Description of problem:
After change of socket location in /etc/my.conf the init script
/etc/init.d/mysqld is unable to establish whether the mysql server
has started

Version-Release number of selected component (if applicable):


How reproducible:
Always

Steps to Reproduce:
1. edit /etc/mysql and change socket option
2. restart mysqld : service mysqld restart
3.
  
Actual results:
mysql is started however the init reported FAILED after a 30 seconds timeout

Expected results:
report OK

Additional info:
Straightforward work around in init script:
specify --socket in mysqladmin call in start() function.
I have "hacked" this in the following way:

     75         # Spin for a maximum of N seconds waiting for the server to come up.
     76         # Rather than assuming we know a valid username, accept an "access
     77         # denied" response as meaning the server is functioning.
>>>  78         eval $(grep ^socket /etc/my.cnf)
     79         if [ $ret -eq 0 ]; then
     80             STARTTIMEOUT=30
     81             while [ $STARTTIMEOUT -gt 0 ]; do
>>>  82                 RESPONSE=`/usr/bin/mysqladmin -uUNKNOWN_MYSQL_USER
--socket=$socket ping 2>&1` && brea        k
     83                 echo "$RESPONSE" | grep -q "Access denied for user" && break
     84                 sleep 1
     85                 let STARTTIMEOUT=${STARTTIMEOUT}-1
     86             done
     87             if [ $STARTTIMEOUT -eq 0 ]; then
     88                     echo "Timeout error occurred trying to start MySQL
Daemon."
     89                     action $"Starting $prog: " /bin/false
     90             else
     91                     action $"Starting $prog: " /bin/true
     92             fi
     93         else
     94             action $"Starting $prog: " /bin/false
     95         fi

A better solution would be if mysql would read in the /etc/my.conf file like
/usr/bin/mysqld_safe does!!!! Couldn't find that option for mysqladmin though...
grrrrr mysql....