Bug #114780 | lost connection in MySQL NDB cluster | ||
---|---|---|---|
Submitted: | 25 Apr 2024 8:48 | Modified: | 25 Apr 2024 12:15 |
Reporter: | CunDi Fang | Email Updates: | |
Status: | Duplicate | Impact on me: | |
Category: | MySQL Cluster: Cluster (NDB) storage engine | Severity: | S2 (Serious) |
Version: | 8.0.35-cluster MySQL Cluster Community S | OS: | Any (20.04) |
Assigned to: | MySQL Verification Team | CPU Architecture: | Any |
[25 Apr 2024 8:48]
CunDi Fang
[25 Apr 2024 8:56]
CunDi Fang
There are a surprising number of such similar SELECT statements, and I haven't found anything in common. The shortest one I've found so far is this SELECT statement: Poc: ``` select subq_0.c0 as c0, ref_2.column4 as c1, ref_2.column5 as c2, subq_0.c0 as c3, (select exec_count from sys.x$statements_with_full_table_scans limit 1 offset 2) as c4, subq_0.c0 as c5 from (select ref_1.column7 as c0 from mytest90.test3 as ref_0 inner join mytest90.test7 as ref_1 on ((ref_1.column6 is not NULL) or (true)) where cast(coalesce(ref_0.column2, cast(null as float)) as float) is NULL limit 113) as subq_0 left join mytest90.test7 as ref_2 on (false) where subq_0.c0 is not NULL limit 97; ``` Because the errors behave so similarly, I'll just write them directly in this comment, instead of opening a separate bug list.
[25 Apr 2024 9:32]
CunDi Fang
And this one: Poc: ``` select subq_0.c1 as c0, ref_0.column8 as c1, subq_0.c0 as c2 from mytest90.test3 as ref_0 right join (select 3 as c0, ref_2.column2 as c1 from mytest90.test8 as ref_1 inner join mytest90.test7 as ref_2 left join mytest90.test8 as ref_3 on (false) on (ref_1.column6 is NULL) where true limit 65) as subq_0 on (subq_0.c0 is NULL) where subq_0.c1 is NULL; ```
[25 Apr 2024 12:15]
MySQL Verification Team
Please use ndb_error_reporter to collect all the required log files https://dev.mysql.com/doc/refman/8.0/en/mysql-cluster-programs-ndb-error-reporter.html Duplicate of Bug #114464