Test Cases Fail on Solaris When Drizzle is Compiled with Debug Info (starting from revision 1060)

Bug #387619 reported by Padraig O'Sullivan
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Drizzle
Fix Released
Critical
Monty Taylor

Bug Description

While testing a patch tonight on Solaris, I noticed a number of test cases failing which should not have been affected by my patch. Thus, I tried the test suite on trunk and I've noticed that a number of test cases fail on Solaris when Drizzle is compiled with debugging information.

The test cases which fail are:

- alter_table
- drizzledump
- frm
- heap
- myisam
- mysqldump-max
- rename

These tests pass when Drizzle is not compiled with debug information. I've only noticed this behavior on Solaris. I reverted the tree back to determine when these failures started happening and I've found that this behaviour was introduced in revision 1060.

As an example of the kind of things that are failing, the following scenario is taken from the alter_table test case:

$ ./dtr --start-and-exit
$ ../client/drizzle --port=9306
posulliv@drizzle-os:~/src/trunk/tests$ ../client/drizzle --port=9306
Welcome to the Drizzle client.. Commands end with ; or \g.
Your Drizzle connection id is 2
Server version: 2009.06.1063-trunk Source distribution

Type 'help;' or '\h' for help. Type '\c' to clear the buffer.

drizzle> use test;
Database changed
drizzle> create table t1 (id int);
Query OK, 0 rows affected (0.01 sec)

drizzle> show table status;
+------+--------+---------+------------+------+----------------+-------------+-----------------+--------------+-----------+----------------+---------------------+-------------+------------+-----------------+----------+----------------+---------+
| Name | Engine | Version | Row_format | Rows | Avg_row_length | Data_length | Max_data_length | Index_length | Data_free | Auto_increment | Create_time | Update_time | Check_time | Collation | Checksum | Create_options | Comment |
+------+--------+---------+------------+------+----------------+-------------+-----------------+--------------+-----------+----------------+---------------------+-------------+------------+-----------------+----------+----------------+---------+
| t1 | InnoDB | 0 | Compact | 0 | 0 | 16384 | 0 | 0 | 0 | NULL | 2009-06-16 16:16:55 | NULL | NULL | utf8_general_ci | NULL | | |
+------+--------+---------+------------+------+----------------+-------------+-----------------+--------------+-----------+----------------+---------------------+-------------+------------+-----------------+----------+----------------+---------+
1 row in set (0.01 sec)

drizzle> show table status like 't1';
Empty set (0 sec)

drizzle>

The second SHOW command is expected to have the same output as the first SHOW command.

-Padraig

Changed in drizzle:
assignee: nobody → Monty Taylor (mordred)
importance: Undecided → Critical
milestone: none → aloha
status: New → Confirmed
Revision history for this message
Lee Bieber (kalebral-deactivatedaccount) wrote :

Compiled with debug options ( --with-debug) on OpenSolaris intel and don't see the problem any more as of build 1103. Machine was drizzle-os.

Stopping All Servers
All 224 tests were successful.
The servers were restarted 17 times
Spent 422.459 of 472 seconds executing testcases

Changed in drizzle:
status: Confirmed → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.