From: Jurgen Haan on
Hi hi,

I've got some network service that spawns agents upon client
connections. But I'm wondering if those things get cleaned up again.
Does anybody know how I can determine/see how long a process exists on a
server?

Thanks.

-R-
From: Chris Davies on
Jurgen Haan <jurgen(a)fake.tld> wrote:
> I've got some network service that spawns agents upon client
> connections. But I'm wondering if those things get cleaned up again.

If it's your code, make it clean up after the protocol ends, or the
sockets are closed. If it's someone else's code, maybe telling us what
it is might help. Some common network services hang around in case
more connections come in. (These are often UDP based servers.) Others
exit as soon as their share of the work has been completed (usually TCP
based servers.)


> Does anybody know how I can determine/see how long a process exists on a
> server?

The STIME field from "ps -f" ("ps -ef") shows you the time (date)
a process was started.

Chris
From: Jurgen Haan on
Chris Davies wrote:

> The STIME field from "ps -f" ("ps -ef") shows you the time (date)
> a process was started.
>
> Chris


Ah! Thanks. Exactly what I needed. :P
A classic case of RTFM!
 | 
Pages: 1
Prev: iptables help
Next: Multiple default routes