S
Sean
Here is the scenario:
Unix system, 100 machines in the "pool": one server, 99 clients. Main
program resides on the "server". On each of the 99 clients, there is
a daemon listening on a TCP port. In the main program (on the
server), it sends a signal out to the clients and waits for their
responses. The clients might take mili-seconds, MANY seconds, or
worst-case scenario lock up forever.
What is the best approach for the main program to wait a total of X
seconds for each of the clients responses (storing the string based
response in a hash array??) and then continuing on with the program so
as not to wait "forever" for the clients to respond?
If I "fork" them (or whatever it is I do), how do I make sure to get
their result?
Thanks for your help and any pointers,
Sean
Unix system, 100 machines in the "pool": one server, 99 clients. Main
program resides on the "server". On each of the 99 clients, there is
a daemon listening on a TCP port. In the main program (on the
server), it sends a signal out to the clients and waits for their
responses. The clients might take mili-seconds, MANY seconds, or
worst-case scenario lock up forever.
What is the best approach for the main program to wait a total of X
seconds for each of the clients responses (storing the string based
response in a hash array??) and then continuing on with the program so
as not to wait "forever" for the clients to respond?
If I "fork" them (or whatever it is I do), how do I make sure to get
their result?
Thanks for your help and any pointers,
Sean