TW>> I'm of the opinion that it can be done, too --
 SD>> Your hopeing so ?
 TW> Yes and no.  Yes because it would be good to wipe the smirk off of the
 TW> face of the supercilious jerk that runs the servers who says it can't be
 TW> done, no way, no how -- and no, because in all the time I've worked with
 TW> NetWare, I've never seen any app or series of commands (or anything else
 TW> for that matter) that would bring down any server, any time.
Tony, I haven't seen it happen since the 2.x days. A concerted effort from a 
large number of machines could exhaust Netware resources, for instance locks. 
If you have access to the System directory and care to remove some choice 
Files, the server might not be able to reboot.
I've heard that a large number of findfirst calls without searching to the 
end of the directory can be a problem, but I can't verify this with a test 
program. Perhaps this reported problem is due to the maximum NCP searches 
that a single workstation can have open (which I _can_ duplicate by running 
60 processes under OS/2), but that just exhausts resources to the rogue 
machine.
If you take a moment to assume there are security holes, typically you see in 
the trade press an announcement by some lab that particular security features 
of a NOS can be exploited. This is usually followed immediately by the 
announcement that the fixes for the hole in question can be downloaded over 
the Internet. Microsoft NT and W95 are continually having their security 
breached, with patch upon patch being provided. I've _never_ seen Netware or 
Warp Server receive this kind of press.
The only reliable way I know of to crash a server is to have remote access to 
a buggy 3rd party NLM running on the server (like Arcserve, Backup Exec, 
F-Prot, etc) and to know how to crash it, which I can do almost without 
trying.
Jeff
--- GoldED/2 2.42.G0615
---------------