a personal space (we want to limit this), and some other ideas to help us monitor and acknowledge our users and environment.
There are several new enhancement in the product road map and one of them is to have more system tables, please see the release notes of every release to track enhancements released