As discussed with Elkin. This (seemingly invalid) error is often seen during test runs:
|
10.8.0 347f6d01e3b570dce49aa1ab42cb83021905a14d (Optimized)
|
2022-01-18 23:23:40 4 [Note] BINLOG_BASE64_EVENT: Rolling back unfinished transaction (no COMMIT or ROLLBACK in relay log). A probable cause is that the master died while writing the transaction to its binary log, thus rolled back too. Internal MariaDB error code: 0
|
However, no replication is active, it is a single server, a single SQL thread, and only a simple testcase is needed to produce either this message and/or a crash on debug:
CREATE TABLE t (c CHAR(1));
|
XA START 'a';
|
SELECT dt DIV 1 AS f,id FROM t GROUP BY f;
|
BINLOG ' O1ZVRw8BAAAAZgAAAGoAAAAAAAQANS4xLjIzLXJjLWRlYnVnLWxvZwAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAA7VlVHEzgNAAgAEgAEBAQEEgAAUwAEGggAAAAICAgC ';
|
CREATE TABLE t (c INT KEY,c2 CHAR(1),c3 DATE);
|
XA START 'b';
|
SELECT INTc,INTc2,CHARc,CHARc2,CHARc3 FROM t WHERE id=0;
|
BINLOG ' SOgWTg8CAAAAbgAAAHIAAAAAAAQANS42LjMtbTUtZGVidWctbG9nAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAABI6BZOEzgNAAgAEgAEBAQEEgAAVgAEGggAAAAICAgCAAAAAAVAYI8=';
|
Optimized may see the message only, and debug the message and crash. The crash is after the message:
|
10.8.0 347f6d01e3b570dce49aa1ab42cb83021905a14d (Debug)
|
2022-01-18 23:23:36 4 [Note] BINLOG_BASE64_EVENT: Rolling back unfinished transaction (no COMMIT or ROLLBACK in relay log). A probable cause is that the master died while writing the transaction to its binary log, thus rolled back too. Internal MariaDB error code: 0
|
mysqld: /test/10.8_dbg/sql/sql_error.cc:335: void Diagnostics_area::set_ok_status(ulonglong, ulonglong, const char*): Assertion `!is_set() || (m_status == DA_OK_BULK && is_bulk_op())' failed.
|
As discussed, it would be great to see this one resolved asap as the message shows up all over the test results, and clogs runs up. I have marked it critical as such for the moment. Thank you!
Bug
Critical
