Kevin McGregor wrote:

Hey all,

Not actually UNIX related, but since a lot of you are network admins you might have some suggestions.

I'd like to keep an eye on a bunch of servers (~20) but all I need to know is which ones are up or down, and if they're down, when they were last "seen". Does anyone know of a simple, free tool that will do that? I don't want details like number of packets lost or any such. Mostly I want to know which machines shut down and in what order.

Thanks in advance, Kevin

Thought I saw something like this on Freshmeat within the last couple of days. It was intended for rack units but it may suffice for your set up. I can't recall the name off the top of my head. It may be something like Rackit...

Later Mike

Kevin McGregor wrote:

Hey all,

Here is the actual link to the project I mentioned. At least I was close with the name...

Later Mike

http://freshmeat.net/projects/racktables

On Thu, 2009-17-09 at 13:27 -0500, Kevin McGregor wrote:

Hey all,

Not actually UNIX related, but since a lot of you are network admins you might have some suggestions.

I'd like to keep an eye on a bunch of servers (~20) but all I need to know is which ones are up or down, and if they're down, when they were last "seen". Does anyone know of a simple, free tool that will do that? I don't want details like number of packets lost or any such. Mostly I want to know which machines shut down and in what order.

Thanks in advance, Kevin

I use Monit.

http://mmonit.com/monit/

Very simple configuration syntax with configurable alert parameters. This example includes checking http service as well as ping. There is also a web interface for checking the status of monitored entities, which shows the "last seen" time, I believe. The "service detail" screenshot shows it:

http://mmonit.com/monit/screenshots/monit_httpd1.html

check host foo with address 192.168.1.20 if failed icmp type echo count 4 with timeout 15 seconds for 2 cycles then alert if failed url http://your.server.com/path/to/whatever with timeout 15 seconds 2 times for 2 cycles then alert alert **********@msg.telus.com alert **********@text.mts.net alert **********@text.mts.net only on { icmp, connection } mail-format { from: someaddress@example.com subject: Foo Server $EVENT message: Monitoring Foo Server on $DATE }