MyonKeminta
MyonKeminta
cc @sticnarf @cfzjywxk @disksing
According to @disksing , DC-belonging of data is usually defined table by table, and it's not guaranteed that a region will never span over multiple tables. As I understood it,...
Is there any problem about the idea mentioned at first in this issue, which means to make async commit work for both local and global transactions?
Did you mean that the transaction's lock can be skipped if we did CheckSecondaryLocks on one if its keys that's not yet locked? Sounds correct, but I'm afraid that if...
I noticed that clients outside TiDB seem to be unable to check the schema version. So perhaps clients other than TiDB should be prohibited to resolve locks of async commit...
> Upgrading tools provided by us (like tidb-operator or tiup) can set these two variables to 1 after upgrading is finished. I think it's better if the user can choose...
@cfzjywxk @sticnarf @youjiali1995 @coocood PTAL. How do you think the future plan should be?
It's exciting that you agree landing it in 5.0. But I'm still afraid that there might still be other corner cases or compatibility issues I didn't notice.
@youjiali1995 Do you know if this affects TiFlash?
For ScanLock and PhysicalScanLock requests, it should update the max_ts, check the memory lock, and finally read the raftkv or the rocksdb. For checking memory locks, there are two possible...