Known Issues and Recommendations
Voice Platform Resource Manager
The Known Issues and Recommendations section is a cumulative list for all 8.5.x releases of Voice Platform Resource Manager. This section provides the latest information on known issues and recommendations associated with this product. It includes information on when individual items were found and, if applicable, corrected. The Corrections and Modifications section for each release may list additional issues that were corrected without first being documented as Known Issues.
See also: Internationalization Issues
The IP ReadMe for this component states incorrectly that GVP 8.5.0 supports 32-bit RHEL 6. The IP ReadMe will be corrected to remove this incorrect information in a future release.
Resource Manager currently does not support OPM parameters that have key name in capital letters at a solution level. Workaround: Use key names with lower case letters only.
| ID: GVP-18808 | Found In: 8.1.700.61 | Fixed In: |
The Resource Manager MIB data (scalar table) may not reflect the correct number when running in an Active-Active configuration.
| ID: GVP-20257 | Found In: 8.1.700.61 | Fixed In: |
In a Resource Manager Active-Active cluster, during the brief period when a node with higher priority comes online, one of the RMs may respond to OPTIONS pings with a 503. The result: calls are directed to that RM for a brief period of time.
| ID: GVP-19668 / ER# 323890741 | Found In: 8.1.700.61 | Fixed In: |
Virtual IP (VIP) Address Takeover for Windows is less reliable than a Windows NLB cluster configuration.
Adding or removing an IP address using VIP Address Takeover is more complicated than enabling or disabling a port in the Windows NLB configuration. VIP Address Takeover fails when either command fails on the backup or the primary. By comparison, in the Windows NLB configuration there is no dependency on the backup command to succeed for most of the failover scenarios.
It takes longer for Netsh (the Microsoft utility that is used for IP Address Takeover) to modify the network configuration than it takes to enable or disable a port in the NLB configuration. In addition, the time may depend on a particular NIC and its configuration. Normally, modification takes less than 15 seconds to execute, but in this situation it can take as long as 30-45 seconds.
Note: The Netsh command can fail if the NIC configuration is already being accessed through the Network Properties User Interface.
InterfaceForARPing
The InterfaceForArping requires the user to specify the correct device for the virtual interface. Follow these steps to get this information from the registry:
- Start Regedit and go to the HKLM\SYSTEM\CurrentControlSet\Control\Network\ directory.
- Identify the Key with the value {Default} and the data Network Adapters.
- If the virtual interface is set for Local Area Connection, then search the listed adapter (in the registry) for the value name that contains the data Local Area Connection. The Key that contains Local Area Connection is the reference to the physical device identifier.
- Pre-append \Device\NPF_ to the Key and set this value for InterfaceForArping.
Example: \Device\NPF_{85FEBE1C-9EEF-4E61-974B-1158DB270F6E} From this key: HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Network\{4D36E972-E325-11CE-BFC1-08002BE10318}\{85FEBE1C-9EEF-4E61-974B-1158DB270F6E}
Virtual IP Interface Takeover Scripts Find these scripts on your Installation CD or in the IP 8.1.502.87 package that you downloaded over the internet:
Ping.vbs
- Ping host and return 1 if ping successful, 0 if not, -1 if target not specified.
Check_ip.vbs
- Check if IP address (arg0 ) can be found on host (arg1 )
- Return 1 if found, 0 if not, -1 if address and host not specified
INIT_IPTakeOver.bat
- Optional. You can manually execute this batch script, to disable the Virtual IP interface in the RM box before starting the RM process.
IPTakeOver.bat
- Enable or disable the Virtual IP interface in the RM box, during RM's own internal election process or when a failover event occurs.
| ID: ER# 298588431 | Found In: Unspecified | Fixed In: |
Resource Manager crashes in a race condition while processing CANCEL . The race condition is caused by SIP Server sending CANCEL after a Media Control Platform crash.
Found In: Fixed In:
| ID: ER# 305137359 | Found In: 8.1.502.76 | Fixed In: |
In active-standby HA mode for RM, if the DataModule cluster initialization is delayed for any reason, the election message processing can be affected and both RMs try to become active. In this scenario, you can check for one or two things:
A possible network issue. If the Reporting client is timing out while trying to connect to the Reporting Server, and therefore affecting initialization.
| ID: ER# 275029916 | Found In: 8.1.502.76 | Fixed In: |
Resource Manager could still route calls to an unassigned GVP Configuration Unit (CU) because Configuration Server does not dynamically update the deletion of the gvp.resources section in that CU if that is the last section to be deleted. To work around this, manually create a dummy section in the CU, with a dummy parameter, so that Configuration Server will always send notifications of deletions when the gvp.resources section is deleted from the CU.
| ID: ER# 242188298 | Found In: 8.1.201.93 | Fixed In: |
Resource Manager (RM) continues to accept announcement calls when the announcement-allowed option is set to false in the IVR Profile. This issue occurs under the following conditions:
Calls comes to RM from a Gateway resource that has CTI Usage set to Based on DN Lookup in the logical group it belongs to.
The mapped IVR Profile has CTI Allowed set to CTI using IVR Server is not allowed.
| ID: ER# 230809547 | Found In: 8.1.101.33 | Fixed In: |
In RM High Availability (HA) configuration, calls are not processed in the first 15 seconds when an HA pair is started simultaneously. To workaround this issue, start RM instances in a safe interval of 15–30 seconds apart, or if started simultaneously, allow some startup time (approximately 15 seconds) for the RM algorithm to elect the active node before taking live traffic.
| ID: ER# 214740001 | Found In: 8.1.001.68 | Fixed In: |
When a SIP resource sends a redirection response (301 or 302) back to RM, if the SIP resource returns multiple SIP destinations to redirect to, RM takes the first of these SIP destinations to redirect the request to.
| ID: ER# 193841929 | Found In: 8.0.004.66 | Fixed In: |
Resource Manager, in an active-active or active-standby HA deployment, terminates in a race condition when a CANCEL from SIP Server and responses for INVITE from the resources managed by RM occur at the same time.
| ID: GVP-21484 | Found In: 8.1.700.68 | Fixed In: |
Internationalization Issues
Information in this section is included for international customers.
There are no internationalization issues for this product.