Professional Documents
Culture Documents
Contents
Getting Started . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
What is ACD Resiliency? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Levels of resiliency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Basic ACD Resiliency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Advanced ACD Resiliency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Full ACD Resiliency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Resilient network topology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 OPS Manager and ACD Resiliency. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 3300 ICP and ACD Resiliency. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Implementing ACD Resiliency for your contact center . . . . . . . . . . . . . . . . . . . . . . . . . 16 Installing OPS Manager. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 About ACD Hot Desk Agents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 3300 ICP ACD Agent Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 OPS Manager and ACD Agent Resiliency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 Configuring ACD Agents with Resiliency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 OPS Manager Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 Resiliency policy settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 Programming your Mitel telephone systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Supported resilient features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Fail-over States. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Failback/Handoff States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Agent Failover and Call Redirection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Installing Contact Center Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 Contact Center Management resiliency concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 Understanding your enterprise structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 Organizing media servers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 Configuration overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Configuring ACD Resiliency in the YourSite database . . . . . . . . . . . . . . . . . . . . . . . . . 31 Configuring enterprise settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Adding sites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Adding 3300 ICP media servers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Configuring data collection settings for a 3300 ICP . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 Configuring data summary options for a 3300 ICP . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 Configuring MiTAI options for a 3300 ICP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Configuring Network Monitor settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Adding employees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Using Quick Setup to add employees (and agent login IDs) . . . . . . . . . . . . . . . . . . . . 33 Adding/modifying employees manually . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Adding agents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 Using Quick Setup to add agents (and employee IDs) . . . . . . . . . . . . . . . . . . . . . . . . . 38 Adding agents manually . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Adding agent groups to the YourSite database . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Associating agents to agent groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
ii
Adding extensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Using Quick Setup to add extensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Adding extensions manually . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Adding queues. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Using Quick Setup to add queues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Adding queues manually. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Adding queue groups to the YourSite database . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Associating agent groups with queues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
42 42 42 43 44 45 47 47
Getting Started
Mitel ACD Resiliency supports the resiliency, scalability, and virtual contact center models of the Mitel telephone platforms. It is an add-on option to Contact Center Enterprise Edition or Mitel Call Accounting and works in conjunction with Mitel OPS Manager. This guide provides information on ACD Resiliency concepts and how to configure your enterprise in Contact Center Management, specifically the YourSite database. To fully understand the information provided in this document, technicians and administrators must have a strong understanding of, or experience with
Mitel 3300 IP Communications Platform Mitel OPS Manager 3300 ICP network planning and configuration 3300 ICP System Administration Tool Mitel IP phone functionality IP Networking
For information about the Mitel 3300 IP Communications Platform, refer to the following online documents at http://edocs.mitel.com:
3300 ICP Technicians Handbook 3300 ICP Hardware Technical Reference Manual 3300 ICP Engineering Guidelines 3300 ICP Resiliency Guidelines Voice Clustering (Portable Directory Number) 3300 ICP Software Installer Tool Help
For information about Mitel OPS Manager, refer to the following online documents at http:// edocs.mitel.com:
Enterprise Manager Technicians Handbook OPS Manager Online Help
For information about any Mitel IP devices mentioned in this document, see the device-specific documentation on http://edocs.mitel.com, under "User Guides".
Resiliency gives your network the ability to maintain calls in progress, handle new incoming and outgoing calls, and continue to provide voice mail service in the event of 3300 ICP failure or a network-level failure. Resiliency is achieved through setting up a network of 3300 ICPs in a resilient cluster, which is a specially configured network of 3300 ICPs that can direct IP phones and route and maintain calls. The transfer of phone service between primary and secondary ICPs as well as the maintenance of calls in progress ensures that most system failures are not noticed by desktop users.
Levels of resiliency
Configurations are classed based on the level of resiliency achieved:
Basic ACD Resiliency Advanced ACD Resiliency Full ACD Resiliency
Levels of resiliency
Failover Conditions
Return-to-Service Conditions
The advantages:
No single point of failure. Most economical two controller solution. 100 percent of agents remain in service. 100 percent of trunks remain in service if each controller is provisioned with the required hardware to support 100% of the trunk traffic.
The disadvantages:
Less resilient than Advanced or Full Resiliency Configurations. Limited scalability. Maximum of 100 IP agents per controller. A controller outage can reduce the number of trunks by 50% if each controller is not provisioned to support 100% of the call traffic.
Basic Resiliency - Configuration 2 In Configuration 2 (Figure 2) exchange calls are split evenly between the ACD controller 1 and ACD controller 2. IP Networking is used between 3300 ICP controllers. Figure 2 Basic Resiliency - Configuration 2
In Configuration 2:
50 ACD hot desk agents are programmed on ACD controller 1 as resilient devices and will fail over to ACD controller 2 in the event that ACD controller 1 fails. 50 ACD hot desk agents are programmed on ACD controller 2 as resilient devices and will fail over to ACD controller 1 in the event that ACD controller 2 fails.
Levels of resiliency
Resiliency is enabled for the agent groups: On ACD controller 1, resilient agent groups are configured with their primary controller as ACD controller 1 and their secondary controller configured as ACD controller 2. The group members are resilient hot desk agents. Their primary controller is ACD controller 1 and their secondary controller is ACD controller 2. On ACD controller 2, resilient agent groups are configured with their primary controller as ACD controller 2 and their secondary controller configured as ACD controller 1. The group members are resilient hot desk agents. Their primary controller is ACD controller 2 and their secondary controller is ACD controller 1. Calls arriving through ACD controller 1 queue locally on group A and then remotely on ACD controller 2 via NETACD on group B. Calls arriving through ACD controller 2 queue locally on group B and then remotely on ACD controller 1 via NETACD on group A. For both configurations (Figure 1 and Figure 2), the following conditions apply:
For both, configurations 1 and 2 (Figures 1 and 2), the following conditions apply
3300 ICP Release 7.0 software or later is required on the controllers. System Data Synchronization must be enabled on the primary and secondary controllers. By default, SDS shares user and device data at the "Resilient Pair" scope. Mitel OPS Manager is required to program resiliency for the group members. Agent group resiliency must be programmed through the System Administration Tool. For call center reporting, Mitel Customer Interaction Solutions collection points are required for both nodes. RADs must be configured on the digital trunking gateway by using the "embedded RAD" functionality that is available through the 3300 ICP embedded voice mail or by using a third-party RAD unit (for example, Mitel Contact Center Intelligent Queue, Interalia, or similar unit). Note that the embedded RADs do not use E2T channels; however, the Intelligent Queue is an IP integration that uses E2T channels when RAD recordings are played. Up to 256 paths are supported on each controller. Up to 1,181 agent IDs are supported on each controller. Up to 100 IP ACD agents are supported on each controller.
Failover Conditions
Return-to-Service Conditions
Any calls that were queued on ACD controller 2 while ACD controller 1 was down will be requeued on the ACD controller 1.
The advantages:
No single point of failure. Most economical two controller solution. 100 percent of agents remain in service. 100 percent of trunks remain in service if each controller is provisioned with the required hardware to support 100% of the trunk traffic.
The disadvantages:
Less resilient than Advanced or Full Resiliency Configurations. Not scalable -- limited to 100 IP agents per controller.
Levels of resiliency
System Data Synchronization (SDS) is sharing data among the network and elements. Mitel OPS Manager is required to program resiliency for the group members. Agent group resiliency must be programmed through the System Administration Tool. For Call Center reporting, Customer Interaction Solutions collection points are required for all nodes.
Failover Conditions
If the primary controller fails, calls that are queued to a resilient agent group on the primary are rerouted to its resilient agent group on the secondary controller. After the primary returns to service, calls are automatically rerouted from resilient agent group on the secondary controller to the resilient agent group on the primary controller. When the primary controller fails, agents may not fail over immediately to the secondary controller. For example, all agents could be engaged in calls that remain up after the primary fails (call survival). If new calls arrive at the secondary controller before any agents have failed over, the calls will follow path unavailable routing on the secondary controller. However, if you set the "Queue Callers to Group When No Local Agents are Logged In" option to "Yes" for the resilient agent groups, the incoming calls will queue to the resilient agent group and wait for the agents to home to the secondary. Note that if this option is enabled, then the last agent who logs out of the group at the end of the shift or workday must manually set the path to unavailable.
Return-to-Service Conditions Trunking gateway detects that the node is in service and once again resumes queuing on the primary controller. 120 ACD hot desk agents home back to their primary controller when idle and health check is complete. After agent devices fail back from the secondary controller to primary controller, the agents are automatically logged into the system and can begin accepting calls. Any calls that are active on the secondary controller will be allowed to complete. After agents hang up, their sets fail back. Any calls that were queued on the secondary controller while the primary was down will be rerouted to the primary controller. 120 agents now reside on the primary controller.
The advantages:
If the primary controller fails 100 percent of agents remain in service 100 percent of trunks remain in service. Agents in talk state can complete active calls before failover. Scalability - Up to 350 active agents can be supported on the primary or secondary controller depending on the number of gateways. In configuration 3 Cx and CXi can support up to 65 agents (all IP phones) with one gateway MX can support up to 128 agents (all IP phones) with one gateway MXe (with 128 MHz processor) and LX can support up to 128 agents (all IP phones) with one gateway. Up to 350 IP phone agents can be supported with additional gateways. NOTE: Only IP phones support ACD hot desk agents.
The disadvantages:
Gateway is a single point of failure (although it resembles a Digital Service Unit (DSU) and is inherently more reliable than the agent controller).
Levels of resiliency
Calls that arrive through trunking gateway 1, queue to a path on the gateway. The gateway then queues the calls on the ACD primary controller. Calls that arrive through trunking gateway 2, queue to a path on gateway 2. The gateway then queues the calls on the ACD primary controller. Physical channels are not used between the gateway and primary controller until calls are distributed to the agents. Under normal operation, calls are only answered from the primary controller. 3300 ICP Release 7.0 software is required on the controllers. Load sharing (queueing) is provided across two trunking gateways. ACD paths are programmed with the same agent groups and same agents on both trunking gateways and handle calls for the same 800 number. Central Office (CO) calls are split evenly between trunking gateways 1 and 2. IP Networking is configured between the controllers. Longest idle agent call distribution is supported. Queued callers listen to RADs and MOH at the trunking gateways and are distributed to agents via Networked ACD as agents become available. Up to 256 paths are supported on the trunking gateways Up to 1181 agent IDs are supported on the controllers Scalability - Up to 350 active agents can be supported on the primary or secondary controller depending on the number of gateways. In configuration 4 Cx and CXi can support up to 65 agents (all IP phones) MX can support up to 128 agents (all IP phones) MXe (with 128 MHz processor) and LX can support up to 256 agents (all IP phones) with two gateways. Up to 350 IP phone agents can be supported with additional gateways. NOTE: Only IP phones support ACD hot desk agents. RADs can be configured on the trunking gateway using the "embedded RAD" functionality available through the 3300 ICP embedded voice mail or by using a third-party RAD box (Intelligent Queue, Interalia, and so forth). System Data Synchronization (SDS) is sharing data among the network and elements. Mitel OPS Manager is required to program resiliency for the group members. Agent group resiliency must be programmed through the System Administration Tool. For Call Center reporting, Customer Interaction Solutions collection points are required for all nodes.
