Search Results

My guess is you are executing "wrapper.sh" instead sourcing.


# try
. ./wrapper.sh

#instead
./wrapper.sh

Hello Kamal,

Following may help you in same.


awk 'BEGIN{split("ABCDEFGHIJKLMNOPQRSTUVWXYZ", A, "")} {print $0 ":" A[++i]} END{ORS="";print "\n"}' ORS=" " Input_file


Output will be as...

Try read -a H, and then read the man bash

Try something like:

#!/bin/bash
nodecount=10
host=na7-db1-1-chi
IFS=- read -a H <<< "$host"
C=({A..Z})
for ((i=1; i<=nodecount; i++))
do
printf "%s "...

If want to avoid the some of the pipes,

ps -ef | awk '/pmon/ && !/asm/ && !/grep/ {x=$1;gsub(/.*_/,_);print x FS $0}'

I guess, the process owner would be always oracle? as its an oracle...

Excellent idea Akshay. Note, however, that some versions of awk sometimes (depending on field widths) use a tab as well as spaces as field delimiters. To be safe you might want to use:
awk -F'[...

Try :

$ ps -ef | grep pmon | grep -v grep | grep -v asm | awk -F'[ _]' '{print $1,$NF}'
$ ps -ef | grep pmon | grep -v grep | grep -v asm | awk '{s=$1;sub(/.*_/,x);print s,$1}'

Try:
ps -ef | grep pmon | grep -v grep | grep -v asm | awk '
{ n = split($NF, u, /_/)
print $1, u[n]
}'

If you change:
for(j = i+1; j <= i+4; j++)in SriniShoo's awk script to:
for(j = i+1; j <= i+3; j++)
it should do what you want on your system. It uses one construct on all of the print...

awk -v q="'" 'BEGIN{FS="\n"; RS="";
HEAD = "set echo on timing on feedback on \
\nspool alter_diskgroup.add_datadg_disks_1.log \
\nalter diskgroup DATADG add disk";
TAIL = ";\nspool off \...

Something like this..?
michaelf>uname
SunOS
michaelf>nawk -F'[: ]' '/Parameter Name/{f=$5;getline;print f"-->"$5}' infile
SNAPSHOTTIMESTAMP-->2014-01-07-15.21.50.022423
DATABASESIZE-->96178176...

yet another awk:
awk -F'[ \t\n.]' '/ORATEXT/{print $10 (++c[$10]), $5}' RS= OFS=" -> " file

try this..


awk -F "[. ]" '{if($0 ~/ORATEXT : cluster_interconnect/){a++;print $3a,s}else if($0 ~/ORATEXT : public/){b++;print $3b,s}else{s=$5}}' OFS=" -> " file

Like this?
awk '/^ORATEXT : (cluster_interconnect|public)/{
sub(/([^.]+[.]){4}/,"",p)
sub(/[.].*/,"",p)
print $3,p}{p=$0}' OFS=' -> ' file
BTW, I like the looks of this regexp... :D

Are you sure the install directives are correct? Maybe change to something like the following to test installation:

%install
make DESTDIR=%buildroot install

To protect it from changes in the order the adapters occur, try this version:
awk 'BEGIN {RS="/"; FS="[= \"]+"}
/cluster_interconnect/ {sub("cluster_interconnect","Private"++ct)}
...

This will do the job for your example:
awk 'BEGIN {RS="/"; FS="[= \"]+"}
/Adapter/ {
for (i=NF;i>=1;i--) if ($i=="Adapter") break;
...

echo $ip | perl -pe 's/ /\n/g'

echo $ip | sed 's/ /\n/g'

echo $ips|xargs -n1

echo $ips | tr -s " " "\n"

If the value of the variable already has embedded newlines, then all you have to do is quote the variable. If not, you can use the tr utility to convert the embedded whitespace to newlines.
...

How about this using bash:

#!/bin/bash
IFS=' .v' V=( $(perl --version) )
[ ${V[3]:-0} -le 5 ] && [ ${V[3]:-0} -ne 5 -o ${V[4]} -lt 8 ] && exit 1
echo "Perl version ${V[3]}.${V[4]}.${V[5]} is...