Hi.  We currently have Netview 6.0.2 on AIX 4.3.   I have found some 
discussions on my question in the archives, but I'm still unclear on quite a 
few things.   

Today, this is implemented on a single machine, but we want to add redundancy.  
We currently configure our devices to send traps to this machine, which 
processes them in a ruleset, and forwards some of the events to T/EC for 
notification.  We also want to move to a more proactive approach by pinging (or 
something else) the devices.  The devices in question are routers, switches, 
hubs, and eventually servers.

Our UNIX support people would like to provide this with HAGEO.  This seems to 
be covered somewhat in the manual...at least HACMP is.  It also looks like 
MLM's could help with this??  Also, I'm unclear as the the purpose of Netview 
clients.

MLM's appear to be a way to localize the polling traffic.  Since none of our 
WAN sites are large enough to warrant an MLM (IMO), and we are willing to 
accept this traffic over the WAN links, I'm leaning toward the HAGEO solution.  
Where do NV clients fit into all of this?

The last thing that complicates this is that we also use CiscoWorks2000.  It 
appears that to get the full integration of the products, they need to live on 
the same machine.  Since this product is only supported on NT and Solaris, the 
common platform for us would be Solaris.  This solution throws the HAGEO 
redundancy solution out the window.

Has anybody else been through this before and would care to share their 
thoughts?  I'm really struggling with what the best solution is...I'm not even 
sure I have a decent handle on all of the pro's and con's of any proposed 
solution.

Thanks very much for any help or insights you can provide.

Craig Treptow
Principal Financial Group
I/S Network Administration