10
Levels of resiliency
11
calls are automatically rerouted from resilient agent group on the secondary controller to the resilient agent group on the primary controller. Any calls that are active prior to the primary controller failure will be allowed to complete (held calls are lost). After the agents hang up, their sets home to their secondary controller. 150 ACD hot desk agents are now available to take calls on the secondary controller.
Return-to-Service Conditions
You can convert a standard ACD system on a single controller to a resilient ACD configuration that consists of a gateway controller, a primary agent controller, and a secondary agent controller.
12
Failure scenarios
13
Phone Support The 5330 IP and 5340 IP Phones are the newest additions to the Mitel family of desktop IP phones. Both feature self-labeling keys (24 on the 5330, 48 on the 5340), a 160x320 graphic LCD display, full-duplex speakerphone and support for Mitel Call Control (MiNet) and SIP protocols. The 5330 IP and 5340 IP Phones support hot desking, resiliency, and the MiXML applications feature key, as well as Line Interface Modules. OPS Manager allows provisioning for 24 self-labeling keys. System Data Synchronization with the 3300 ICP With 3300 ICP System Data Synchronization (SDS) feature now synchronizes the data of resilient users and devices between primary and secondary controllers. After you set up data sharing between the primary and secondary controllers, the application automatically maintains the synchronization of user and device data. This feature keeps user and device data, such as call forward programming and phone settings, synchronized between the primary and secondary controllers regardless of whether the users make the changes while they are on their primary or secondary controller. OPS Manager Release 7.4 manages the replication of changes on the platform. When SDS is configured to share user and device data, OPS Manager ensures that resilient and non-resilient devices are synchronized and that network resources are not allocated to the unnecessary replication of data between home and secondary elements. SDS now propagates changes to name, directory number, department, location, privacy, device type, CESID data, and MAC address. SDS also distributes changes for resilient groups, networked hunt groups, and clustered pickup groups. With 3300 ICP Release 7.0 you can also use the Shared Data Updates form (on the platform) to managed out-of-sync data. Software Parity OPS Manager Release 7.4 software supports Mitel 3300 ICP Release 7.0 and SX-2000 LIGHTWARE 34.2.
NOTE:
Agent IDs are not manageable by OPS Manager. In OPS Manager user interfaces, ACD Hot Desk Agents appear as "ACD Agent". Traditional ACD Agents appear as "ACD Agent ID". In this Help system, the term "ACD Agent" indicates an ACD Agent with Hot Desking and Resiliency capability, and "ACD Agent ID" indicates a traditional ACD agent.
14
You can decide which users/devices in your system to make resilient. You may want to make all devices resilient or only those required for critical services. You can also select which 3300 ICP each device will fail over to. Any agents that are not on calls (that is, agents who are logged in but idle, in Make Busy, or on Work Timer) rehome to their secondary controller. Calls in progress at ACD agents on the primary controller are maintained (call survival). However, the agents are unable to access any features or dialing functions because the primary is out of service. After an agent ends the current call, the agents ACD set fails over to the secondary controller. As agent members fail over to the secondary, they begin to answer the calls that were requeued by the trunking gateway to the agent groups on the secondary controller. Eventually, all ACD agents fail over to the secondary controller and are available to answer calls to the agent groups. While on the secondary controller, agents have access to the same ACD features that they would have on their primary controller. During a failover or failback, the state of an agent -- for example, Make Busy, Do Not Disturb, and so forth -- is maintained. If an agents set is in Work Timer when the agents set fails over or fails back, the Work Timer is cancelled. During a failover, the agent group that is configured on the secondary performs the same functions as if it was still on the primary by
Servicing calls that are queued in the path Presenting calls to the agents Recording group statistics for the gateway controller as if the failure of the primary controller had not occurred (group statistics are recorded on the secondary controller but the statistics are only collected for calls that arrive on the secondary controller).
After the agent groups primary controller returns to service, the primary controller notifies the trunking gateway. Calls that are currently queued to the agent group on the secondary controller, but are not ringing a agent, are re-queued on the primary controller. Calls that are ringing an agent or are in progress are not affected. As agents on the secondary controller become idle, they fail back to the primary and begin answering the re-queued calls on the primary controller. When all the agents are back on the primary controller and all existing calls are queued on the primary, failback is complete. The ACD agent hot desking feature allows an agent to log into any ACD set and have the system apply the agents personal phone profile to that ACD set. After the agent logs in to the ACD set, the agent has access to his or her own personal speed calls, features, and phone settings. If you enable ACD agent hot desking in your call center, you do not have to provide a separate phone for an agents personal use. Instead, you can make a pool of shared phones available to many agents and any ACD set that the agent logs into also functions as the agents personal phone profile. NOTE: Only ACD hot desk agents can be resilient. Even if the agents do not require hot desking capability, they must be configured as ACD hot desk agents in order to be resilient. To convert a standard agent to an ACD hot desk agent, you set the agents User Type to "ACD Agent". When a resilient ACD hot desk agent fails over to the secondary controller or fails back to the primary controller, the states of the following ACD features are maintained:
ACD Login and Logout status Make Busy On/Off status (reason code is not maintained)
In addition, when a resilient ACD hot desk agent fails over to the secondary controller or fails back to the primary controller, the settings of features such as auto answer, do not disturb, and the configured value for the Work Timer are maintained.
