The
ConnectionProfiles key contains an organized list of the connection information
of all configured CTI OS Servers present in the corporate network that you can
access by a client application. The connection profiles are defined in the
registry of each server at the following location:
HKEY_LOCAL_MACHINE\SOFTWARE\Cisco Systems
Inc.\CtiOs\<CTIOS
InstanceName>\<CTIOSServerName>\ EnterpriseDesktopSettings\All
Desktops\Login\ConnectionProfiles\Name\<CtiOsProfileName>
To create
a profile for a given server, you must define a subkey under
ConnectionProfiles\Name with the following format:
[HKEY_LOCAL_MACHINE\Software\…\ConnectionProfiles\Name\CtiOsProfileName]
"PeripheralID"=dword:5000
"Heartbeat"=dword:00000000
"MaxHeartbeats"=dword:00000005
"CtiOsA"="HostName_A"
"CtiOsB"="HostName_B"
"PortA"=dword:0000a42c
"PortB"=dword:0000a42c
"AutoLogin"=dword:00000001
“ShowFieldBitMask”=dword:00000023
"WarnIfAlreadyLoggedIn"=dword:00000001
"RejectIfAlreadyLoggedIn"=dword:00000000
"DisableSkillGroupStatistics"=dword:00000001
"DisableAgentStatistics"=dword:00000001
"UCCESilentMonitorEnabled"=dword:0x00000001
"WarnIfSilentMonitored"=0x00000000
The
following table describes the required ConnectionProfiles key registry values.
Table 18 ConnectionProfiles Key Registry Values
SubKey/Value
|
Description
|
CtiOsProfileName
|
The name
given to the profile. This string appears on the Login Dialog when a user is
about to log in using the CTI OS Agent State Control.
|
PeripheralID
|
The numeric
value of the peripheral to which the CTI OS Server connects.
|
Heartbeat
|
Time
interval between heartbeat messages between the client and CTI OS Server.
|
MaxHeartbeats
|
Maximum
number of heartbeats that can be missed by the CTI OS Client Session before
failover occurs.
|
CtiOsA
|
DNS name of
IP Address of the primary CTI OS Server to which a client application can
connect.
|
CtiOsB
|
DNS name of
IP Address of the secondary CTI OS Server to which a client application can
connect.
|
PortA
|
TCP/IP port
number assigned to the primary server.
|
PortB
|
TCP/IP port
number assigned to the secondary server.
|
AutoLogin
|
Indicates if
the client must automatically log in an agent or supervisor after it recovers
from a system failure. For all peripherals other than UCCE you must set this
field to 0x00000000. For UCCE, set this field to 0x00000001.
|
ShowFieldBit
Mask
|
Indicates what fields appear in the CTI OS Login dialog box.
Fields appear on the dialog box only if their corresponding bit in the mask is
on. The possible fields and their corresponding masks are shown in the table
"ShowBitFieldMask Fields" below. The default value at setup
for ShowFieldBit Mask is 0x00000023 (AgentID, Instrument, and Password
displayed).
|
WarnIfAlready LoggedIn
|
Indicates
whether to display a warning but still permit login if an agent who is already
logged in attempts to log in again. A value of 1 (default) enables the warning;
a value of 0 disables the warning. This value is relevant only if
RejectIfAlreadyLoggedIn is 0.
|
RejectIfAlready LoggedIn
|
Indicates
whether or not to permit an agent who is already logged in to log in again. A
value of 0 (default) permits an agent to log in again. A value of 1 prohibits
an agent from logging in again.
|
DisableSkillGroup Statistics
|
Indicates
whether skill group statistics are enabled for the agent using this connection
profile. A value of 1 disables statistics. If this value is 0 (default) or not
present, skill group statistics are enabled for this agent.
|
DisableAgent
Statistics
|
Indicates
whether agent statistics are enabled for the agent using this connection
profile. A value of 1 disables statistics. If this value is 0 (default) or not
present, statistics are enabled for this agent.
|
IPCCSilent
MonitorEnabled
|
Indicates
whether silent monitor is enabled for the clients using this connection
profile. A value of 0x00000001 (default) enables silent monitor. If this value
is 0x00000000 or not present, silent monitor is disabled for this client. For
all peripherals other than UCCE, you must set this field to 0x00000000.
|
WarnIfSilent
Monitored
|
Indicates
whether to display an indicator on the agent desktop when the agent is silent
monitored by the team supervisor. A value of 0x00000001 causes a message to
appear on the agent desktop when the supervisor is silent monitoring this
agent. If this value is 0x00000000 (default) or not present, no message appears
on the agent desktop when the supervisor is silent monitoring this agent.
|
RasCallMode
|
Indicates
the agent work mode options for the mobile agent login dialog box. Valid values
are 0 (agent chooses), 1 (call by call), and 2 (nailed up).
|
Table 19 ShowBitFieldMask Fields
Field
|
Mask
|
Instrument
|
0x00000001
|
Password
|
0x00000002
|
Work Mode
|
0x00000004
|
Position
ID
|
0x00000008
|
Skillgroup
|
0x00000010
|
AgentID
|
0x00000020
|
Login Name
|
0x00000040
|
Mobile
Agent
|
0x00000080
|
The
heartbeating mechanism uses the MaxHeartbeats and Heartbeat values together to
determine when a client must send heartbeat requests to the server and when the
client must connect to the other server.
MaxHeartbeats is the max number of missed heartbeats before
failover. (Default = 5)
Heartbeat is the time interval between consecutive heartbeats.
(Default = 5)
This is
how the heartbeating mechanism works on the CTI OS client:
- After 5 seconds, if the
client does not receive a response from the server, it sends a heartbeat
request 1.
- After 5 seconds, if the
client does not receive a response from the server, it sends a heartbeat
request 2.
- After another 5 seconds, if
the client does not receive a response from the server, it sends a heartbeat
request 3.
- After yet another 5
seconds, if the client does not receive any response from the server, it
connects to an alternative server.
Note |
The amount of
time it takes a client to reconnect to the other server depends on the type of
failure that occurs.
|
The
heartbeat parameters above are only a factor if the TCP/IP socket is not
broken. For example, if you disconnect the network cable to the CTI OS Server,
TCP/IP does not break the socket. In this case, the client uses the
heartbeating mechanism listed above to detect the failure.
In a
different case, however, if the CTI OS Server process crashes or the machine is
turned off, the socket breaks and the client immediately knows that the
connection has failed. In this case, the client directly connects to the other
server without heartbeat attempts.
Note |
In either case,
although the socket connection might get established right away, it might take
a few more seconds for the agents to fully recover their previous, pre-failure
state. This delay might particularly be experienced if many agents are failing
over at the same time, or if the system is experiencing a heavy call load at
the time of the failure.
|