tpcc_start get segmentation fault after printing RAMP-UP TIME

Bug #1081821 reported by Adam Scott
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
percona-tools
New
Undecided
Unassigned

Bug Description

after running in gdb here is the context messages I receive

...
...
...
[New Thread 0x7ffff67a3700 (LWP 4067)]
[Thread 0x7ffff67a3700 (LWP 4067) exited]
[New Thread 0x7ffff67a3700 (LWP 4068)]

Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7ffff67a3700 (LWP 4068)]
my_stat (path=0x7ffff67a24d0 "/usr/local/mysql/share/charsets/Index.xml", stat_area=0x7ffff67a23a0, my_flags=0)
    at /home/mysql/mysql-5.5.28/mysys/my_lib.c:413
413 my_errno= errno;

I resolved this, inspired by the MySQL bug report at http://bugs.mysql.com/bug.php?id=50565 by

sudo ln -s /opt/mysql/5.5.28 /usr/local/mysql

I have mysql untarred to /opt/mysql/5.5.28 and running from that directory
I compiled tpcc_start from source after creating doing a cmake for source lieing in /home/mysql/mysql-5.5.28
and I have to do a
     export LD_LIBRARY_PATH=/home/mysql/mysql-5.5.28/libmysql

before running tpcc_load and tpcc_start

I

Revision history for this message
priyanka (pkc431) wrote :

Hi ,

1. Try to fix the bug by running the tpcc_start from the source directory where you make the file.
Since its just make and not make install, it requires you to run the benchmark from the source directory and not the mounted directory.
2. Also, try to kill the processes that are running for the tpcc_start.
   i.e "kill <PID>"
3. give the correct parameters, like the correct database and the number of warehouses.
First load the database and then start the benchmark.

Hope this helps.

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.