Description:
I do not find a category for "mysqldump", so I use "Server: general".
At first sight, the symptom looks very similar to bug#37938 (closed by now, and there is an entry that the fix got pushed into this version).
But the significant difference is the "Out of memory" line here:
=====
main.mysqldump [ fail ]
Test ended at YYYY-MM-DD HH:MM:SS
CURRENT_TEST: main.mysqldump
mysqldump: Couldn't find table: "non_existing" mysqldump: Got error: 1356: View 'test.v1' references invalid table(s) or column(s) or function(s) or definer/invoker of view lack rights to use them when using LOCK TABLES
mysqldump: Couldn't execute 'SHOW FIELDS FROM `v1`': View 'test.v1' references invalid table(s) or column(s) or function(s) or definer/invoker of view lack rights to use them (1356) ERROR 5 (HY000) at line 220: Out of memory (Needed 40800 bytes)
mysqltest: At line NNN: command "$MYSQL < $MYSQLTEST_VARDIR/tmp/bug21527.sql" failed
The result from queries just before the failure was:
< snip > use mysqldump_myDB;
create user myDB_User@localhost;
grant create, create view, select, insert on mysqldump_myDB.* to myDB_User@localhost;
create table t1 (c1 int); insert into t1 values (3);
use mysqldump_myDB;
create table u1 (f1 int);
insert into u1 values (4); create view v1 (c1) as select * from t1;
use mysqldump_myDB;
drop view v1;
drop table t1; drop table u1;
revoke all privileges on mysqldump_myDB.* from myDB_User@localhost;
drop user myDB_User@localhost; drop database mysqldump_myDB;
flush privileges;
# Bug#21424 continues from here.
# Restore. Flush Privileges test ends. #
More results from queries before failure can be found in /PATH/mysql-test/var/log/mysqldump.log
=====
In the build of cluster-6.2.17, I have this failure in all configurations on Solaris 8 and 9, Sparc, 64bit.
(Solaris 8 is not supported in this version, got built by accident, so Solaris 9 is the relevant platform).
The symptom is new: This "Out of memory" does not show in any other bug report mentioning "mysqldump" and "non_existing", bugs 28372, 32991, 33208, 36966, 37938, 43079.
Cluster-6.2.17 is based on the server version 5.1.32, but I could not find this symptom in the test logs of 5.1.32 on the affected hosts, so there is some chance the effect is specific to the CGE tree.
OTOH, from the symptom it seems no cluster-specific component which fails, so I am hesitant to file it as a cluster issue.
How to repeat:
Found by running the test suite in a cluster-6.2.17 build.
© 2024, Oracle Corporation and/or its affiliates