MySQL: Fix problem of starting up mysql 5.7.33 version when user is root

[seveneves]

Fixes following problem in console of mysql container

java.lang.IllegalStateException: Container did not start correctly.
	at org.testcontainers.containers.GenericContainer.tryStart(GenericContainer.java:435)
	at org.testcontainers.containers.GenericContainer.lambda$doStart$0(GenericContainer.java:325)
	at org.rnorth.ducttape.unreliables.Unreliables.retryUntilSuccess(Unreliables.java:81)
	at org.testcontainers.containers.GenericContainer.doStart(GenericContainer.java:323)
	at org.testcontainers.containers.GenericContainer.start(GenericContainer.java:311)

[ERROR] [Entrypoint]: MYSQL_USER="root", MYSQL_USER and MYSQL_PASSWORD are for configuring a regular user and cannot be used for the root user
    Remove MYSQL_USER="root" and use one of the following to control the root user password:
    - MYSQL_ROOT_PASSWORD
    - MYSQL_ALLOW_EMPTY_PASSWORD
    - MYSQL_RANDOM_ROOT_PASSWORD

Fixes issue #3893

[rnorth]

Thanks @seveneves
Please could you add some tests which demonstrate the problem and can guard against regressions?

#2689 is a draft PR I created to demonstrate #2627 - this might be something you could use as a basis.

[seveneves]

@rnorth
Hi. Please check my test

[rnorth]

I've noticed a bad (existing) bug when testing this locally... Firstly, I noticed that your testRootAccountUsageWithEmptyPassword test takes >2mins to run.

It seems that setting the password to null causes a NullPointerException here, which is in turn swallowed here.

The terrible thing about the second piece of code is that the connection check fails continuously until timeout is reached, but upon hitting timeout it just continues.

We're going to need to fix that; I think we'll also need to prevent null passwords, to. I will dig a little more and may add a commit to your PR. Please stand by.

[rnorth]

I've noticed a bad (existing) bug when testing this locally... Firstly, I noticed that your testRootAccountUsageWithEmptyPassword test takes >2mins to run.

It seems that setting the password to null causes a NullPointerException here, which is in turn swallowed here.

The terrible thing about the second piece of code is that the connection check fails continuously until timeout is reached, but upon hitting timeout it just continues.

We're going to need to fix that; I think we'll also need to prevent null passwords, to. I will dig a little more and may add a commit to your PR. Please stand by.

[seveneves]

It seems that setting the password to null causes a NullPointerException

@rnorth I noticed the slow test run but I didn't dig into the reason. Thanks for finding the reason.

I pushed a "fix" and replaced the null with an empty string. This, of course, is not a solution to the NPE but I'd like to have this PR merged separately from NPE problem. which would require a separate issue and PR.

What do you think?

[rnorth]

@seveneves yes sure, I was going to suggest that. I'll follow up with a change to get null-checks where they need to be.

If you could please add my suggested change, I think we're good.

[rnorth]

Approved, thanks for this @seveneves!
Fixes #3893, #2627