Yesterday, I increased the memory in my mysql server from
2 GByte to 4 GByte.
Here the log file as it died:
Number of processes running now: 1
mysqld process hanging, pid 1015 - killed
020808 09:40:12 mysqld restarted
020808 9:40:12 Can't start server: Bind on TCP/IP port: Address already in use020808
9:40:12 Do you already have another mysqld server running on port: 3306 ?
020808 9:40:12 Aborting
020808 9:40:12 /usr/local/mysql/bin/mysqld: Shutdown Complete
020808 09:40:12 mysqld ended
The database was under heavy load at the time. The machine is
a 'vanilla' dual 1 GHz P-III running RedHat Linux and a 2.4.17 kernel
Server version: 3.23.49a-max-log
After doing the memory upgrade, I also adjusted some of the
parameters, in particular, I increased the key buffer to 2 Gig.
The database has a very large (100,000,000 row, 30 GigByte)
MyISAM table, and a few smaller myisam and innodb tables. It ran
ok for at least half a year (last machine reboot). So far, it does
not look like any data was corrupted.
Usually, only a small number of client (10 or so) is connected.
Here my my.cnf file:
set-variable=max_connections=200
set-variable=max_allowed_packet=10M
set-variable=key_buffer=2000M
set-variable=join_buffer_size=200M
set-variable=sort_buffer=150M
set-variable=record_buffer=50M
set-variable=record_rnd_buffer=50M
set-variable=table_cache=200
set-variable=thread_concurrency=4
set-variable=long_query_time=5
innodb_data_home_dir=/usr/local/mysql/innodb
innodb_data_file_path=data1/ibdata1:10000M;data2/ibdata1:10000M
set-variable=innodb_buffer_pool_size=50M
set-variable=innodb_additional_mem_pool_size=10M
innodb_log_group_home_dir=/opt/innodb_logs
innodb_log_arch_dir=/opt/innodb_logs
innodb_log_archive=1
set-variable = innodb_log_files_in_group=3
set-variable = innodb_log_file_size=3M
set-variable = innodb_log_buffer_size=100
innodb_flush_log_at_trx_commit=1
set-variable = innodb_file_io_threads=4
set-variable = innodb_lock_wait_timeout=50
