11-19-2008 4:36 AM
Hi I have been trying to understand the working of the F&R Dispatcher and I have some basic queries to which I am yet to get a satisfactory answer. I was wondering if I could get some help on the same. Please find my query below.
1. Could someone please explain as to how should we interpret the LATENCY TIME? What does it exactly mean?
2. I am also not clear as to what the 2 values of latency time mean? For e.g. 2 2; 3 3 and so on!
3. Also, how do I read/ interpret the logs?
4. How do I know if the F&R Dispatcher is running OK?
Thanks in advance!
Vismay Jani
11-19-2008 12:41 PM
Hi,
From the doc :
1.
Latency Time Settings for the Dispatcher Processes
Each process leaves an update time stamp in the database. The
dispatcher's processes coordinate themselves by checking the update
time information in each other's database. The latency time is added to
the received time. If the result is earlier than the current time, the
process to be checked is assumed to be dead and it will be restarted.
2. explained in 1
3. from transaction /FRE/FRP_DISP_ACT - Activate and Start Dispatcher button log
4. from transaction /FRE/FRP_DISP_ACT - Activate and Start Dispatcher
Thanks,
regards,
Nicolas
11-19-2008 12:41 PM
Hi,
From the doc :
1.
Latency Time Settings for the Dispatcher Processes
Each process leaves an update time stamp in the database. The
dispatcher's processes coordinate themselves by checking the update
time information in each other's database. The latency time is added to
the received time. If the result is earlier than the current time, the
process to be checked is assumed to be dead and it will be restarted.
2. explained in 1
3. from transaction /FRE/FRP_DISP_ACT - Activate and Start Dispatcher button log
4. from transaction /FRE/FRP_DISP_ACT - Activate and Start Dispatcher
Thanks,
regards,
Nicolas
11-20-2008 9:50 AM
Nicolas, thanks for your response. I find this explanation on LATENCY TIMES in the system itself, when I do an F1. I am looking for some more information. Why are there 2 sets of times shown? I still have no clue about that. Could you throw some light on that? What does 2 2, 3 3 and so on 11 11, or 12 12 mean? Why is there not just ONE time?
11-21-2008 12:05 PM
here it possible to differentiate the latency time (performance purpose) for
- dispatcher
-controller
-event handler
.
In general 99 % of the customers have the same latency time for the 3.