Since a week I have a problem with my DB's replication, the problem is that there is difference between Read_Master_Log_Pos and Exec_Master_Log_Pos. This doesn't occur always, by momments all works fine, but there are other that don't. Weird here is that we wait some minutes and all work fine again in some other case I have to resynchronized it.

Mysql v. 5.1.4 Centos 5.8

mysql -e "Show slave status\G;"

*************************** 1. row ***************************
           Slave_IO_State: Waiting for master to send event
              Master_Host: IP Master
              Master_User: repl
              Master_Port: 3306
            Connect_Retry: 60
          Master_Log_File: remote-bin.000183
      Read_Master_Log_Pos: 50996883
           Relay_Log_File: local-relay-bin.000003
            Relay_Log_Pos: 444613
    Relay_Master_Log_File: remote-bin.000183
         Slave_IO_Running: Yes
        Slave_SQL_Running: Yes
          Replicate_Do_DB: xxxDB
      Replicate_Ignore_DB:
       Replicate_Do_Table:
   Replicate_Ignore_Table:
  Replicate_Wild_Do_Table:
  Replicate_Wild_Ignore_Table:
               Last_Errno: 0
               Last_Error:
             Skip_Counter: 0
      Exec_Master_Log_Pos: 50949964
          Relay_Log_Space: 868147
          Until_Condition: None
           Until_Log_File:
            Until_Log_Pos: 0
       Master_SSL_Allowed: No
       Master_SSL_CA_File:
       Master_SSL_CA_Path:
          Master_SSL_Cert:
        Master_SSL_Cipher:
           Master_SSL_Key:
    Seconds_Behind_Master: 2501
Master_SSL_Verify_Server_Cert: No
            Last_IO_Errno: 0
            Last_IO_Error:
           Last_SQL_Errno: 0
           Last_SQL_Error:

Here mys /etc/my.cnf

[mysqld]
datadir=/var/lib/mysql
socket=/tmp/mysql.sock
port = 3306
user=mysql
log-error=/var/log/mysqld.log
pid-file/var/run/mysqld.pid
log-bin=local-bin
max_connections=350
expire_logs_days = 2

ignore_builtin_innodb
plugin-load=innodb=ha_innodb_plugin.so;innodb_trx=ha_innodb_plugin.so;

innodb_data_home_dir = /var/lib/mysql/
innodb_data_file_path = ibdata1:50M:autoextend
#skip-external-locking


key_buffer_size = 8M
max_allowed_packet = 128M
table_cache = 2048
table_definition_cache = 1024
sort_buffer_size = 8M
read_buffer_size = 8M
read_rnd_buffer_size = 8M
myisam_sort_buffer_size = 64M
thread_cache_size =32
query_cache_size= 316M
thread_concurrency = 1
join_buffer_size = 5M


wait_timeout=50
#query_cache_type=1
interactive_timeout=60
connect_timeout=15

#wait_timeout  = 10000000

server-id=103
replicate-do-db = xxxDB
binlog-do-db=xxxDB

relay-log=local-relay-bin

#log-slow-queries
low_priority_updates=1
#log-queries-not-using-indexes
tmpdir=/var/tmpfs

max_heap_table_size = 20K
tmp_table_size = 20K

#long_query_time         = 5
#log-slow-queries        = /var/log/mysql/mysql-slow.log

log-queries-not-using-indexes
#basedir=/usr/local

slave-skip-errors = 1054
[client]
socket=/tmp/mysql.sock
port = 3306    

What could be the problem?

Please let me know if you need anything more

Yes I know... it's an all system that have a very old web system... but it worked fine :) ...until now :(