Intake file
website1 ip 20 website1 ip 30 website1 ip 10 website2 ip 200 website2 ip 10 website3 ip 150 I need it to sort so that the sum of the 3rd with the associated website will show first with the highest total with the website
website2 has 210 website 3 has 150 website1 has 60
website2 ip 200 website2 ip 10 website3 ip 150 website1 ip 30 website1 ip 20 website1 ip 10 ive tried sort -k3n -k1n but that is not the results i am looking for 230 website2 ip 200
You can use awk to have 2 pass and sum the column 3 per website (column 1) and append a new column in the output. Then sort using new column in output and finally strip first column from output using cut:
awk 'FNR==NR{sum[$1]+=$3; next} {print sum[$1] "\t" $0}' file file | sort -k1nr -k4nr | cut -f2- website2 ip 200 website2 ip 10 website3 ip 150 website1 ip 30 website1 ip 20 website1 ip 10 Output of awk command:
awk 'FNR==NR{sum[$1]+=$3; next} {print sum[$1] "\t" $0}' file file 60 website1 ip 20 60 website1 ip 30 60 website1 ip 10 210 website2 ip 200 210 website2 ip 10 150 website3 ip 150 
awk 'FNR==NR{sum[$1]+=$3; next} {print sum[$1] "\t" $0}' file file | sort -k1nr -k4nr | cut -f2-
sortsolution.sortcompares on a line-by-line basis but your sorting criteria depends on sorting in groups, then lines.