We have a java related app running in a server and many scripts to monitor the java app functionality scheduled in crons. We have similar setup for many servers, but in only select servers, we find few defunct process (java apps). At times servers becomes inaccessible and we depend on our onsite guys to do a hard reboot...in most cases its kernal panic...and when ever there is kernel panic, there are many defunct process.....
as zombie does not occupy resources, but it does occupy the PIDs....but i m not sure whether many zombie will lead to kernel panic......

Zombies by themselves don't cause a panic. They are simply processes who exited and are still waiting for the parent process to clean them up. I.e. each process will be a zombie when it exits, but under normal circumstances it will disappear quickly.

It could be helpful to know what panic occurs.

@Iyyappan: Zombie processes can be the (indirect) cause for a kernel panic.

In normal circumstances a process is cleaned up (as described by berndbausch), but in that case we do not talk about a zombie process (although technically they are for a very short time).

A "real" zombie process doesn't have a parent any more and as a result the process will stay in the process list and use 1 PID (resources ARE used,). If the system runs out of PID's it can assign the result will be a kernel panic (or at least a completely unresponsive system).

If a badly written program (one that creates one or more zombie processes for instance) is launched from cron on a very regular base the system will run out of PID's it can assign at some point.

There isn't enough information in your first post to determine what is going on in your case and the cause might be something else altogether. My guess would be that one of the (java) applications causes the kernel panic.

Admittedly I only know UNIX at that level; Linux may be different. On a UNIX system, a process always has a parent, and if that parent dies, the process is "adopted" by process #1, i.e. the init process. If the adopted process is a zombie, init will immediately clean it up.
But of course, this is not what the original poster wanted to know.

As you hinted, the careless creation of zombie processes could be a cause for exhausting kernel resources and, as a consequence, a panic or at least an inaccessible system. But to be more certain we would need to know what kind(s) of panic occurred.