When I run top -n 1 -d 2 | head -n 12; it returns processor usage for some processes sorted by %cpu desc as desired, but I'm not convinced that the results are aggregated as they should be. I'm wanting to put these results in a file maybe like
while true; do
top -n 1 -d 2 | head -n 12;
done > top_cpu_users;
When I run top -d 2; interactively, I first see some results, then two seconds later I see the results updated and they appear to be aggregated over the last two seconds. The first results do not appear to be aggregated in the same way.
How do I get top cpu users every two seconds aggregated over the previous two seconds?
top will always capture a first full scan of process info for use as a baseline. It uses that to initialize the utility's database of values used for later comparative reporting. That is the basis of the first report presented to the screen.
The follow-on reports are the true measures for the specified evaluation intervals.
Your code snippet will therefore never provide what you are really looking for.
You need to skip the results from the first scan and only use the follow on reports, but the only way to do that is to generate them from a single command by specifying the count of scans desired, then parse the resulting combined report.
To that end, here is a proposed solution:
#!/bin/bash
output="top_cpu_users"
rm -f ${output} ${output}.tmp
snapshots=5
interval=2
process_count=6 ### Number of heavy hitter processes being monitored
top_head=7 ### Number of header lines in top report
lines=$(( ${process_count} + ${top_head} )) ### total lines saved from each report run
echo -e "\n Collecting process snapshots every ${interval} seconds ..."
top -b -n $(( ${snapshots} + 1 )) -d ${interval} > ${output}.tmp
echo -e "\n Parsing snapshots ..."
awk -v max="${lines}" 'BEGIN{
doprint=0 ;
first=1 ;
}
{
if( $1 == "top" ){
if( first == 1 ){
first=0 ;
}else{
print NR | "cat >&2" ;
print "" ;
doprint=1 ;
entry=0 ;
} ;
} ;
if( doprint == 1 ){
entry++ ;
print $0 ;
if( entry == max ){
doprint=0 ;
} ;
} ;
}' ${output}.tmp >${output}
more ${output}
The session output for that will look like this:
Collecting process snapshots every 2 seconds ...
Parsing snapshots ...
266
531
796
1061
1326
top - 20:14:02 up 8:37, 1 user, load average: 0.15, 0.13, 0.15
Tasks: 257 total, 1 running, 256 sleeping, 0 stopped, 0 zombie
%Cpu(s): 0.5 us, 1.0 sy, 0.0 ni, 98.5 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st
MiB Mem : 3678.9 total, 157.6 free, 2753.7 used, 767.6 buff/cache
MiB Swap: 2048.0 total, 1116.4 free, 931.6 used. 629.2 avail Mem
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
31773 root 20 0 0 0 0 I 1.5 0.0 0:09.08 kworker/0:3-events
32254 ericthe+ 20 0 14500 3876 3092 R 1.0 0.1 0:00.04 top
1503 mysql 20 0 2387360 20664 2988 S 0.5 0.5 3:10.11 mysqld
2250 ericthe+ 20 0 1949412 130004 20272 S 0.5 3.5 0:46.16 caja
3104 ericthe+ 20 0 4837044 461944 127416 S 0.5 12.3 81:26.50 firefox
29998 ericthe+ 20 0 2636764 165632 54700 S 0.5 4.4 0:36.97 Isolated Web Co
top - 20:14:04 up 8:37, 1 user, load average: 0.14, 0.13, 0.15
Tasks: 257 total, 1 running, 256 sleeping, 0 stopped, 0 zombie
%Cpu(s): 1.5 us, 0.7 sy, 0.0 ni, 97.4 id, 0.4 wa, 0.0 hi, 0.0 si, 0.0 st
MiB Mem : 3678.9 total, 157.5 free, 2753.7 used, 767.6 buff/cache
MiB Swap: 2048.0 total, 1116.4 free, 931.6 used. 629.2 avail Mem
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
3104 ericthe+ 20 0 4837044 462208 127416 S 3.0 12.3 81:26.56 firefox
1503 mysql 20 0 2387360 20664 2988 S 1.0 0.5 3:10.13 mysqld
32254 ericthe+ 20 0 14500 3876 3092 R 1.0 0.1 0:00.06 top
1489 root 20 0 546692 61584 48956 S 0.5 1.6 17:23.78 Xorg
2233 ericthe+ 20 0 303744 11036 7500 S 0.5 0.3 4:46.84 compton
7239 ericthe+ 20 0 2617520 127452 44768 S 0.5 3.4 1:41.13 Isolated Web Co
top - 20:14:06 up 8:37, 1 user, load average: 0.14, 0.13, 0.15
Tasks: 257 total, 1 running, 256 sleeping, 0 stopped, 0 zombie
%Cpu(s): 0.6 us, 0.4 sy, 0.0 ni, 99.0 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st
MiB Mem : 3678.9 total, 157.5 free, 2753.7 used, 767.6 buff/cache
MiB Swap: 2048.0 total, 1116.4 free, 931.6 used. 629.2 avail Mem
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
1489 root 20 0 546700 61584 48956 S 1.5 1.6 17:23.81 Xorg
3104 ericthe+ 20 0 4837044 462208 127416 S 1.5 12.3 81:26.59 firefox
1503 mysql 20 0 2387360 20664 2988 S 0.5 0.5 3:10.14 mysqld
2233 ericthe+ 20 0 303744 11036 7500 S 0.5 0.3 4:46.85 compton
2478 ericthe+ 20 0 346156 10368 8792 S 0.5 0.3 0:22.97 mate-cpufreq-ap
2481 ericthe+ 20 0 346540 11148 9168 S 0.5 0.3 0:41.73 mate-sensors-ap
top - 20:14:08 up 8:37, 1 user, load average: 0.14, 0.13, 0.15
Tasks: 257 total, 1 running, 256 sleeping, 0 stopped, 0 zombie
%Cpu(s): 0.6 us, 0.5 sy, 0.0 ni, 98.9 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st
MiB Mem : 3678.9 total, 157.5 free, 2753.6 used, 767.7 buff/cache
MiB Swap: 2048.0 total, 1116.4 free, 931.6 used. 629.3 avail Mem
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
32254 ericthe+ 20 0 14500 3876 3092 R 1.0 0.1 0:00.08 top
3104 ericthe+ 20 0 4837044 462208 127416 S 0.5 12.3 81:26.60 firefox
18370 ericthe+ 20 0 2682392 97268 45144 S 0.5 2.6 0:55.36 Isolated Web Co
19436 ericthe+ 20 0 2618496 123608 52540 S 0.5 3.3 1:55.08 Isolated Web Co
26630 ericthe+ 20 0 2690464 179020 56060 S 0.5 4.8 1:45.57 Isolated Web Co
29998 ericthe+ 20 0 2636764 165632 54700 S 0.5 4.4 0:36.98 Isolated Web Co
top - 20:14:10 up 8:37, 1 user, load average: 0.13, 0.13, 0.15
Tasks: 257 total, 1 running, 256 sleeping, 0 stopped, 0 zombie
%Cpu(s): 2.5 us, 0.9 sy, 0.0 ni, 96.6 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st
MiB Mem : 3678.9 total, 157.5 free, 2753.6 used, 767.7 buff/cache
MiB Swap: 2048.0 total, 1116.4 free, 931.6 used. 629.3 avail Mem
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
3104 ericthe+ 20 0 4837076 463000 127416 S 7.5 12.3 81:26.75 firefox
1489 root 20 0 546716 61584 48956 S 1.5 1.6 17:23.84 Xorg
1503 mysql 20 0 2387360 20664 2988 S 1.0 0.5 3:10.16 mysqld
32254 ericthe+ 20 0 14500 3876 3092 R 1.0 0.1 0:00.10 top
2233 ericthe+ 20 0 303744 11036 7500 S 0.5 0.3 4:46.86 compton
2481 ericthe+ 20 0 346540 11148 9168 S 0.5 0.3 0:41.74 mate-sensors-ap