Maybe we can verify dataSource in ReadWrite-splitting rules，before create ReadWrite-splitting rules.

Feature Request

For English only, other languages will not accept.

Please pay attention on issues you submitted, because we maybe need more details. If no response anymore and we cannot make decision by current information, we will close it.

Please answer these questions before submitting your issue. Thanks!

Is your feature request related to a problem?

NO

Describe the feature you would like.

First of all, we can use DistSQL create two READWRITE_SPLITTING RULE in one logic schema.

eg:

CREATE READWRITE_SPLITTING RULE ms_group_0 (
WRITE_RESOURCE=write_ds,
READ_RESOURCES(read_ds_0,read_ds_1),
TYPE(NAME=random)
);
CREATE READWRITE_SPLITTING RULE ms_group_1(
WRITE_RESOURCE=write_ds,
READ_RESOURCES(read_ds_0,read_ds_1),
TYPE(NAME=random)
);

But when we use select in this logic schema,we can’t figure out how to route,because of we have same table in two rules.

And it is not legal, because we cant’t have two same table in one schema.

So,can we verify dataSource in ReadWrite-splitting rules? Ensures that a logical schema does not exist the same table and cannot be routed.

[strongduanmu]

@wsm12138 I found RuleConfigurationChecker a good place to add validation logic, but currently Proxy doesn’t use it. I will add a RuleConfigurationChecker call to the Proxy and migrate various rule checking logic to RuleConfigurationChecker.

[RaigorJiang]

Yes, by design, read resource and write resource may be the same.