День добрый.
ос: Gentoo Linux(32Bit), kenel-2.6.22
машина: 2xXEON 3.2Ghz, 8Gb RAM, 2x73 SCSI
работает каквыделенный сервер mysql-5.0.44

проблема с доступом к серверу, в основном по сети
выпадает ошибка:
"Can't create a new thread (errno 11); if you are not out of available memory, you can consult the manual for a possible OS-dependent bug"

my.cnf
[mysqld]
character-set-server                            = latin1
default-character-set                           = latin1
user                                            = mysql
port                                            = 3306
socket                                          = /var/run/mysqld/mysqld.sock
pid-file                                        = /var/run/mysqld/mysqld.pid
log-error                                       = /var/log/mysql/mysqld.err
basedir                                         = /usr
datadir                                         = /var/lib/mysql
skip-locking
key_buffer                                      = 8M
max_allowed_packet                              = 1M
table_cache                                     = 64
sort_buffer_size                                = 512K
net_buffer_length                               = 128K
read_buffer_size                                = 256K
read_rnd_buffer_size                            = 512K
myisam_sort_buffer_size                         = 8M
language                                        = /usr/share/mysql/english
bind-address                                    = 0.0.0.0
log-bin
server-id                                       = 1
tmpdir                                          = /tmp/
innodb_buffer_pool_size         = 2G
innodb_additional_mem_pool_size = 64M
innodb_data_home_dir            = /var/lib/mysql/
innodb_log_arch_dir             = /var/mysqld/
innodb_log_group_home_dir       = /var/mysqld/
innodb_data_file_path           = ibdata1:1024M:autoextend
innodb_log_file_size            = 250M
innodb_log_buffer_size          = 16M
set-variable = innodb_log_files_in_group=2
innodb_flush_log_at_trx_commit = 1
innodb_lock_wait_timeout = 50

#
-mysqld(5362)-+-{mysqld}(5378)
        |              |-{mysqld}(5379)
        |              |-{mysqld}(5380)
        |              |-{mysqld}(5381)
        |              |-{mysqld}(5398)
        |              |-{mysqld}(5399)
        |              |-{mysqld}(5400)
        |              |-{mysqld}(5401)
        |              |-{mysqld}(5576)
        |              |-{mysqld}(5675)
        |              |-{mysqld}(5676)
        |              |-{mysqld}(5677)
        |              |-{mysqld}(5678)
        |              |-{mysqld}(5679)
        |              |-{mysqld}(5681)
        |              |-{mysqld}(5682)
        |              |-{mysqld}(9444)
        |              `-{mysqld}(9556)
Хотя в процессах виден только один mysqld, но потоков много...
На данный момент tcp-соединений к mysql - 10, да и вообще кол-во соединений не очень большое, параметр по умолчанию max_connection = 100...

Насколько удалось понять - проблема вызванна из-за памяти, неправильной выделения или утечки...
Если есть какие здравые мысли/идеи прошу поделитесь...