HPUNIX Monitoring tool

[sendilkumar]

CAn any one suggest Opensource Monitoring tool for HPUNIX machine. We tried with Glance.Mainly to check the performance of machine when load testing is done.

[Peter Godron]

Hi,
and welcome to the forums !

If you really did not like glance:
http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=1019371
http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=92369

Please also read:
http://forums1.itrc.hp.com/service/forums/helptips.do?#33 on how to reward any useful answers given to your questions.

[Asif Sharif]

Hi Sendilkumar,

Glanceplus is good tool for performacne.
http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=1019371

Regards,
Asif Sharif

[Steve Steel]

Hi

Indeed glance rocks and is the best solution.


Otherwise look on

http://hpux.connect.org.uk/

http://docs.hp.com/en/5991-7469/index.html


steve Steel

[spex]

Hi,

The sar (system activity reporter) package works well for historical data, and comes pre-installed under HP-UX. For more information, check out sar(1M) and sa1(1M).

PCS

[Dave Wherry]

You said "when load testing is done."
Glance does not have historical data so it is only usefull during the load test.
As was mentioned, sar is included in the OS. Read up on it and you can script it to monitor the metrics you are interesed in while testing and redirect the output to a file for later study.

[Geoff Wild]

For "monitoring" have a look at BigBrother:

http://bb4.org/

If you are talking performance, then use built in tools like sar and vmstat.

Rgds...Geoff

[Nate Schuchard]

Hi,

I used SAM > Resource Management > Event Monitoring Service and set up notifications for limmits that I did not want to excceed.
You may want to do this in GUI mode as opposed to test mode. it is easier to set up the monitoring. It was very handy when we were bumping up on some resources.

Regards
Natalie

[sendilkumar]

Glance doesn't store the data i think,so that we can analysis later point of time. I need something to monitor and collect the data while testing and redirect the output to a file for later study.

[Berd]

Perfview will allow you to perform historical analysis. You forward your glance data to Perfview on a daily cycle and it builds up a historical view.

Berd

[Geoff Wild]

If you have measure ware, then you can use extract:

In root cron, put something like:

55 23 * * 1-5 /opt/perf/bin/extract -xp -r /home/gwild/zgbltemplate -g -b today 7:00 -e today 18:00 -f stdout | /bin/mailx -s 'system performance report' you@yourdomain.com >/tmp/performance.log 2>&1


zgbltemplate is where you specify parameters you want to report on:

# cat /home/gwild/zgbltemplate
REPORT "MWA Export on !SYSTEM_ID"
FORMAT ASCII
HEADINGS ON
SEPARATOR="|"
SUMMARY=60
MISSING=0
DATA TYPE GLOBAL
YEAR
DATE
TIME
GBL_CPU_TOTAL_UTIL
GBL_CPU_SYS_MODE_UTIL
GBL_CPU_USER_MODE_UTIL
GBL_CPU_SYSCALL_UTIL
GBL_CPU_INTERRUPT_UTIL
GBL_PRI_QUEUE
GBL_CPU_CSWITCH_UTIL
GBL_SWAP_SPACE_UTIL
GBL_DISK_UTIL_PEAK
GBL_DISK_SUBSYSTEM_QUEUE
GBL_MEM_UTIL
GBL_MEM_CACHE_HIT_PCT
GBL_MEM_PAGEOUT_RATE
GBL_MEM_SWAPOUT_RATE
GBL_MEM_QUEUE
GBL_NET_PACKET_RATE
GBL_NET_OUTQUEUE
GBL_NETWORK_SUBSYSTEM_QUEUE


You can also just run extract from the command line:


What was my server doing on Oct 23 from midnight to 13:00:

/opt/perf/bin/extract -xp -r /home/gwild/zgbltemplate -g -b 10/23/06 0:00 -e 10/23/06 13:00 -f stdout | /bin/mailx -s 'system performance report' me@mydomain.com

Rgds...Geoff

[Coolmar]

I have monitored systems on many sites using Big Brother. The main screen will list all your servers and have green, amber, and red lights for server status. There are many many modules also to monitor apps like oracle on your systems as well. I find it excellent for a free tool.

http://www.bb4.org/

[Raj D.]

Hi Sedilkumar,

You can check this :

http://www.sarcheck.com/schp.htm

But its not free. ($$.)

Cheers,
Raj.

[sendilkumar]

Thanks for your feedback. I planned to work obn glance. i'm getting error as belo. I would appreciate your valuable response.
MeasureWare scope status:
WARNING: scopeux is not active (MWA data collector)

MeasureWare background daemon status:
(Should always be running when the system is up)
Running ttd (Transaction Tracker daemon) pid 29794 5111

MeasureWare server status:
Running alarmgen (alarm generator) pid 5602
Running agdbserver (alarm database server) pid 5601
Running perflbd (location broker) pid 5124

The following data sources have running repository servers:
PID DATA SOURCE
Running rep_server 5596 SCOPE

Status .scope details

**** /opt/perf/bin/scopeux : 12/07/06 17:57:39 ****
scopeux C.03.72.00 COLLECTOR BEGIN.

**** /opt/perf/bin/scopeux : 12/07/06 18:02:49 ****
scopeux C.03.72.00 COLLECTOR BEGIN.

**** scopeux : 12/07/06 18:06:51 ****
scopeux C.03.72.00 COLLECTOR BEGIN.

**** scopeux : 12/07/06 18:06:51 ****
ERROR: RUN file exists, collector may already be running. (PE221-22)

**** scopeux : 12/07/06 18:06:51 ****
COLLECTOR END. program terminated abnormally.