Solaris command: prstat

Total Physical Memory

2. >prtdiag -v | grep Memory
3. Memory size: 8192 Megabytes
5. >prtconf | grep Memory
6. Memory size: 8192 Megabytes

Total Physical Memory

2. >vmstat 5 10
3. r b w swap free re mf pi po fr de sr vc vc vc vc in sy cs us sy id
4. 0 0 0 14841224 772880 23 147 19 6 6 0 1 1 0 2 2 3775 2990 3820 2 2 96
5. 1 0 0 20482808 4992304 3 3 0 0 0 0 0 1 0 0 0 527 474 482 13 0 87
6. 0 0 0 20483192 5000272 1 5 0 3 3 0 0 0 0 0 1 583 645 606 13 1 86
7. 0 0 0 20485848 5033400 0 0 0 0 0 0 0 0 0 0 0 528 463 492 13 1 86
8. 0 0 0 20485848 5033344 0 0 0 0 0 0 0 0 0 0 1 514 469 475 13 0 87
9. 0 0 0 20485848 5033288 0 0 0 0 0 0 0 0 0 0 2 513 450 476 13 0 87
10. 0 0 0 20485848 5033288 0 0 0 0 0 0 0 1 0 0 0 516 468 476 13 0 87
11. 0 0 0 20485848 5033240 0 0 0 0 0 0 0 0 0 0 0 510 451 467 13 0 87
12. 0 0 0 20485848 5033184 0 0 0 0 0 0 0 0 0 0 0 518 468 475 13 0 87
13. 0 0 0 20485848 5033128 0 0 0 0 0 0 0 0 0 0 0 523 456 474 13 1 86

Find Overall CPU Usages

* Use multiple processor stat tool ‘mpstat’
* Columns to note
– ‘usr’: percent user time
– ‘sys’: percent system time
– ‘idl’: percent idle time
* Example

2. # Generate 2 processor reports with 5 second interval
3. mpstat 5 2
4. CPU minf mjf xcal intr ithr csw icsw migr smtx srw syscl usr sys wt idl
5. 0 14 0 42 35 18 48 0 1 8 0 139 1 0 0 99
6. 1 7 0 27 87 55 65 0 1 8 0 92 0 1 0 99

Find CPU Usage by Processes

* Use process stat tool ‘prstat’
* Similar to Linux ‘top’ command
* Examples

2. # Sort by cpu usage (default)
3. prstat
5. # Sort by memory usage
6. prstat -s rss
8. # Sort by virtual memory usage
9. prstat -s size
11. # Sort by process execution time
12. prstat -s time
14. # Find top 5 processes
15. prstat -n 5
17. # Also print process numbers each user is using
18. prstat -a
20. # Follow a particular process id
21. prstat -p pid
23. # Follow threads of a particular process id
24. prstat -L -p pid

Find Paging Activities

* Use system activity report tool ‘sar’
* Examples

2. # Find top 3 paging activities with 5 second interval
3. sar -g 5 3
4. 18:57:12 pgout/s ppgout/s pgfree/s pgscan/s %ufs_ipf
5. 18:57:17 0.00 0.00 0.00 0.00 0.00
6. 18:57:22 0.00 0.00 0.00 0.00 0.00
7. 18:57:27 0.00 0.00 0.00 0.00 0.00
9. Average 0.00 0.00 0.00 0.00 0.00
11. # Total memory
12. /usr/sbin/prtconf | grep Memory | awk -F”:” ‘{print $2}’
13. Memory size: 8192 Megabytes
15. # Find free memory and swap space based on 3 reads with 5 second read intervals
16. sar -r 5 3
17. 18:56:24 freemem freeswap
18. 18:56:29 175049 28289968
19. 18:56:34 175031 28289843
20. 18:56:39 175026 28289952
22. Average 175036 28289921

Process Memory Usage

2. >ps -A -o pmem -o pcpu -o args
4. >ps -e -o “rss(k) vsz(k) pmem pid comm” | qrep -v COMMAND | sort -rn | awk ‘{print $1 “\t” $2 “\t” $3 “\t” $4 “\t” $5;}’

Sample Sessions

* Follow a particular process id in 5 second intervals and send result to server.out file

2. bash-3.00$ prstat -L -p 16401 5 > server.out &
3. [1] 16530
4. bash-3.00$ tail -f server.out
5. PID USERNAME SIZE RSS STATE PRI NICE TIME CPU PROCESS/LWPID
6. 16401 oracle 5270M 4125M sleep 59 0 0:00:00 0.0% oracle/1
7. Total: 1 processes, 1 lwps, load averages: 0.03, 0.04, 0.04
8. PID USERNAME SIZE RSS STATE PRI NICE TIME CPU PROCESS/LWPID
9. 16401 oracle 5270M 4125M sleep 59 0 0:00:00 0.0% oracle/1
10. Total: 1 processes, 1 lwps, load averages: 0.03, 0.04, 0.04
11. PID USERNAME SIZE RSS STATE PRI NICE TIME CPU PROCESS/LWPID
12. 16401 oracle 5270M 4125M sleep 59 0 0:00:00 0.0% oracle/1
13. Total: 1 processes, 1 lwps, load averages: 0.03, 0.04, 0.04
14. PID USERNAME SIZE RSS STATE PRI NICE TIME CPU PROCESS/LWPID
15. 16401 oracle 5270M 4125M sleep 59 0 0:00:00 0.0% oracle/1
16. Total: 1 processes, 1 lwps, load averages: 0.03, 0.04, 0.04
17. ^C
18. bash-3.00$ jobs
19. [1]+ Running prstat -L -p 16401 5 >server.out &
20. bash-3.00$ fg %1
21. prstat -L -p 16401 5 >server.out
22. ^C
23. bash-3.00$