Sponsored Content
Full Discussion: Performance Monitoring
Operating Systems Solaris Performance Monitoring Post 302264027 by jlliagre on Wednesday 3rd of December 2008 03:38:05 AM
Old 12-03-2008
You can add "mpstat", "lockstat", "fsstat", "zpool iostat" and "kstat" to the list. The latter displays most of raw counters used by the remaining commands.
 

10 More Discussions You Might Find Interesting

1. UNIX for Advanced & Expert Users

performance monitoring

hi, can any one tell me, is there is any way i can check the performance of my solaris 8 os on an Ent 3500. Other than top to check for the top most processes, how to make the calculations with vmstat, iostat, mpstat and nfsstat. Or is there any other tools that i can use? cheers. (3 Replies)
Discussion started by: i2admin
3 Replies

2. AIX

Performance monitoring

Hi All I am looking for a script that would collect statistics in a summarised format. CPU, Memory,Swap, Wait queue, Run queue and disk activity. Something that would allow me to profile the environment based on a 1 line output that I could run every 15 min. Thx Junaid (1 Reply)
Discussion started by: jhansrod
1 Replies

3. UNIX for Advanced & Expert Users

Performance Monitoring

Hi all The place I work for is about to to place there database server under heavy load for testing and would like the effect recorded as much as possible. Can anyone point me in the right direction with respect to real time system monitoring. I am aware of of 'sar', vmstat etc and hope to... (2 Replies)
Discussion started by: silvaman
2 Replies

4. UNIX for Dummies Questions & Answers

Performance monitoring

Hello, I am trying to find a way to view current CPU and disk usage. I used to use nmon which worked fine but since an upgrade to our servers this is no longer available. I have tried to get it reinstalled to no avail! Are there any other commands you can use within unix which will allow me... (4 Replies)
Discussion started by: johnwilliams
4 Replies

5. UNIX for Dummies Questions & Answers

Unix Performance Monitoring

In the vmstat , there are many columns you can see. Can someone tell me what is the most important column that i need to be watched on, and what value or average value should i watch inorder to determine that im experiencing a cpu bottle neck. What should be my basis. or if you use glance ... (2 Replies)
Discussion started by: kaibiganmi
2 Replies

6. Shell Programming and Scripting

Performance monitoring help needed.

How would i check for following? 1)open ports in my linux machine. 2)Hard disk read speed. 3)Hard disk write speed. (2 Replies)
Discussion started by: pinga123
2 Replies

7. AIX

Performance Monitoring of FileSystem

As I am new to the Unix field, I would like to get the clarification regarding the Filesystem. The scenario is.. The filesystem (/drbackup) is getting monitored and if it exceeds the threshold, we will receive an alert from it. The issue is that we receive an alert with the description of... (2 Replies)
Discussion started by: A.Srenivasan
2 Replies

8. Solaris

Performance / Batch monitoring

What tools can I use to look "deeper" into a process to see if the job is actually running or just hanging. What is the best method to accomplish this? SunOS 5.10 Generic_142900-14 sun4v sparc SUNW,T5240 (2 Replies)
Discussion started by: Harleyrci
2 Replies

9. AIX

Need some help for AIX performance monitoring

Hello I am new user of AIX; I have only basic knowledge of the UNIX commands, and I want to create script that will monitor the performance and resources usage on AIX 6.1 machine. Basically I wan to start a loop that will grab, every 10 seconds, the CPU usage, the memory usage, the disk usage,... (1 Reply)
Discussion started by: adaher
1 Replies

10. Red Hat

Performance Monitoring - RHEL 7.4

Have a question about doing some performance monitoring - how to approach it. This is on RHEL 7.4 We are moving a rather large application to new hardware in the future. I would like to find a way to compared performance on the new hardware at different stages and at different times. I'm... (4 Replies)
Discussion started by: Overcast451
4 Replies
LOCKSTAT(8)						    BSD System Manager's Manual 					       LOCKSTAT(8)

NAME
lockstat -- display kernel locking statistics SYNOPSIS
lockstat [-ceflMmpstx] [-b nbuf] [-E event] [-F func] [-L lock] [-N nlist] [-o file] [-T type] command ... DESCRIPTION
The lockstat command enables system wide tracing of kernel lock events, executes the specified command, and when finished reports statistics to the user. Tracing may be ended early by sending SIGINT (Ctrl-C) to the process being executed by lockstat. The lockstat pseudo-device driver must be present in the kernel, and the lockstat command may only be used by the root user. The options are as follows: -b nbuf Adjust the number of trace buffers allocated by the kernel to nbuf. -c Report percentage of total events by count, and sort the output by number of events. The default is to key on event timings. -E event Limit tracing to one type of event. Use the -e option to list valid events. -e List valid event types for the -E option and exit. -F func Limit tracing to locking operations performed within the specified function. func must be the name of a valid function in the kernel. -f Trace only by calling functions; do not report on individual locks. -L lock Limit tracing to one lock. lock may either be the name of a lock object in the kernel, or a kernel virtual address. -l Trace only by lock; do not report on calling functions. -M Merge lock addresses within unique objects. -m Merge call sites within unique functions. -N nlist Extract symbol information from the nlist file. -o file Send output to the file named by file, instead of the standard output (the default). -p Show the average number of events and time spent per CPU. The default is to show the total values. May be used in conjunction with the -s option. -s Show the average number of events per second, and the average time spent per second. The default is to show the total values. -T type Limit tracing to one type of lock. Use the -t option to list valid lock types. -t List valid lock types for the -T option and exit. -x Summarize events, and do not report on lock types. FILES
/dev/lockstat lockstat control device /dev/ksyms namelist EXAMPLES
# lockstat -L uvm_pageqlock sleep 10 Elapsed time: 10.01 seconds. -- Adaptive mutex spin Total% Count Time/ms Lock Caller ------ ------- --------- ---------------------- ------------------------------ 100.00 1281 0.78 uvm_pageqlock <all> 39.81 385 0.31 uvm_pageqlock uvm_fault_internal+11cc 30.98 358 0.24 uvm_pageqlock uvm_fault_internal+bb1 28.06 522 0.22 uvm_pageqlock uvm_anfree+132 0.51 5 0.00 uvm_pageqlock ubc_fault+28f 0.20 4 0.00 uvm_pageqlock uvm_fault_internal+12b6 0.18 2 0.00 uvm_pageqlock uao_detach_locked+58 0.11 2 0.00 uvm_pageqlock uvm_fault_internal+7d5 0.08 2 0.00 uvm_pageqlock ufs_balloc_range+160 0.07 1 0.00 uvm_pageqlock uvm_fault_internal+107b DIAGNOSTICS
lockstat: incompatible lockstat interface version The kernel device driver does not match the version of the lockstat command. lockstat: overflowed available kernel trace buffers Increase the number of buffers using the -b option. lockstat: ioctl: Invalid argument The number of trace buffers is outside the minimum and maximum bounds set by the kernel. SEE ALSO
ps(1), systat(1), vmstat(1), iostat(8), pstat(8) HISTORY
The lockstat command appeared in NetBSD 4.0. BSD
March 21, 2009 BSD
All times are GMT -4. The time now is 05:49 PM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy