Search Results

Try:
LC_ALL=C tree

You provided two values for sid, is the second one correct?
my $sid = "dbd";
Please just post the exact text from your terminal including the commands executed.

Could you post the exact output from your terminal from both the successful and the failing execution?

No, as far as I know ...

Double check the SQL statement:
select tname from tab where rownum <6
Try to execute it from some other clinet (like sqlplus) with the user defined in the Perl script.

I suppose you'll need to change you LD_LIBRARY_PATH when switching between Oracle versions.
I would use a wrapper shell script to invoke the Perl script correctly,something like this:
#!/bin/bash
...

Do you get the error while executing the script interactively - on the command line - or you get it on a scheduled execution (via cron)?

Try addning the ORACLE SID:

my $ORACLE_SID="your_sid";
$ENV{ORACLE_SID}=$ORACLE_SID;

export is a shell builtin, remove the line that starts with "export" and change the declaration like this:
my $ORACLE_HOME = "/oracle/product/10.2.0";
$ENV{ORACLE_HOME} = $ORACLE_HOME;

No, you need:
my $ORACLE_HOME = "/oracle/product/10.2.0";
#You may also need to set the ORACLE SID
#my $ORACLE_SID="your_sid";

$ENV{ORACLE_HOME}=$ORACLE_HOME;
# Uncomment if you set ORACLE SID...

Just put them on top of your script. See orafaq (http://www.orafaq.com/wiki/Perl) "Setting the Oracle environment" for more.

The error indicates that the Perl script is executed in a different environment (most probably via cron). Make sure all necessary Oracle related variables are set before invoking the Perl script:...

You could try something like this:

arr1=(a1 b1 c1)
arr2=(a2 b2 c2)

a() {
typeset -n _my_arr1=$1 _my_arr2=$2
printf '%s\n' first: "${_my_arr1[@]}"
printf '%s\n' second:...

Following up on MadeInGermany's more standard sed solution,
I suppose the OP needs something like this:
sed '/^dns\.com$/ {N;N;d;}' /tmp/temp.txt

Try:
sed '/^dns\.com$/,+2d' /tmp/temp.txt

With Perl:
ls -lh | perl -e'
@l{ K, M, G, T, P, E, Z, Y } = ( 1 .. 8 );
print sort {
($aa) = $a =~ /(\w)\s+/;
($bb) = $b =~ /(\w)\s+/;
$l{$aa} <=> $l{$bb} || $a <=> $b
}<>'

Either use a VM with Linux and ask here or, if you choose Windows, post on the Oracle Forums (https://community.oracle.com/community/database/general_questions/installation).

You're welcome! Just use the correct forum :)

Yes, you can. You can also use Oracle VirtualBox on Windows to setup
a VM with Oracle Linux (or CentOS) and install Oracle Database on it.

Yes,
you may be able to install Oracle 11.2 XE (Express Edition) on Ubuntu,
but it's still not supported.
As the post suggests, you need to install oracle-xe-11.2.0-1.0.x86_64.rpm, not only the...

As I said, you should use a supported platform (not Debian/Ubuntu!).
You can use either the official documentation (http://docs.oracle.com/database/121/LTDQI/E54543-03.pdf) or the quick guides on...

You have installed only the client. You're not able to connect as there's no database ...
Oracle Database Server is not supported on Ubuntu, you can use a free supported
platform like Oracle Linux.

Check Troubleshooting Guide for ORA-12545 / TNS-12545 Connect failed because target host or object does not exist (Doc ID 553328.1) on MOS (https://www.unix.com/support.oracle.com).

For multi-line output use a loop:
zsh-5.0.2[user]% awk 'BEGIN {
while (("ls /etc | head" | getline _out) > 0)
print "Got:", _out
}'
Got: alternatives
Got: bash.bash_logout
Got:...

You can store the output of an external command invoked from inside awk,
but you should not use the system function for that:
zsh-5.0.2[user]% awk 'BEGIN { "date" | getline dt; print "Got", dt }'...