10.2.7.2. Monitoring the Runtime for Signaling Messages (Round
Trip Delay) 
If a malfunction occurs in the IP network, the effective bandwidth
available drops and the time between the dispatch of a message and receipt
acknowledgement rises.
A maximum permissible message runtime can be defined for the signaling
connection for every access point. This runtime includes the entire time
from sending the message to receiving the acknowledgement, in other words,
the “Round Trip Delay“. The runtime is evaluated with every signaling
message. Measurement is therefore independent of the signaling load.
If the runtime exceeds the limit set, the system outputs the error
message F8290 (see Section 2.7.6.2,
“F8290 - Net Weakness Begin”). The exceeding of the
maximum message runtime can also be a criteria for changing from the
signaling path to the signaling survivability path (see Section 2.7.4,
“Advanced Criteria for Signaling Survivability”).
The message F8291 is output to signal a return by the runtime to permitted
values (see Section 2.7.6.3,
“F8291 - Net Weakness End”).
Monitoring the maximum message runtime is only active on the signaling
path over LAN. The modem connection is not affected in the case of signaling
survivability.
Procedure
An average runtime value is calculated and stored every three seconds
for all messages acknowledged in this interval.
The three-second average values are used to generate further average
values over a short (15 seconds: SHORT) and long monitoring time (60
seconds: LONG).
The error message is generated whenever either of the two values (SHORT/LONG)
exceeds the limit configured.
The limit is set for the long monitoring duration (LONG). The limit
for the short duration (SHORT) is set to factor 2 times the value.
The SHORT interval makes sense with a higher limit because if the
network quality jumps, the system can respond faster, that is, within
15 seconds. This interval would be 60 seconds in the case of exact measurement.
The measurement is made at “application level“, in other words,
over the TCP/IP stack. On account of stack dynamics, measured values
cannot be compared directly with values that are measured directly on
the network.
The runtime includes:
- Sender of packet from TCP level in the OpenScape 4000 central system
- Transmission of packet in the network
- Receipt of packet on HG 3575 up to TCP
- Creation and sending of acknowledgement on the HG 3575
- Transmission of acknowledgement in the network
- Receipt of acknowledgement in the OpenScape 4000 central system
Since TCP can delay the acknowledgement of a packet in order to combine the acknowledgement with that of another packet, values under 400 ms are impractical.Keep alive packets are not included in the measurement. If no packets are sent or acknowledged within a 3-second interval, a long-term mean value is used that is determined by the TCP.
Generation
 |
Configuration Management > System Data
> IPDA > Access point
Click Search and enter
or change the maximum runtime on the Quality of Service tab in
the Signaling Quality of Service section, then Save.
|
 |
CHA-STMIB:MTYPE=NCUI2,TYPE=SIGQOS,LTU=99,MAXRTD=500; |
The message runtime is expressed here in milliseconds. The value
MAXRTD=0
disables message runtime monitoring. The setting becomes effective immediately.IMPORTANT:
Measurements are performed at 10-ms intervals.
As a result, it only makes sense to set values divisible by 10 (without
remainder), for example, 400, 410, 420, etc.