I use badTEST for that where necessary. Depending on what MTA you're running, you may be able to boost the refusal level from defaults without compromising server operation. With sendmail, refuseLA defaults to a load average of 12*number-of-processors, and most relatively modern Linux servers will continue operating reasonably at 50 or higher if they're dedicated MTAs. I used to have that level frequently on Dell Poweredge 2550's with dual 1Ghz processors and 4Gb RAM running RHEL 3. In a spam storm, the undeliverable bounces could get out of hand, but that's a problem that should get attention.

On Mon, January 26, 2009 16:18, Olivier Beau wrote:

Hi,

I would use the badTEST feature (cf. man bb-hosts)

Olivier

On 26/01/2009 22:08, Martha McConaghy wrote:

...

I'm monitoring some very busy SMTP servers that handle all our incoming email. Because they are so busy, at times, Xymon will get a connection refused condition when testing SMTP, which generates a red alert. Its fine at the next test cycle, but the false alerts concern me. (People get used to seeing false alerts, they ignore the real ones.)

Does anyone have any suggestions? Is there I way I can alter the SMTP test to not generate this condition as an alert, or perhaps an alert script that won't send out an email?

Martha McConaghy Marist College

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