15
NOTE: If an agents set is in Work Timer when the agents set fails over or fails back, the Work Timer is cancelled. The following conditions apply to ACD Resiliency:
Release 7.0 or later software is required on the gateway controllers, primary controllers, and secondary controllers to support ACD hot desk agents and ACD Resiliency. You must use either ACD hot desk agents or standard ACD agents. The use of ACD hot desk agents and standard ACD agents in the same ACD system, is not supported. The primary and secondary controllers must be clustered and connected to each other by IP trunks (fully meshed cluster). System Data Synchronization (SDS) must be sharing data between the primary and secondary controllers. You must use OPS Manager to initially configure resilient ACD agents on the primary and secondary controllers. You then use the System Administration Tool to configure resilient agent groups. After you have configured the agents and agent groups, SDS automatically keeps the required user and device data synchronized. Ensure that you assign the agent group members to the same secondary controller as the agent group (that is, both the agent group members and the agent group to which the members are assigned must have the same secondary controller). If you do not assign the agent group members to the same secondary controller as the agent group, the agents will not receive ACD calls during a failure of the primary controller. You cannot copy a resilient agent group from an agent group's secondary controller. When you click Copy, if the agent group's primary element is not local, you will receive an error message. You can only copy a resilient agent group from its primary controller. If the primary controller fails, any active Request Help calls, Silent Monitor sessions, or Record-aCalls that are in progress are dropped. Both the initial call and the Request Help call, Silent Monitor session, or Record-a-Call are dropped. Only ACD hot desk agents can be resilient. In the User Configuration form, you configure an agent as an ACD hot desk agent by setting or converting the User Type to "ACD Agent". Standard agents cannot be resilient. ACD hot desk agents can only log into ACD sets. ACD hot desk agents MUST NOT have "Hot Desk Login Accept" or "Hot Desk Remote Login Accept" enabled in its Class of Service. These COS options apply to standard hot desk users only. After a system reset, ACD hot desk agents are automatically logged back into the ACD system once the set re-establishes contact with the agent's primary or secondary controller. ACD hot desk agents are not logged out unless the set loses power or the agent explicitly logs out. The login and logout procedures for resilient agents are the same as those for standard agents. However, if a resilient agent logs in and the ACD set rehomes to the primary or secondary controller, there could be a delay of several seconds before login is achieved. If an ACD hot desk agent fails over to the secondary controller and then logs out of the agent session, the phone can take up to 45 seconds before it will return to service. After the phone returns to service, LOGGED OUT appears in the phone display. If a standard ACD agent logs out during a call, the system generates the report log immediately (that is, during the call). For resilient agents and ACD hot desk agents, the system does not generate the report until the call ends. In the Telephone Directory Editor and Telephone Directory Utilities applications of OPS Manager, ACD hot desk agents are identified as "Hot Desk or ACD Agent". In addition, the Mitel tab on the Directory Server application displays "Hot Desk" for both hot desk users and ACD hot desk agents. The Group Administration Tool identifies ACD hot desk agents as "Hot Desk" users. You cannot make ACD sets and agent IDs members of hunt groups.
16
You can make ACD hot desk agent IDs members of pickup groups. However, standard ACD agent IDs cannot be members of pickup groups. You can park calls against ACD hot desk agents and ACD sets (whether the ACD set is assigned to a standard ACD agent ID or an ACD hot desk agent). However, you cannot park calls against standard ACD agent IDs, ACD paths, or ACD agent groups. In the ACD real time events, the directory number and agent number for an ACD hot desk agent is recorded as the agent's ID.
While only 3300 ICPs, Release 4.0 or later can be resilient controllers, resilient clusters can contain pre-4.0 3300 ICPs and Mitel legacy SX-2000 PBXs. These other devices cannot function as secondary controllers, but they can be part of a resilient solution as boundary nodes and transit nodes:
Boundary node Release 4.0 or later 3300 ICPs and SX-2000 PBXs with LIGHTWARE 32, Release 1.1 or later can function as boundary nodes, actively routing calls between resilient 3300 ICPs and non-resilient (preRelease 4.0) 3300 ICPs. Transit node All cluster elements can function as transit nodes, passively allowing calls to be routed through them, on their way to a resilient ICP.
17
In OPS Manager user interfaces, ACD Hot Desk Agents appear as "ACD Agent". Traditional ACD Agents appear as "ACD Agent ID". In this Help system, the term "ACD Agent" indicates an ACD Agent with Hot Desking and Resiliency capability, and "ACD Agent ID" indicates a traditional ACD agent. Converting an ACD Agent ID to a ACD Agent When you create an ACD Agent, OPS Manager MAC checks for existing Agent IDs on the home element as well as on the secondary element. If any exist, you are prompted to confirm the conversion process. You may select either "Yes", "No", or "Cancel". If you select either "No" or "Cancel", the MAC cancels the operation and returns you to the MAC workform to make any necessary changes. If you answer Yes, MAC completes the conversion process. NOTE: Scheduled MACs automatically convert ACD Agent IDs to ACD Agents without prompting you to confirm the conversion. The Agent ID may already have one or more corresponding entries in the Telephone Directory. Only one can be the Prime Name. MAC does not allow you to add duplicate names or numbers. However, MAC does allows you to add an entry with a new name in the Telephone Directory as part of the conversion process. Existing entries remain intact. If you wish to reuse an existing ACD Agent name, you must:
1. 2. 3. 4. 5. Take note of fields such as Name, Private, Department, Location and Distribution List Delete the entry in the Telephone Directory in OPS Manager Synchronize the Telephone Directory in order to update the host of the Agent ID Copy the values previously noted in the MAC Workform Commit MAC changes to complete the conversion
Conditions
18
19
the 3300 ICP that is sharing ACD group data ensures that the group resilient pairs are up-to-date and that the agent and groups share the same resilient pair. Only ACD Agents can be configured as resilient. For additional information about the ACD Agent feature and conditions. In the Telephone Directory Editor and Telephone Directory Utilities applications of OPS Manager, ACD Agents and Hot Desks are identified "Hot Desk or ACD Agent". In addition, the Mitel tab on the Directory Server application displays "Hot Desk" for both Hot Desk users and ACD Agents. You cannot make ACD sets and agent IDs members of hunt groups. You can make ACD Agent IDs members of pickup groups. However, traditional ACD agent IDs cannot be members of pickup groups. Selecting an ACD Agent for a 3300 ICP pre-Release 7.0 or SX-2000 systems causes MAC functions to fail.
20
Set to Resilient or Not Resilient. Set to Resilient or Not Resilient. Set to Resilient or Not Resilient. Set to Resilient or Not Resilient. Set to Resilient or Not Resilient.
21
Table 2 Resiliency policy settings Resiliency policy Call Forwarding Resiliency options Key Assignment Resiliency options Key Assignment Setting Call Forwarding is not resilient. Set to Resilient or Not Resilient. The Resiliency setting applies to all keys that can be resilient (see the Resilient Keys table below). TIP: If you have set the Enable Voice Messaging field to Resilient in the Voice Messaging section of the Resiliency Policy form, you should make key assignments resilient to support the use of speed dial keys for accessing voice mail. NOTE:
If the user has key resiliency enabled and the administrator disables it, the keys will not be provisioned or synchronized as resilient from that point on. The keys will not be deleted from the secondary element, however, until they are deleted from the primary element. While Single Line, Multicall, and System keys can be made resilient, unlike other Resilient keys, they will not be automatically programmed on a secondary element when the Key Assignment field is set to Resilient in the Resiliency Policy for a MAC operation. You must make these keys resilient, using the Telephone Directory Editor application.
22
The following table indicates the keys that will work or partially work on the secondary element.
Table 3: Resilient Keys table Resilient Keys Key Type Auto Answer Callback Camp-on Direct Page Headset Operation Intrude Meet Me Answer Non-networked group page will not work if the set attempting the MMA is hosted on a different switch from the paging party. Networked group pagepage group can only be programmed on one switch. MMA will only work from sets still homed to that switch. At times, callbacks will fail to mature properly because of the two parties is on the wrong switch. Key-Behavior Comments
Music Night Answer Override Paging Privacy Release Call Pickup Release Speed Dial Swap Single Flash Double Flash Single Line, Multicall, System Swap Message Waiting Indicator Emergency Call Account Code Verified Account Code Non-Verified Tag Call Must use the Telephone Directory Editor to make resilient. May change Day/Night service on wrong system.
23
System Data Synchronization The System Data Synchronization (SDS) feature now synchronizes the data of resilient users and devices between primary and secondary controllers. After you set up data sharing between the primary and secondary controllers, the application automatically maintains the synchronization of user and device data. This feature keeps user and device data, such as feature keys and phone settings, synchronized between the primary and secondary controller regardless of whether the users make the changes while they are on their primary or secondary controller. You must enable the System Data Synchronization (SDS) feature to share data in the cluster. For ACD resiliency, you must configure SDS to share the following forms:
Class of Service Class of Restriction Feature Access Codes Independent Account Code Definition
24
Note that SDS only distributes the Remote Directory Number (RDN) information of the agent groups to the resilient secondary controller. OPS Manager distributes the RDN information of the agent groups to the other network elements in the cluster if the agent groups are programmed in the Telephone Directory. Refer to System Data Synchronization in the Features book of the System Administration Tool online help for instructions. T1/E1 trunk resiliency The new T1/E1 Combo MMC supports T1/E1 trunk resiliency. If a site's primary controller fails, this feature automatically transfers the T1/E1 trunks to a T1/E1 Module on the secondary controller. Portable Directory Number Resiliency is primarily an IP-enabled capability that builds on existing 3300 ICP Voice Clustering (Portable Directory Number [PDN]), and call-routing principles. Existing clustering techniques are used to set up a cluster, which is then made resilient through the programming of boundary nodes, transit nodes, and call routing ARS. Hot Desking (Resilient Clustered) The 3300 ICP clustered hot desking feature supports resiliency. It is recommended that all clustered hot desk users and devices be provisioned to be resilient. (Nodal Hot Desking does not support resiliency because resiliency requires clustering.) For more information about clustered hot desking, refer to the 3300 ICP System Administration Tool Online Help. As part of the Resiliency solution, resilient clustered hot desking offers device, call, voice mail, and feature resiliency, as well as the following features related specifically to resiliency in hot desking:
Hot desk user-profile resiliency Persistent login Hot desk login, logout, and remote logout on the primary and secondary controller. With resilient clustered hot desking, a hot desk user can be provisioned to be resilient and to support persistent login. Persistent login means that an active hot desk login survives during failover and failback. Resilient hot desk users can log in and log out of a hot desk session as long as they are in service on either their primary or secondary controller and as long as the hot desk device that they are using is in service on either a primary or secondary controller.
You program ACD hot desk agents with resiliency from OPS Manager. Then, you must perform a Telephone Directory Synchronization to distribute the resilient ACD hot desk agents to the other network elements. After you have programmed ACD resiliency, the System Data Synchronization feature keeps the user and device data, such Make Busy status, and the Do Not Disturb settings, synchronized with the data on the secondary. For an agent group to be fully resilient, all its member agents must be configured as ACD resilient hot desk agents and they must have the same primary and secondary controller as the resilient agent group.
25
Fail-over States
Fail-over is initiated when a resilient device loses connectivity to its current ICP (primary or secondary). The device progresses through the following states of resilient operation:
Early warn state The early warn state occurs when a phone detects the possible loss of its link to its current ICP. A phone can recover from this state and resume normal operation on the ICP if a failure is averted. Fail state Fail state follows early warn state if an ICP or network failure is not averted. Once in fail state, a device will Fail-over to its alternate ICP. Homing/Registration Once a failure has occurred, a resilient device homes to and registers on its alternate ICP. If it is unable to do so, the device goes out of service. This state occurs during all Fail-over, Fail-back, and hand-off events, and it occurs only when a phone is idle (see Idle Devices on page 271). A phone does not enter this state until any resilient calls on it are ended. Operation on alternate ICP Once a device finishes registering on its alternate ICP (primary or secondary), it is in service on that ICP. NOTE: Registration is transparent to an end user.
Failback/Handoff States
When a resilient device is in service on its secondary ICP and the ICP determines that the primary ICP of the device has regained healthy status, the secondary ICP initiates a Fail-back or handoff of the device to its primary ICP. When a device fails back or is handed off to its primary ICP, it progresses through three main states of resilient operation:
Fail-back/Handoff Initiation The secondary (and current) ICP of a resilient device initiates the Fail-back/handoff process. Homing/Registration The device rehomes to and registers on its primary ICP. Normal operation on primary ICP Once the device finishes registering on its primary ICP, it is in service on that ICP.
While the calls are being redirected, a situation can occur where the gateway attempts to queue calls on the secondary controller before the first agent becomes available in the resilient agent group on the secondary. In this case, the agents have not arrived on the secondary because they are engaged in existing calls (call survival) or they are in the process of failing over and registering on their secondary controller. Agents dont begin to fail over until their current calls are completed. If the gateway attempts to queue calls on the secondary controller before the first agent becomes available in the resilient agent group, the gateway views the groups on the secondary as having no agents logged in during this transition period. You have two options for handling the redirection of the queued calls. NOTE: The same situation can occur when the primary controller returns to service and calls once again commence queuing at the primary before the first agent has failed back and registered in the agent group on the primary controller.
26
Option 1 In the Agent Group Assignment form, leave the option "Queue calls when no local agents are logged in" set to "No" for the resilient agent groups. Then, any calls that are redirected to the secondary controller prior to the first agent failing over are not queued to the agent groups because there are no agents logged in. Thus, calls that are queued to the path on the gateway before the primary controller failed:
First follow Overflow programming when they attempt to queue on the secondary when no agents are logged in, overflow immediately Then follow Interflow handling when the Interflow Timer expires
All new calls that attempt to queue to the path on the gateway after primary agent controller fails but before the first agent is available on the secondary controller:
First follow overflow programming and overflow immediately without queueing to the group Then follow Path Unavailable Answer Point handling
Option 2 Set the "Queue Callers to Group When No Local Agents are Logged In" option to "Yes" in the Agent Group Assignment form of the resilient agent groups. Then, if the gateway redirects calls to the resilient agent group on the secondary before the first agent has registered, the calls are queued against the resilient agent group. Be aware that if you enable this option, calls will always queue to the agent groups on either the primary or secondary controller when no agents are logged into the group. After all agents log out of an agent group at the end of day, the calls will not be routed to the Path Unavailable Answer Point. However, you can still achieve Path Unavailable Answer Point routing by manually putting the Path DN in Do Not Disturb at the end of the day. You can also use Option 2 if you do not require the Path Unavailable Answer Point to automatically route calls to a "Business is closed for the day" greeting. Instead, you could use the Mitel Intelligent Queue solution to act as a front end with Time of Day routing to the greeting.
27
Table 4 3300 ICP Enterprise Node vs. Gateways 3300 ICP Enterprise Node Queuing Gateway Agent/Extension Gateway NOTE:
To configure ACD Agents to be resilient, you must first configure the agents as ACD Hot Desk Agents (that is, not as ACD Agent IDs). Queues are configured on trunking gateways (3300 ICP Enterprise Nodes or 3300 ICP Queuing Gateways) to direct calls to the resilient ACD Agent groups you configured on both the primary and secondary controllers. The 5330 IP and 5340 IP desk phones support hot desking and resiliency.
