I think it would be nice to have the green/yellow/red based on the
GB/MB/KB size since some of our disks are presented from a SAN and can
be close to 180GB, even 1% of that is HUGE...
I was thinking about this just the other day, so thanks to Anatoli for
bringing it up :)
Adam
-----Original Message-----
From: Henrik Stoerner [mailto:user-ce4a2c883f75@xymon.invalid] Sent: Thursday, October 06, 2005 9:24 AM
To: user-ae9b8668bcde@xymon.invalid
Subject: Re: [hobbit] mailq script & absolute values for disk checks
On Thu, Oct 06, 2005 at 03:06:36PM +0200,
user-4d3800b5a33a@xymon.invalid wrote:
i am trying to implement the mailq check feature, but it seems that i
am
missing the right mailq script. If i use mailq.sh script from deadcat
the
values for mail queue size and length are reported properly, but no
RRD
file is created and thus no graph appears. Log files remains clean.
Within do_mailq.c there is a notice about "Erik's enhanced mailq
script"
Unfortunately i cant find this script. Maybe anybody is using this one
and
could provide it?
Heh :-) the "Erik" mentioned here is a colleague of mine who handles a bunch of Sendmail servers. That's a custom script for a virus-scanning
mail-gateway we have, where we want to check on both incoming and
outgoing mails - the default mailq scripts I've seen only looks at the
outbound queue size. So Erik's enhanced mailq script is pretty specific
to our setup, i think.
The normal mailq script I used for this was the one that comes with larrd 0.43c (the "bf/nmailq-bf.sh" script in LARRD).
Would you mind to think about the option to enhance disk checks for handling of absolute disk usage (KB/MB/GB) and not only the percent values?
It's already there - just ask for the "disk1" graph instead of "disk".
Or were you thinking of having the status (red/yellow/green) depend on the # of GB's used ?
Henrik