If you have all services (sites) on one server included in a single WUG
object (server/icon), then WUG will ping the server, if it is up then it
will test the services / sites.  

If you have each site as a separate object / icon, then you also need to
create one for the main server itself.  Then create an icon for each web
site. Set it's pulling method as TCP, add a http service to each icon.  Then
set each site icon to be up dependant on the main server icon.  It will only
poll the site icons if the main server icon is up.  However the server could
go down after it has been polled, and would show all of the sites as down.
But then that is only an issue if you have alerting on the first bad poll.

Jeff Cook
Network Administrator
Whatcom Educational Credit Union


-----Original Message-----
From: John Batdorf [mailto:[EMAIL PROTECTED]]
Sent: Monday, February 10, 2003 2:53 PM
To: [EMAIL PROTECTED]
Subject: [WhatsUp Forum] Quick question about Web Site monitoring.... 


Hello, new to the forum, and just wanted to run this scenario past the 
vetran WUG users!  Please bear with me, it will take a minute to explain.

Network Infrastructure:

Our network has a load balancer, with three web servers in the webfarm.  
Monitoring the site through the load balancer is done as a service, and the 
domain name.  Easy.  Now when I reboot a server in the web farm, no 
problems, as the Load Balancer (LB) detects a server is not available, and 
routs traffic around the missing server.

We also are monitoring *each* site on *each* LB server.  So we have created 
a vast array of host names, web0x-project.test.url.com (where URL) is our 
company name.  We've added all these host headers to our DNS servers, and 
set them up as services in WUG.  Getting to my question.. really.  So these 
can all be monitored at an INDAVIDUAL level, each server's IP and using host

headers through the service, as well as THROUGH the LB.

The map for monitoring the indavidual server's sites is three servers that 
must respond to ICMP ping request, and then each server has 40 or so sites 
that are under it.  I had set up a UP dependency on the server itslelf 
before checking the web sites, but found out this is not how WUG works, and 
you can only have dependencies between devices and not between a device and 
a service.

Ok... here's the question:  So is there a way, that when I reboot a server, 
if WUG determines the server is down, to NOT check the sites on that server?

  Make sense?  I spent a long time explaining this to the TECHNICAL SUPPORT 
MANAGER at IPSWITCH.  Here's another way to look at the problem.  Say you 
have a server with 10 web sites on it.  You reboot the server, how do you 
not get inandated with s bunch of pages.

We've set the poll cycle to 100 seconds, and to not alert us until 4 missed 
poll cycles, this seems to be ok for most reboots, but I'd like to figure a 
way to tell WUG to not even check the sites if the server is unavailable.

Thanks... sorry so long, or if it's confusing.

John Batdorf

_________________________________________________________________
MSN 8 with e-mail virus protection service: 2 months FREE*  
http://join.msn.com/?page=features/virus


Please visit http://www.ipswitch.com/support/mailing-lists.html 
to be removed from this list.

An Archive of this list is available at:
http://www.mail-archive.com/whatsup_forum%40list.ipswitch.com/

Please visit http://www.ipswitch.com/support/mailing-lists.html 
to be removed from this list.

An Archive of this list is available at:
http://www.mail-archive.com/whatsup_forum%40list.ipswitch.com/

Reply via email to