28
ACD Resiliency depends on the concept of clusters. Figure 5 illustrates the relationship of the telephone systems (voice media servers) at three offices (sites) in two clusters. The following points explain attributes of media servers in a resilient environment:
Clusters vs. sites Clusters can comprise one or more sites, or a subset of media servers at a site. You can also have more than one cluster at a site. In the example, the main office in New York and the Boston branch are both in CLUSTER 1. Chicago comprises CLUSTER 2. Media Servers in different Clusters cannot be resilient with each other Telephone systems in different clusters cannot be resilient with each other. In the example, the 3300 ICP Enterprise Node in Chicago belongs to CLUSTER 2, which means media servers from CLUSTER 1 cannot failover to it. Therefore, the media server in Chicago is not resilient. However, the 3300 ICP Enterprise Node could operate as a resilient media server if it were in CLUSTER 1. Media servers resilient with each other must belong to the same cluster In a resilient environment only telephone systems (voice media servers) within the same cluster can be resilient with each other. In the example, the 3300 ICP Primary Queueing Gateway, 3300 ICP Secondary Queuing Gateway, 3300 ICP Primary Agent/Extension Gateway, and 3300 ICP Secondary Gateway are resilient media servers. They all belong to CLUSTER 1. Agent/Extension Gateway cannot failover to a Queuing Gateway In the event of an outage, a Agent/Extension Gateway can failover to another Agent/Extension Gateway or an Enterprise Node. Agent/Extension Gateways cannot failover to Queuing Gateways because you cannot configure agents, agent groups, extensions, or extension groups on Queuing Gateways. In the example, the 3300 ICP Primary Agent/Extension Gateway fails-over to the 3300 ICP Secondary Agent/Extension Gateway. Any type of telephone system can belong to a cluster All telephone systems within the same cluster need not be resilient. In the example, the SX-2000 is not resilient, yet it belongs to CLUSTER 1. A non-resilient telephone system need not belong to a cluster Telephone systems may or may not belong to a resiliency cluster. In the example, the SX-2000 belongs to CLUSTER 1 by virtue of the cluster name assigned to the SX-2000 in YourSite. It is not resilient so it could also be configured to belong no cluster at all. Device names and numbers must be unique in a cluster All devices, such as queues, agents, and extensions, belonging to the same cluster must have unique names and reporting numbers. In the example, Queue 1000 is associated with the 3300 ICP Primary Queuing Gateway in New York. Queue 1000 cannot be associated with any other media server in CLUSTER1. Although the SX-2000 is not resilient with the other telephone systems in New York, you cannot create and assign another Queue 1000 to it. Conversely, Queue 2000 is associated with the 3300 ICP Primary Agent/Extension Gateway in Boston, yet in Chicago another Queue 2000 is associated with the 3300 ICP Enterprise Node. This is possible because the telephone systems belong to separate clusters.
NOTE: Each Mitel telephone system has a unique cluster ID. The cluster ID is not the same as the Cluster Name. A Mitel telephone system will have a cluster ID whether or not it is in a resilient environment. However, a telephone system must be assigned a cluster name if it is resilient.
Configuration overview
29
Configuration overview
For reporting purposes you must configure the Contact Center Management YourSite database to mirror the telephone system. You configure your sites and telephone systems (voice media servers) in YourSite Enterprise. You configure other devices and device groups, (including queues, queue groups, employees, agents, agent groups, and extensions), in YourSite Configuration. The following devices have specific configuration requirements in order to support resiliency. See Figure 6.
Media servers Employees Agent IDs Agent groups Extensions
30
31
4. 5. 6. 7.
Adding sites
When you add a site, you add the name of the site, the time zone, and smart algorithms (optional). By making each site distinct, you can restrict user access to that site. To add a site
1. 2. Click YourSite=>Enterprise. Click Add site. Configure the SMTP Mail Server Configure smart algorithms for sites
32
If you have ACD Resiliency, after This media server functions as a, select the functionality. Otherwise, skip to step 9. 7. After This media server will failover to, select the failover media server. 8. If you must create a failover media server, click Add media server and create a failover media server. 9. If you have hot desking agents or ACD Resiliency, under All ACD PhoneSet Manager and ACD Softphone agents associated with this media server are, click ACD hot desking agents. 10. Specify the data collection settings. See Configuring data collection settings for a 3300 ICP on page 32.
6.
Adding employees
33
Adding employees
You must add all contact center employees (agents, supervisors, and managers) to the YourSite database so you can run reports on employee activities, view employee performance in real time, and use Contact Center Management. Each employee can have one employee ID only. You must add agent login IDs to the YourSite database for employees you want to track in real time and reporting. When you add an employee, you can simultaneously create an agent login ID for the employee on the Create associated devices tab, provided you want the agent login ID to be the same as the employee ID. NOTE: You must assign each employee a unique ID number. If an employee needs to access the Contact Center Management website at more than one enterprise site, then you must assign the employee a unique employee ID for each site.
34
6.
7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21.
Adding employees
35
9. 10.
11.
12. 13.
36
Creating associated devices You can associate agent IDs, extension numbers, and account codes to employees. After you create associated devices, set up credentials for the Contact Center Management user. See Configuring employee user settings on page 37.
Creating agent login IDs for employees
To create login IDs for each media server that the employee will log into
1. 2. In YourSite=>Configuration, Employee=>Employee, on the Add/Edit employee window, click the Create associated devices tab. Under Automatic agent login ID creation, select the check boxes of the media servers to which the employee will log in.
Adding employees
37
5. 6. 7. 8.
38
Adding agents
You must add agent login IDs to the YourSite database for employees you want to track in real time and reporting. An employee can have more than one agent login ID but only one employee ID. You assign multiple agent login IDs to employees who are cross-trained to handle calls for different departments (different queues). For example, Carol Jones handles calls for English, French, and Spanish queues. In the YourSite database, you assign Carol three agent login IDs, one for each queue that she handles. When you add an agent login ID, you must associate it with an employee so you can identify the employee.
