What you wrote is 'a one to nine digits, then space, then the word "failed"'.

No. /\d{1-9}/ matches a digit followed by a "{" followed by a "1" followed by a "-" followed by a "9" followed by a "}".

To match any number (for some definition of number), /\d+/ would do.

To make this answer a little more helpful:
The correct expression to match "a series of 1 to 9 digits" is:
/\d{1,9}/ or /[0-9]{1,9}/ - note the comma, not dash

1 to 9 digits between 1 and 9:
/[1-9]{1,9}/

Any number of digits between 1 and 9: /[1-9]+/

Oh, thank you, my fault; I saw comma inplace of dash

Actually the OP's open:

open (FILE,$file) || die ("cant open file");
[download]

is correct, but old school and frowned on. As you suggest the three parameter version of open is better, but even better is to use it with a lexical file handle. Using a lexical file handle ensures the file is closed when the file handle variable goes out of scope. The improved version of open is:

open my $fileIn, "<", $file or die "Can't open $file: $!";
[download]

Note too the file name mentioned in the die and the use of $! to give the system provided failure message. Both of those can help a lot when figuring out why an open failed.

All,
Thanks a lot
My requirement is such that if the log contains the string
like "2 failed" or "3 failed"..."<any number except 0>" failed.
I have some other processing to do. So will the below code work.

my $file = "D:/perl/02.log";
open my $inFile, '<', $file or die "cant open $file: $!";

# Look for errors
while (<$inFile>) {
    if (/\d{1-9} failed/) {
        print "There are build errors (see line $.)\n";
        exit 0;
    }
}
[download]

It's a slightly tricky match to get right depending on just how fussy you really want to be. Using a negative look ahead helps cover most bases though. Consider:

use strict;
use warnings;

my $str = <<STR;
Errors: 0 failed
Errors: 01 failed
Errors: 1 failed
Errors: 10 failed
Errors: None failed
Ben10 failed
STR

open my $fIn, '<', \$str;

while (<$fIn>) {
    print if /\b(?!0+\b)\d{1,9} failed/;
}
[download]

Prints:

Errors: 01 failed
Errors: 1 failed
Errors: 10 failed
[download]

The (?!0+\b) is the negative look ahead that fails the match if a zero value (regardless of the number of 0 digits) is found. The \b's are there to ensure all the digits in the number are matched.

True laziness is hard work

Won't work. Use /[1-9] failed/ for that. \d{1-9} would actually parse a digit followed by '{1-9}' or a number with 1 to 9 digits if the '-' were a ','. More info in perlre.