Document query.reuse-connection configuration property #14870
Conversation
| @@ -51,3 +51,7 @@ connector: | |||
| Using a large timeout can potentially result in more detailed dynamic filters. | |||
| However, it can also increase latency for some queries. | |||
| - ``20s`` | |||
| * - ``query.reuse-connection`` | |||
There was a problem hiding this comment.
This is a kill switch, added only in case of bugs. I would prefer not document this and possibly remove in future once we see there are no issues related to this new feature.
There was a problem hiding this comment.
It was added to the release notes...
There was a problem hiding this comment.
I would say that if users run into issues, ideally they should be able to figure out how to work around it by looking at connector docs, not by looking at release notes.
If/when we decide to remove the property, it should be easy to also remove the docs at that juncture, yeah?
There was a problem hiding this comment.
I don't think we document all the kill switches.
There was a problem hiding this comment.
So how are users supposed to know about it, if we don't document or expose it? I think we should document them. And if we want to remove them at some stage .. we can.
Or are you saying you want to prevent users from knowing about it and/or using them. If so .. then why did we add it in the first place. Purely for developers?
There was a problem hiding this comment.
It is here, so in case we hear a feedback that it does not work then we may want to suggest to use it. I plan to remove the kill switch after some time if there is no feedback.
I we place this in documentation then we may not hear the feedback at all and we could just have bugs that people bypass while I would like to fix these bugs.
There was a problem hiding this comment.
the problem with documenting kill switches is that users enable those to avoid change (I know I did) and then you can never know if people actually use the new code and then one day you remove kill switch and people start filing bugs that thing was broken.
It's tricky to balance.
There was a problem hiding this comment.
So here is what I would do. Always do one or the other:
- implement a kill switch and document it
- don't implement a kill switch
If you need to later remove a kill switch .. you could leave it behind as a no-op with a warning for a while and then remove it.
Description
Docs for #14653 - added to JDBC fragment to propagate to all connector pages.
Non-technical explanation
Docs only.
Release notes
(x) This is not user-visible or docs only and no release notes are required.
( ) Release notes are required, please propose a release note for me.
( ) Release notes are required, with the following suggested text: