Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Enable SQL by default #7793

Closed
yurmix opened this issue May 29, 2019 · 2 comments · Fixed by #7808
Closed

Enable SQL by default #7793

yurmix opened this issue May 29, 2019 · 2 comments · Fixed by #7808

Comments

@yurmix
Copy link
Contributor

yurmix commented May 29, 2019

Description

According to #3682, SQL is disabled by default to save additional metadata queries by the broker (Per new segment, not per query).

However, I think that since SQL is no longer considered experimental, we should enable SQL by default (set the default of druid.sql.enable to true).

If there is any performance optimization needed to be done, we should probably do it anyways (if SQL is to be broadly adopted).

I admit this could lead to cluster operators getting unexpected SQL queries from end users. However, if clearly documented in release notes, operators can simply disable through configuration.

Motivation

  • Increase adoption by letting end users easier access to SQL.
  • Bring SQL one step closer to a first-class citizen.
@vogievetsky
Copy link
Contributor

It should also be noted that the new web console heavily relies on SQL and while the functionality is there to make it work without SQL having SQL enabled unlocks many of its features due to the heavy use of Druid system tables

@gianm
Copy link
Contributor

gianm commented May 30, 2019

I think the only real downside is that there will be slightly more load on historicals (due to extra metadata queries) and memory use on the broker will be somewhat higher (enabling SQL activates a metadata cache). But these effects should be small and I think they are worth it, since SQL is rad. So I think enabling it by default is good. By the way, most of our customers (at Imply) have SQL enabled.

SQL can only do things that native queries could also do (they simply translate into native queries) so I don't think cluster operators getting unexpected SQL queries would be a serious issue.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging a pull request may close this issue.

3 participants