[hobbit] Setting up Disk usage Alarm level

Eric Meddaugh etmsys at rit.edu
Fri Sep 21 13:28:37 CEST 2007


One other item I did touch on would be the purpose of the disk.  For
example disk based backups you might want at higher levels just to catch
something before the night when all backups normally occur.  This does
go along the same lines as 1 & 2, but adds a little different thought
into planning the levels.

---Eric

-----Original Message-----
From: Ryan Jay B. Lapuz [mailto:rlapuz at fcpp.fujitsu.com] 
Sent: Thursday, September 20, 2007 20:48
To: hobbit at hswn.dk
Subject: Re: [hobbit] Setting up Disk usage Alarm level

Thanks Eric,

Base on what you said, I added some points to consider and here it is:

Guidelines in setting up disk monitoring alarm levels:

1. How fast the disk fills up?
        a. Temporary disk usage
        b. Regular disk usage increase (Monthly, Weekly)
2. How fast you can respond whena warning/alarm nitification goes out?
        a. During the alarm, make sure that the remaining disk space can

maintain system availability while addressing the problem.
        b. Identify what files to move/delete - this is to speed up the 
recovery

Can you please further eloborate if I miss other factors cause I think I

still do.
Maybe other menbers can also add their ideas so we can make a more
effective 
guidlines for disk monitoring.
I think this is very helpful specially to those that do not have a lot
of 
experience like me.

Thanks and regards,
Ryan


----- Original Message ----- 
From: "Eric Meddaugh" <etmsys at rit.edu>
To: <hobbit at hswn.dk>
Sent: Thursday, September 20, 2007 7:50 PM
Subject: RE: [hobbit] Setting up Disk usage Alarm level



I probably depends on how fast the disk fills up and how quickly you can
respond to clean it.

Our "default" levels are 500Mg warning, and 100Mg red.  On some other
systems that have web servers that we want to be alarmed sooner, we have
the warning at 1g, while the alarm is 500Mg.

We've been looking at changing some of the Oracle servers we have to
higher than 500/100 since when someone is processing a giant load the
archive logs can fill the disk up way to fast.

On a system that uses disk based backups, the level are even much
higher, around 100g warning, and 70g alarm.

It'll all depend on the individual system and how fast you can respond
when a warning/alarm notification goes out.

---Eric

-----Original Message-----
From: Ryan Jay B. Lapuz [mailto:rlapuz at fcpp.fujitsu.com]
Sent: Thursday, September 20, 2007 07:40
To: hobbit at hswn.dk
Subject: [hobbit] Setting up Disk usage Alarm level

Good day!

Guys, I need your help again.
Is there someone here who have guidelines for setting up the alarm level
for
disk usage?
It's just that I can't defend it if I'm just going to set something
without
reference.

Please kindly share it to me. I just need it so badly.

Thank you very much!

Ryan



To unsubscribe from the hobbit list, send an e-mail to
hobbit-unsubscribe at hswn.dk



To unsubscribe from the hobbit list, send an e-mail to
hobbit-unsubscribe at hswn.dk





To unsubscribe from the hobbit list, send an e-mail to
hobbit-unsubscribe at hswn.dk





More information about the Xymon mailing list