tc-latency-poll-interval
Section: TServer
Default Value: 10
Valid Values: Any positive integer
Changes Take Effect: On the next call
Introduced: 8.1.103.58
Set in the VOIP Service DN with service-type=sip-cluster-nodes, this option specifies the time interval, in seconds, that each T-Controller polls its peers and updates *_QUEUE_LEN statistics. If set to 0 (zero), no polls are performed, all statistics are set to "disconnected value" and neither statistic from this feature is updated. See Troubleshooting T-Controllers communication issues for details.
Configuring Switch and DN Objects for SIP Cluster
Complete these steps to create configuration objects required for the SIP Cluster environment.
- Configure the SIP Cluster Switch
- Configure the SIP Cluster Node DN
- Configure the SIP Outbound Proxy DN
- Configure MSML DNs
- Configure Softswitch DNs
- Configure Trunk DNs
- Configure Routing Point DNs
- Configure Extension DNs
- Configure the Switch and DNs for Virtual Queues
Keep in mind the following:
- Agent Logins should not be configured in the SIP Cluster environment.
- All SIP Server applications representing SIP Cluster nodes must use the same switch.
- ACD Queue DNs are not supported in the SIP Cluster.
Configuring the SIP Cluster Switch
Create a Switch object of type SIP Switch dedicated to the SIP Cluster. with the name, for example, SIP_Cluster.
Configuring the SIP Cluster Node DN
The SIP Cluster Switch must contain a SIP Cluster Node DN. All SIP Server and SIP Proxy applications in the Cluster use the parameters configured in this DN.
- Under Switches > SIP_Cluster > DNs, create a DN of type Voice over IP Service named, for example, SIP_Cluster_DN.
- In the Annex > [TServer] section, configure the following mandatory options:
Name | Value | Notes | Example |
---|---|---|---|
addp-trace | full | The trace level for ADDP messages. | |
addp-remote-timeout | 11 | T-Controller ADDP protocol setting. | |
addp-timeout | 7 | T-Controller ADDP protocol setting. | |
agent-state-auto-restore | false | Enables restoration of agent states when the DN ownership changes between SIP Cluster nodes. When set to false, agents are logged out on the DN ownership change and must log in manually. When set to true, agent states are restored on the new primary SIP Cluster node and agents are logged in automatically.
Limitation: When an agent manually sets to NotReady state with ACW mode, SIP Server generates the ManualSetACWPeriod reason code automatically and does not restore the agent state on the new agent state owner (the SIP Cluster node). |
|
applications | <A comma-separated list of primary SIP Server application names in the cluster> | All SIP Server instances, except dummy SIP Server applications used by ICON, for all data centers. | applications=SIPS_sfo_1,SIPS_sfo_2,SIPS_ash_1,SIPS_ash_2 |
dn-owner-applications | <A comma-separated list of application names that can take DN ownership for a node> | The application name is sent to SIP Server as AttributeApplicationName in TRegisterClient. This name is used to identify a desktop and maintain DN ownership for remote agents. Usually, it is set to the name used by Genesys Web Services to register with SIP Server. | dn-owner-applications=Cluster |
replace-agent-phone | false | When set to true, enables modification of dial digits that are required to reach an agent DN. This feature supports remote agents to use external numbers that are not provisioned in the Configuration Database. This option setting affects all cluster internal DNs.
Feature limitations:
|
|
service-type | sip-cluster-nodes | ||
sip-enable-strict-auth | false | When set to true, enables SIP Server in SIP Cluster mode to mandate authorization of internal devices on REGISTER and INVITE requests. To register or establish communication, devices must not use empty passwords or passwords equal to the DN name. When this option is set to false, an internal device can register or establish communication with SIP Cluster without any authorization.
You can define the sip-enable-strict-auth option at the following levels listed in order of priority:
|
|
sipp-oos-recheck | false | When set to true, enables the Active Out-of-Service (OOS) check procedure for a VOIP service or Trunk DN to ensure that this DN is placed out of service only when SIP Proxy processes the SIP traffic. If set to true, SIP Server places the DN out of service only after rechecking that SIP Proxy is available, and does not place any other DN out of service if the SIP Proxy DN is already out of service. If set to false (the default), SIP Server behavior is not changed. As soon as the SIP Proxy DN is placed out of service, SIP Server initiates a switchover if the switchover-on-sipp-oos option is set to true. This enables establishing a connection to SIP Proxy from another HA-peer application. | |
sipproxy-applications | <A comma-separated list of all SIP Proxy application names in the SIP Cluster> | All SIP Proxy instances for all data centers. | sipproxy-applications=SIPProxy_sfo_1,SIPProxy_sfo_2,SIPProxy_ash_1,SIPProxy_ash_2 |
smart-proxy-enabled | false | Set to true only when the Smart Proxy module is enabled and all Stat Server applications are connected to the SmartProxy port. Setting to true optimizes event distribution between T-Controllers and improves SIP Cluster performance. If the option is set to false, T-Controllers work in normal mode and Smart Proxy modules, if enabled, work in compatibility mode. | |
switchover-on-sipp-oos | false | If set to true, SIP Server attempts to initiate a switchover immediately after the SIP Proxy VOIP DN is placed out of service. The switchover is initiated by the SERVICE_UNAVAILABLE message sent to LCA/SCS. When SIP Server is switched to backup mode by SCS, SIP Server issues the SERVICE_AVAILABLE message to SCS. The sipp-oos-recheck option must be set to true. | |
tc-reconnect-timeout | 0-60 | Specify the timeout, in seconds, during which T-Controller tries to reconnect to another SIP Cluster node when the connection between SIP Cluster nodes is lost. When the timeout expires and the connection is not restored, DNs owned by the disconnected T-Controller are declared out of service. The default value of 0 (zero) disables this functionality. | |
tc-latency-poll-interval | 10 | See Troubleshooting T-Controllers communication issues for details. |
Configuring the SIP Outbound Proxy DN
The SIP Cluster Switch must contain one SIP Outbound Proxy DN. All SIP Server and SIP Proxy applications in the cluster use the parameters configured in this DN.
- Under Switches > SIP_Cluster > DNs, create a DN of type Voice over IP Service named, for example, SIPProxy_DN.
- In the Annex > [TServer] section, configure the following mandatory options:
Name | Value | Notes | Example |
---|---|---|---|
contact | <The SIP access point of SIP Cluster (geo-aware SRV FQDN resolved to data center's SIP Proxy instances) with reliable transport> | Set this option to the Fully Qualified Domain Name (FQDN) that resolves to the SRV list of the SIP Proxy addresses. SIP Server uses this list to load-balance the traffic across all SIP Proxy instances available in a data center that is defined by geo-location. If there are multiple data centers within the cluster, this FQDN must be resolved to SIP Proxy instances located in the same data center with SIP Server, which is resolving the SRV FQDN. In other words, SIP Server must be aware only of those SIP Proxy instances installed in the same data center where SIP Server resides. It is the responsibility of the DNS to make this differentiation. This contact must be configured with the TCP transport to allow passing call attached data in SIP messages to GVP. | contact=spx-srv.example.com;transport=tcp |
external-contact | <SIP access point of SIP Cluster (geo-aware SRV FQDN resolved to data center's SIP Proxy instances)> | SIP Proxy sends the value of external-contact to its clients (e.g. GVP RM, SBC, SIP phones). SIP Cluster clients should be able to support the SRV FQDN. | external-contact=spx-srv.example.com |
oos-check | 10 | These two options enable active out-of-service detection. In this specific example, SIP Server is checking for SIP Proxies (by using the value of the contact option) every 10 seconds and sets them in the out-of-service state 2 seconds after the last check for which a response was not received. | |
oos-force | 2 | ||
service-type | sip-outbound-proxy |
Configuring MSML DNs
- Under Switches > SIP_Cluster > DNs, create a DN of type Voice over IP Service for each data center containing contacts of all Resource Managers in the environment.
