Unix-Process-ids and tuxedo - Behaviour ?

Question

We have a situation where different tuxedo servers are acquiring a process id which was allocated to a different one after some time.
Is this an error.
Here is an example :
Tux Servers Process ID
Server 1 100
Server 2 200
and at a later time :
Tux Servers Process ID
Server 1 200
Per Tuxedo server 2 is in sleep mode and the server 1 is running , is this normal.
Server 1 and Server 2 are in same group.

Answer 1

A process on Unix or Linux will retain its process ID for as long as it is
alive. Once the process dies, its process ID can be reused by another
process.
If server1 and server2 are shutdown or die and server1 is rebooted or
restarted, it is possible for server1 to have the same process ID that
server2 originally had.
Process IDs on Unix have traditionally gone up to 30000, which was fine to
ensure that a process ID would not be quickly reused back in the days when
computers were slower and supported fewer concurrent processes. In recent
decades, IBM and some other Unix vendors have increased the allowable range
for a process id well beyond 30000, which results in process numbers being
reused less frequently.
If the reuse of process IDs is causing problems for scripts in your
application, you may want to look at the process start time as well as the
process ID. In "ps -f" output, this is under the STIME column. If using
the T_CLIENT MIB class, you can look at the TA_TIMESTART attribute, and if
using the T_SERVER MIB class, you can look at the TA_TIMERESTART MIB
attribute.
Ed
<hunki> wrote in message news:[email protected]..
We have a situation where different tuxedo servers are acquiring a process
id which was allocated to a different one after some time.
Is this an error.
Here is an example :
Tux Servers Process ID
Server 1 100
Server 2 200
and at a later time :
Tux Servers Process ID
Server 1 200
Per Tuxedo server 2 is in sleep mode and the server 1 is running , is this
normal.
Server 1 and Server 2 are in same group.