We have a mixed environment of Suns, SGIs, and HPs. The biggest group
are the Suns, and there are several of them running several different
versions of Solaris, ranging from 2.4 to 2.5.1. The SGIs are running
Irix 5.3. And there is one HP 712 workstation running HPUX 9.03.
Okay, now that we've described the environment we have to work with,
this is what we want to do. We want to get some recommendations on
server/network management software. There seems to be a bewildering
array of available choices, and we're not entirely sure what we want it
to do for us. In general, we want it to warn of low disk space, cpu and
i/o load averages, number of processes, there's going to be some servers
with Sybase on them, and we'd like it to log and graph any of this data
and be managed from a single remote control machine.
We'd like an opinion on these specific products: a) Tivoli, b) BMC
Patrol, c) HP OpenView, d) Solstice Management Suite, e) CA Unicenter,
and f) anything else you can think of.
Some of these products don't seem to be in competition with each other,
but are rather complementary products, designed to be plugins for the
other products mentioned up above. This is confusing, as we said, we're
not entirely sure of where to start and how things fit with each other.
We preferably want a single solution, without having to buy pieces from
several independent vendors.
Now, few specific questions:
-Do the two products created by the workstation manufacturers (HP
Openview and Sun Solstice) support other platforms besides their own.
For example is there an HP Openview for Sun, or a Solstice for IRIX? If
this is going to present a problem, I'll just let you know that we
consider the Suns more important to find a solution for, over the SGI or
HP boxes, since we have so many more of them. There's only two SGI
boxes, and one HP box, but dozens of Sun boxes.
-I've used BMC Patrol (briefly) in the past. As far as I recall, Patrol
installs a whole bunch of daemons into each machine, and this by itself
can lead to serious depletion of resources on each machine. The resource
monitoring utilities are themselves causing the resources to go down, in
other words. It could've been a bug in the version I used back then, or
it could be still there as a problem. Can anybody with BMC care to
elaborate on what problems they had?
-what is best to use for cross-platform compatibility?