5.
Adding agents
39
40
Adding agents
41
Adding agent groups manually To add an agent group to the YourSite database
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Click YourSite=>Configuration. Click Agent login=>Agent group. The Agent group window appears. Click Add. See Figure 12. After Name, type the agent group name. After Reporting number, type the agent group reporting number listed on the telephone system assignment forms. After Agent group ID, type the agent group ID number listed on the telephone system assignment forms. After Site, select the site of the agent group. After Media server, select the media server associated with the group. If the agent group is resilient, after Failover media server, select the failover media server associated with the group. If you want all agents associated with the media server assigned to this agent group to belong to this agent group, select the Automatically add all agents associated with this media server to this agent group check box. Click Save.
11.
42
4.
The Edit agent group members window appears. Under Available agent login IDs, select the check boxes of the members you want to add to the group and click Add >>.
Adding extensions
Using Quick Setup to add extensions
To add a series of agent groups, extensions, trunks, DNIS, or voice mail extensions to the YourSite database
Click YourSite=>Configuration. On the Configuration menu, click the device to which you want to add a series. The device window appears. 3. Click Quick Setup. The Quick Setup window appears. 4. After Associate the entries to this site, select a site. 5. After Associate the entries to this media server, select the media server. 6. If the agents are resilient, you can optionally select the failover media server for the agents. 7. In the Active number from and Active number to boxes, enter a range of reporting numbers. 8. Optionally, create employee IDs for extensions. 9. Click the General tab. 10. Under Advanced options, If the extension is resilient, click This extension is resilient. If the extension will be costed with Mitel Call Accounting and you are not licensed for Call Accounting resiliency, click This extension is costed. If you are licensed for Call Accounting resiliency and the extension will be resilient and costed, click This extension is resilient and costed. To apply the advanced option to all extensions associated with the media server, select the Apply this setting to all extensions associated with this media server check box. 11. If you want to add a prefix or post fix to the device reporting number/ID number, select the Append a prefix and/or a suffix to the reporting/ID number check box and type the prefix and/or suffix. 12. Click Run. YourSite populates the database with the agent groups you created. 1. 2.
5. 6. 7. 8.
Adding queues
43
If the extension is resilient, after Failover media server, select the failover media server associated with the group. 10. Under Advanced options, If the extension is resilient, click This extension is resilient. If the extension will be costed with Mitel Call Accounting and you are not licensed for Call Accounting resiliency, click This extension is costed. If you are licensed for Call Accounting resiliency and the extension will be resilient and costed, click This extension is resilient and costed. To apply the advanced option to all extensions associated with the media server, select the Apply this setting to all extensions associated with this media server check box. 11. Click Save.
9.
Adding queues
You must add queues to the YourSite database so you can run reports on queue activities and view queue performance in real time. When you add a queue to the YourSite database, the configuration options available depend upon the media server you select. When you add a queue to the YourSite database, you associate the queue with the agent groups that handle calls for the queue. You do this so you know which agent groups handle which queues. You must associate at least one agent group with a queue in order to produce reports on the queue. When you add a queue to the YourSite database, the configuration options available to you depend upon the media server you select. For queue programming purposes, media servers fall under five general categories: voice (a telephone system with or without MiTAI enabled), Intelligent Queue, email, chat, and fax. If you purchased the Interactive Visual Queue add-on application, the Interactive Visual Queue tab appears on the Add/Edit queue windows.
44
Adding queues
45
11.
12.
13.
46
Configuring queue spectrum settings To configure the reporting intervals for the abandon, answer, interflow, and handle statistics for the queue
Click the Queue spectrum tab. See Figure 16. 2. Select and edit the time intervals. When you run spectrum reports the statistics will be spread across the time intervals selected. 3. Optionally select the Apply spectrums to all queues associated with this media server check box. For example, if you selected the SX-2000 with MiTAI media server for the queue, enabling this option applies these spectrum values to all of the queues associated with the SX-2000 with MiTAI. 4. Specify the business hours for the queue. See Configuring business hours for a queue on page 46. 1.
Configuring business hours for a queue To configure business hours for a queue
Click the Business hours tab. See Figure 17. 2. After Business-hour schedule for this queue, select a schedule from the list. 3. If you must create or modify a schedule for the queue, click Manage schedule and configure a schedule. For voice queues, this schedule specifies when there should be no real-time data filed to the SQL database. For chat and email queues, the business-hours schedule specifies the time of day customers receive the business-hours auto-acknowledgment message. 4. If you want to apply this schedule to all of the queues with the same media server, select the Apply this business-hour schedule to all queues associated with this media server check box. 5. If the queue is a voice queue and you want to exclude statistics generated for calls received or dialed outside of business hours, select the Generate real-time statistics and reports only within the business hours selected check box. Enabling this option prevents data from being filed to the SQL database outside of business hours. 6. If the queue is a voice queue, click Save. Otherwise continue with step 10. 7. If you will use Interactive Visual Queue, specify Interactive Visual Queue options. 8. If the queue is a voice queue using a media server with MiTAI, specify Interactive Contact Center queue control. 9. If the queue is an email queue, specify Multimedia Contact Center email options. 10. If the queue is a chat queue, specify Multimedia Contact Center chat routing options. 11. If the queue is a fax queue, specify Multimedia Contact Center fax options. 1.
Adding queues
47
48