I'm in favour of moving forward, possibly even to 8.0. But it seems that Cloud
SQL is tardy and haven't supported a new major version even though it has been
out for almost two years: https://cloud.google.com/sql/docs/db-versions
I think given Cloud SQL now defaults to 5.7 for MySQL that 5.7 making 5.6
"unsupported" should be fine, Unsupported in the case of "we'll accept patches
to fix 5.6 behavoiur, but don't explicitly test it".
The other question is do we want to do anything about mariadb etc versions? (I
don't know how they map on to mysql versions, or if they even do.) I'm def not
in favour of increasing the number of build jobs we have!
But then I'm not a MySQL user, so happy to be ignored here.
-a
On Feb 24 2020, at 1:27 pm, Jarek Potiuk <[email protected]> wrote:
> Currently we are running our tests for MySQL using both client and server
> of mysql in 5.6 version and I wonder how we should proceed with it?.
>
> Is there any reason why we would stick to 5.6 and not use 5.7 (or maybe we
> should do both)? While 5.6 is still "alive", it's been released 7 years ago
> and it's end of support is February 2021
>
> Here is some information about the released (from
> https://en.wikipedia.org/wiki/MySQL)
>
> Release General availability Latest minor version Latest release End of sup.
> 5.1 - November 14, 2008; 11 years ago 5.1.73 2013-12-03 December 2013
> 5.5 - December 3, 2010; 9 years ago 5.5.62 2018-10-22 December 2018
> 5.6 - February 5, 2013; 7 years ago 5.6.47 2020-01-13 February 2021
> 5.7 - October 21, 2015; 4 years ago 5.7.29 2020-01-13 October 2023
> 8.0 - April 19, 2018; 22 months ago 8.0.19 2020-01-13 April 2026
>
> What do you think ? Should we move to 5.7 as a base with 2.0? Or should we
> test both?
>
> J.
> --
> Jarek Potiuk
> Polidea | Principal Software Engineer
>
> M: +48 660 796 129
