Details
-
Bug
-
Status: Open (View Workflow)
-
Major
-
Resolution: Unresolved
-
10.6
Description
CURRENT_TEST: sys_vars.slave_parallel_threads_basic
|
--- /source/mysql-test/suite/sys_vars/r/slave_parallel_threads_basic.result 2024-08-22 05:32:43.941289817 +0000
|
+++ /build/mysql-test/var/1/log/slave_parallel_threads_basic.reject 2025-03-26 00:36:25.576440837 +0000
|
@@ -1,6 +1,6 @@
|
SET @save_slave_parallel_threads= @@GLOBAL.slave_parallel_threads;
|
SELECT IF(COUNT(*) < 20, "OK", CONCAT("Found too many system user processes: ", COUNT(*))) FROM information_schema.processlist WHERE user = "system user";
|
-IF(COUNT(*) < 20, "OK", CONCAT("Found too many system user processes: ", COUNT(*)))
|
+Name_exp_1
|
OK
|
SELECT @@GLOBAL.slave_parallel_threads as 'must be 20 because of .cnf';
|
must be 20 because of .cnf
|
@@ -17,6 +17,6 @@
|
@@GLOBAL.slave_parallel_threads
|
10
|
SELECT IF(COUNT(*) < 10, "OK", CONCAT("Found too many system user processes: ", COUNT(*))) FROM information_schema.processlist WHERE user = "system user";
|
-IF(COUNT(*) < 10, "OK", CONCAT("Found too many system user processes: ", COUNT(*)))
|
+Name_exp_1
|
OK
|
SET GLOBAL slave_parallel_threads = @save_slave_parallel_threads;
|
|
|
Result length mismatch
|
Attachments
Issue Links
- is part of
-
MDEV-27691
make working view-protocol
-
- Closed
-
Activity
| Field | Original Value | New Value |
|---|---|---|
| Link |
This issue is part of |
yep that's the general fix.
Name_Exp_X is from sql/sql_view.cc.
Most are fixed, its just the odd test or two that have been added/modified without view protocol being in CI.
| Assignee | Daniel Black [ danblack ] |
| Component/s | Replication [ 10100 ] |
danblack Looks like you just need to add an AS ProcessCheck explicit column name to the SELECT to make it the same in view-protocol and normal (I'm guessing - not familiar with view-protocol).
But in general, I think a lot of test cases would fail like this if the column names of the SELECT outputs differ in view-protocol from other modes. Where does this "Name_exp_1" name come from, why doesn't it give the full expression from the SELECT?
So this seems like something that needs to be fixed generally, not for a specific test case only?
Hope this helps,
- Kristian.