two log file comparison and display if needed...

Shell Programming and Scripting
1

Hi All,

i've written script to collect the tablespaces and the datafiles related to each tablespaces with free space available and directed to two logfiles.. ( pasted below)

  1. datafile_test.log
    PSAPSR3
    PSAPSR3700
    PSAPSR3USR

  2. freedata_test.log
    /oracle/PJ1/sapdata3/sr3_14/sr3.data14 PSAPSR3 640
    /oracle/PJ1/sapdata3/sr3_14/sr3.data14 PSAPSR3 114624
    /oracle/PJ1/sapdata1/sr3700_1/sr3700.data1 PSAPSR3700 50112
    /oracle/PJ1/sapdata1/sr3700_2/sr3700.data2 PSAPSR3700 33728
    /oracle/PJ1/sapdata1/sr3700_3/sr3700.data3 PSAPSR3700 14272
    /oracle/PJ1/sapdata1/sr3700_4/sr3700.data4 PSAPSR3700 17344
    /oracle/PJ1/sapdata2/sr3700_5/sr3700.data5 PSAPSR3700 15296
    /oracle/PJ1/sapdata2/sr3700_6/sr3700.data6 PSAPSR3700 256
    /oracle/PJ1/sapdata2/sr3700_7/sr3700.data7 PSAPSR3700 1984
    /oracle/PJ1/sapdata2/sr3700_8/sr3700.data8 PSAPSR3700 960
    /oracle/PJ1/sapdata3/sr3700_10/sr3700.data10 PSAPSR3700 3008
    /oracle/PJ1/sapdata3/sr3700_11/sr3700.data11 PSAPSR3700 960
    /oracle/PJ1/sapdata3/sr3700_12/sr3700.data12 PSAPSR3700 960
    /oracle/PJ1/sapdata3/sr3700_9/sr3700.data9 PSAPSR3700 960
    /oracle/PJ1/sapdata4/sr3700_13/sr3700.data13 PSAPSR3700 6080
    /oracle/PJ1/sapdata4/sr3700_14/sr3700.data14 PSAPSR3700 9152
    /oracle/PJ1/sapdata4/sr3700_15/sr3700.data15 PSAPSR3700 30656
    /oracle/PJ1/sapdata4/sr3700_16/sr3700.data16 PSAPSR3700 52160
    /oracle/PJ1/sapdata1/sr3usr_1/sr3usr.data1 PSAPSR3USR 11968

Now i want to write a script which need to take the first tablespace name from log 1 and then grep that tablespace in log 2 and search if any of the datafile is having less than 6000 mb then "echo .." And then take second tablespace name from log 1 and repeat the same...

Any help is appreciated..

2 
grep -f datafile_test.log freedata_test.log| awk '$3 < 6000 {print $1}'
3

Assuming your 3rd column is in MB.

awk '{ if (FNR==NR) { a[$2]+=$3; next } if (a[$1] < 6000) { print $1 } }' freedata_test.log datafile_test.log

4

Use nawk or /usr/xpg4/bin/awk on Solaris.
I would do it all in sql though ...

awk 'NR == FNR { tbs[$1]; next }
$2 in tbs && $NF < 6000 {
  printf "tablespace %s, datafile %s has %d MB free\n",
    $2, $1, $NF
  }' datafile_test.log freedata_test.log
5

The OP said:

But, of course, your idea makes more sense.

6

Thanks for the correction

7

perl:

@arr=('PSAPSR3','PSAPSR3700','PSAPSR3USR');
open FH,"<a.txt";
while(<FH>){
	@_= split(" ",$_);
	map {tr/\n//d} @_;
	$_{$_[1]}.=$_ if ($_[2]<6000);
}
close FH;
for($i=0;$i<=$#arr;$i++){
	print $_{$arr[$i]},"\n";
}