Index: trunk/ippMonitor/supervision/check_if_mysql_slave_is_running_fine.py
===================================================================
--- trunk/ippMonitor/supervision/check_if_mysql_slave_is_running_fine.py	(revision 34015)
+++ trunk/ippMonitor/supervision/check_if_mysql_slave_is_running_fine.py	(revision 34015)
@@ -0,0 +1,78 @@
+#!/usr/bin/env python
+
+# This script is supposed to be run on a host which is running a mysql
+# replication slave. It should be started from the crontab every
+# hour(?).
+#
+# This script makes sure that the mysql replication slave is fine on
+# the localhost. If there is any problem, that is, if the replication
+# is broken because of an error or if the lag between the master and
+# the server is too important (see the MAX_LAG_SECONDS parameter),
+# then an e-mail is sent to the ps-ipp-ops mailing list (actually the
+# RECIPIENT parameter).
+#
+
+import MySQLdb
+import smtplib
+from email.mime.text import MIMEText
+import socket
+
+def send_email(recipient, subject, body):
+    message = MIMEText(body)
+    message['Subject'] = subject
+    _from = 'ipp@ifa.hawaii.edu'
+    message['From'] = _from
+    _to = [recipient]
+    message['To'] = recipient
+    SMTP_CONNECTION = smtplib.SMTP('ippc18.ifa.hawaii.edu')
+    SMTP_CONNECTION.sendmail(_from, _to, message.as_string())
+    SMTP_CONNECTION.quit()
+
+if __name__ == '__main__':
+    # Configuration parameters
+    MAX_LAG_SECONDS = 1000
+    RECIPIENT = 'schastel@ifa.hawaii.edu'
+
+    # Don't change anything after this
+    db = MySQLdb.connect('localhost', 'root')
+    cursor = db.cursor()
+    cursor.execute('SHOW SLAVE STATUS')
+    ( Slave_IO_State, Master_Host, Master_User, Master_Port,
+      Connect_Retry, Master_Log_File, Read_Master_Log_Pos, Relay_Log_File,
+      Relay_Log_Pos, Relay_Master_Log_File, Slave_IO_Running, Slave_SQL_Running,
+      Replicate_Do_DB, Replicate_Ignore_DB, Replicate_Do_Table, Replicate_Ignore_Table,
+      Replicate_Wild_Do_Table, Replicate_Wild_Ignore_Table, 
+      Last_Errno, Last_Error, Skip_Counter, Exec_Master_Log_Pos,
+      Relay_Log_Space, Until_Condition, Until_Log_File, Until_Log_Pos,
+      Master_SSL_Allowed, Master_SSL_CA_File, Master_SSL_CA_Path, Master_SSL_Cert,
+      Master_SSL_Cipher, Master_SSL_Key, Seconds_Behind_Master) = cursor.fetchall()[0]
+    if Seconds_Behind_Master is None:
+        nodename = socket.gethostbyaddr(socket.gethostname())[0].split('.')[0]
+        if Last_Error != '':
+            send_email(RECIPIENT, 
+                       'MySQL replication problem on %s' % (nodename),
+                       """Last_error value when executing SHOW SLAVE STATUS on that host is:
+%s
+""" % (Last_Error) )
+        else:
+            send_email(RECIPIENT, 
+                       'MySQL replication problem on %s' % (nodename),
+                       """Slave_IO_State value when executing SHOW SLAVE STATUS on that host is:
+%s
+""" % (Slave_IO_State) )
+            print 'Replication problem: %s (is master down?)' % Slave_IO_State
+    elif Seconds_Behind_Master > MAX_LAG_SECONDS:
+        nodename = socket.gethostbyaddr(socket.gethostname())[0].split('.')[0]
+        send_email(RECIPIENT, 
+                   'MySQL replication is late on %s' % (nodename),
+                   """The Seconds_Behind_Master value when executing the 'SHOW SLAVE STATUS' mysql statement on that host is:
+   %s
+
+For info, the threshold is %s seconds.
+
+Check if there is a blocking statement executing on the slave
+""" % (Seconds_Behind_Master, MAX_LAG_SECONDS) )
+    else:
+        pass
+    cursor.close()
+    db.close()