- In the Annex > [TServer] section, configure the following configuration options:
Name | Value | Notes | Example |
---|---|---|---|
service-type | msml | ||
contact | <RM-SRV-FQDN>;transport=tcp | The SRV FQDN resolved to a pair of A-record FQDNs or IP addresses pointing to two Resource Managers configured as an active-active HA pair which are deployed to serve a particular data center. | contact=sfosrv-rm.example.com;transport=tcp |
geo-location | <String> | A string identifying the data center to which this DN belongs. | geo-location=sfo |
cpd-capability | mediaserver | ||
make-call-rfc3725-flow | 1 | ||
oos-check | 10 | ||
oos-error-check | true | ||
oos-force | 2 | ||
prefix | msml= | ||
refer-enabled | false | ||
ring-tone-on-make-call | false | ||
sip-uri-params | gvp-tenant-id=<IVR_Profile_for_DC> | SIP Server sends the name of the IVR profile to be used to Resource Manager as a parameter of the INVITE’s Request-URI header. The IVR profile will be used both for media services and recording. One IVR Profile must be created for each data center. | sip-uri-params=gvp-tenant-id=sfo-15 |
subscription-id | Environment | ||
userdata-map-filter | * | ||
userdata-map-format | sip-headers-encoded |
Configuring Softswitch DNs
Softswitch DNs are mandatory for remote agents.
- Under Switches > SIP_Cluster > DNs, create a DN of type Voice over IP Service for each data center in the SIP Cluster environment.
- In the Annex > [TServer] section, configure the following mandatory options:
Name | Value | Example |
---|---|---|
contact | <SBC-FQDN or IP-address>:<Agent-SIP-port> Note: The contact pointing to the SBC realm used to access PSTN agents. |
contact=sbc-example.com:5060 |
dual-dialog-enabled | false | |
geo-location | <A string identifying the data center to which this DN belongs> | geo-location=sfo |
make-call-rfc3725-flow | 1 | |
oos-check | 10 | |
oos-force | 2 | |
oos-options-max-forwads | true | |
record | true | |
refer-enabled | false | |
service-type | softswitch | |
sip-error-conversion | 408=486 | |
sip-proxy-headers-enabled | false | |
sip-ring-tone-mode | 1 |
Configuring Trunk DNs
- Under Switches > SIP_Cluster > DNs, create a DN of type Trunk for each data center configured for a particular tenant.
- In the Annex > [TServer] section, configure the following configuration options:
Name | Value | Notes | Example |
---|---|---|---|
contact | <SBC-FQDN or IP-address>:<SIP-port> | The contact pointing to the SBC realm. | contact=sbc-example.com:5060 |
geo-location | <String> | A string identifying the data center to which this DN belongs. | geo-location=sfo |
dual-dialog-enabled | false | ||
oos-check | 10 | ||
oos-error-check | true | ||
oos-force | 2 | ||
oos-options-max-forwards | 5 | ||
prefix | This option can be used the same way as in a standard SIP Server deployment. | + | |
priority | This option can be used the same way as in a standard SIP Server deployment. | 1 | |
refer-enabled | false | ||
sip-proxy-headers-enabled | false | ||
sip-ring-tone-mode | 1 |
Configuring Routing Point DNs
Under Switches > SIP_Cluster > DNs, configure a DN of type Routing Point to be used for default routing in the SIP Cluster. Configure other Routing Point DNs as required for your environment.
Configuring Extension DNs
- Under Switches > SIP_Cluster > DNs >, create DNs of type Extension for all agent phones for a particular tenant.
- For agents using PSTN phones, configure extension DNs by specifying their phone numbers.
- For agents using SIP phones, in the Annex > [TServer] section, configure the following configuration options:
Name | Value |
---|---|
contact | * |
dual-dialog-enabled | true |
make-call-rfc3725-flow | 1 |
refer-enabled | false |
use-register-for-service-state | true |
use-contact-as-dn | true |
sip-ring-tone-mode | 1 |
sip-cti-control | talk,hold (or talk,hold,dtmf) |
Configuring a Switch and DNs for Virtual Queues
Large contact center deployments may generate significant traffic on the Virtual Queue (VQ) DNs to satisfy complex reporting and monitoring requirements. In the SIP Cluster architecture, this traffic is isolated on a dedicated VQ switch served by several HA pairs of SIP Servers running in a non-cluster mode with one pair deployed in each data center. SIP Server instances serving the VQ Switch are also used for agent reservation.
Create a Switch object of type SIP Switch dedicated to Virtual Queue DNs with the name, for example, VQ-switch. Under the VQ-switch, configure DNs of type Virtual Queue. Only DNs of this type must be configured under this VQ-switch.
You will assign this VQ Switch to a SIP Server pair dedicated to VQ DNs.