UNIX for Dummies Questions & Answers

Hi,

While comparing primary key data of two tables thr bteq script I am getting this Error. This script is a shell script.

*** Error: The following error was encountered on the output file.
Script.sh[485]: 3043492 Segmentation fault(coredump)


Please let me know how to get through it.


Monika

The [485] is the line number of the shell script, where the problem is occuring. I guess it's calling some binary program that might have an error trying to allocate or write into memory, but that is a different thing.

Actually at line 485 its simply reading the name of table entered by user and nothing else.
This error occurs whenever I include the code to match Primary key data.
If I remove this code it works fine.

Please let me know how to get thr this

You wrote your own answer -> check you matching code for those keys. Maybe put a "set -x" in front of the the line where the error occurs and check what it does. Or echo your variable names to see what's stored in them.
I can't try to fix it, because I have no access to your code since you did not post it, nor am I sure if I could So the next logical thing would be to fix that code or ask someone who can fix it.

GL

Hi,

Thanks for the Reply.
I am getting the same Error.Table name is correct.

Code on line no. 485 is:

table1_entry()
{
new_table=''
while [ -z "$new_table" ]
do
echo " "
echo "Enter name of table 1:"
read new_table
echo "t2::$new_table"
done

}


If I remove this functionality then I am getting correct ans without any Errors.But if I am comparing data for Primary key then I am getting that Error. there is additional code in this function to match other columns which r not primary key.It works fine for them.
key_query()
{

bteq<<-EOD
.logon $PASSWORD;

.set width 64000
.export report file=$path/match1.msq
.titledashes off

((select $pk_key
from $new_database.$new_table
minus
select $pk_key
from $base_database.$base_table) UNION
(select $pk_key
from $base_database.$base_table
minus
select $pk_key
from $new_database.$new_table));

.export reset

}

In the function "table1_entry()":
Looks like it is missing another " or some other value inside "". But I guess that's not the reason for a segmentation fault - the script should just hang and wait for a complete value.
Interessting might be the calling of that function "table1_entry()" too. If the function is causing of this little typo up there unusable data, the binary that is processing it, might produce the segm. fault.

Fix that new_table=" and let's see if it works.