MySQL Bugs: #14563: 5.0-max build hangs in mysqldump.test

Bug #14563	5.0-max build hangs in mysqldump.test
Submitted:	2 Nov 2005 6:09	Modified:	2 Dec 2005 17:02
Reporter:	Sergey Petrunya	Email Updates:
Status:	No Feedback	Impact on me:	None
Category:	MySQL Server	Severity:	S2 (Serious)
Version:	5.0-bk	OS:	x86 linux
Assigned to:		CPU Architecture:	Any

Description:
Current 5.0 build (tip cset: 1.1957.1.11, 2005-11-01 12:15:08-08:00, jimw@mysql.com) hangs in mysqldump.test when built with -max build.

The issue seems to be a race condition as it doesn't happen every time.
Connecting to the hung server and running "SHOW PROCESSLIST" yields this:

+----+------+-----------+------+---------+------+---------------------------------+------------------------+
| Id | User | Host      | db   | Command | Time | State                           | Info                   |
+----+------+-----------+------+---------+------+---------------------------------+------------------------+
|  1 | root | localhost | test | Query   |  121 | Waiting for release of readlock | drop table t1          |
| 72 | root | localhost | test | Query   |  121 | closing tables                  | show create table `v2` |
| 74 | root | localhost | NULL | Query   |    0 | NULL                            | show processlist       |
+----+------+-----------+------+---------+------+---------------------------------+------------------------+

"mysqladmin debug" hangs too, here are the only lines printed in server stderr:

Running threads: 3  Stack size: 196608
Current locks:
lock: 0x8eea454:

How to repeat:
compile 5.0 tree with BUILD/compile-pentium-debug-max
Run ./mysql-test-run t/mysqldump.test
See it hang.

I was not able to repeat with today's 5.0.16-BK (ChangeSet@1.1957.1.11, 2005-11-01 12:15:08-08:00, jimw@mysql.com) on my Fedora Core 1:

[openxs@Fedora mysql-5.0]$ cd mysql-test/
[openxs@Fedora mysql-test]$ ./mysql-test-run t/my
myisam-blob-master.opt      mysqlbinlog-master.opt
myisam-blob.test            mysqlbinlog.test
myisam.test                 mysqlbinlog2.test
mysql.test                  mysqlcheck.test
mysql_client_test.test      mysqldump-max.test
mysql_delimiter.sql         mysqldump.test
mysql_delimiter_source.sql  mysqlshow.test
mysql_protocols.test        mysqltest.test
[openxs@Fedora mysql-test]$ ./mysql-test-run t/mysqldump.test
Installing Test Databases
Removing Stale Files
Installing Master Databases
running  ../sql/mysqld --no-defaults --bootstrap --skip-grant-tables     --based
ir=. --datadir=./var/master-data --skip-innodb --skip-ndbcluster --skip-bdb
--language=../sql/share/english/ --character-sets-dir=../sql/share/charsets/
Installing Slave Databases
running  ../sql/mysqld --no-defaults --bootstrap --skip-grant-tables     --based
ir=. --datadir=./var/slave-data --skip-innodb --skip-ndbcluster --skip-bdb     -
-language=../sql/share/english/ --character-sets-dir=../sql/share/charsets/
Manager disabled, skipping manager start.
Loading Standard Test Databases
Starting Tests

TEST                            RESULT
-------------------------------------------------------
mysqldump                      [ pass ]
-------------------------------------------------------

Ending Tests
Shutting-down MySQL daemon

Master shutdown finished
Slave shutdown finished
All 1 tests were successful.

I built it using the following commands:

BUILD/autorun.sh
./configure --prefix=/home/openxs/dbs/5.0

(I simply has not enough space to build -max. But according to the parameters used upon startup, --skip-innodb --skip-ndbcluster --skip-bdb, I do not completely understand how -max may influent the results of this test.)

Have you any ideas about the possible reason for the failure in your cases? Or I just have to check on -max?

No feedback was provided for this bug for over a month, so it is
being suspended automatically. If you are able to provide the
information that was originally requested, please do so and change
the status of the bug back to "Open".