NMX 5.5.4.

0 Release Notes

NMX Digital Service Manager™

Notes for Release 5.5.4.0.127

Harmonic Inc. U.S.A.

4300 North First Street Release 5.5.4.0.127
San Jose, CA 95134 Revision 1
+1.800.286.1600
+1.408.944.6700
FAX +1.408.944.6705 October 11, 2010
http://www.harmonicinc.com/

This document contains information that is proprietary and confidential to Harmonic Inc. and is intended for the specific use
of the recipient for the express purpose of operating the Harmonic broadcast encoding system. This document is provided to
the recipient with the expressed understanding that the recipient will not divulge its contents to other parties or otherwise
misappropriate the information contained herein.

NMX Digital Service Manager, ProStream, Mentor, DiviTrackXE, DiviTrackMX, DiviTrackIP are all trademarks of Harmonic
Inc.

Copyright 2010 Harmonic Inc. All rights reserved. All information subject to change without notice.

Page 1
 NMX 5.5.4.0 Release Notes

Table of Contents
1 Introduction ............................................................................................................................... 3
1.1 Architecture Overview ............................................................................................................... 3
1.2 New Capabilities and Features.................................................................................................. 3
1.3 Release Details ......................................................................................................................... 5
1.4 NMX Network Security .............................................................................................................. 5
1.5 Accessing Online Help .............................................................................................................. 6
1.6 Viewing Manuals in PDF ........................................................................................................... 6
2 Issues and Limitations .............................................................................................................. 7
2.1 Resolved Issues from Previous Releases ................................................................................. 7
2.1.1 NMX Problem Reports Resolved in Release 5.5.0.0 ........................................................ 7
2.2 Known Issues in this Release.................................................................................................... 9
2.2.1 3rd Party Devices .............................................................................................................. 9
2.2.2 Automation Server and Automation Admin GUI ............................................................. 10
2.2.3 Element Managers .......................................................................................................... 10
2.2.4 Graphical User Interface (GUI) ....................................................................................... 12
2.2.5 Installation ....................................................................................................................... 13
2.2.6 Stream Editor .................................................................................................................. 13
2.2.7 Redundancy .................................................................................................................... 15
2.2.8 Performance ................................................................................................................... 16
2.2.9 Provisioning .................................................................................................................... 16
2.3 Legacy Known Issues in NMX Released Versions ................................................................. 19
rd
2.3.1 3 Party Devices – CherryPicker .................................................................................... 19
2.3.2 Third-Party Devices – Other devices .............................................................................. 22
2.3.3 Automation Server and Automation Admin GUI ............................................................. 22
2.3.4 Element Managers .......................................................................................................... 22
2.3.5 Graphical User Interface (GUI) ....................................................................................... 23
2.3.6 Installation ....................................................................................................................... 24
2.3.7 NMX Server Data Base................................................................................................... 24
2.3.8 VOD Statistics ................................................................................................................. 25
2.3.9 Stream Editor .................................................................................................................. 25
2.3.10 Redundancy .................................................................................................................... 26
2.3.11 Performance ................................................................................................................... 27
2.3.12 Provisioning .................................................................................................................... 27

Page 2
 NMX 5.5.4.0 Release Notes

1 Introduction
NMX Digital Service Manager™ is a suite of software that provides configuration, control,
monitoring, and fault management in Harmonic open digital compression systems. NMX supports
geographically diverse broadcast sites through client/server architecture, and is highly scalable in
its ability to support small networks with only a few devices as well as larger networks with a
greater number of managed devices.
This document contains only information specific to open issues for the NMX 5.5.4.0.127 release
and problems fixed since the NMX 5.4.x releases. Refer to the NMX Digital Service Manager™
Installation and Startup guide for installation information, the NMX Digital Service Manager
Getting Started guide for general configuration of your NMX system, and the NMX Online Help for
configuration and operational information.

1.1 Architecture Overview

A complete description of the NMX architectural components and functionality can be found in the
various documents provided with your purchase of NMX Digital Service Manager™. These are
contained within the NMX software package

1.2 New Capabilities and Features

NMX 5.5.4.0.127 is a maintenance patch release see section 2.1.1 for list of issues
resolved
Please also read updated release notes for the ProStream 1000 and Europa Encoders for a list
of issues resolved on these products.

NMX 5.5.3.0.116 is a maintenance patch release

 NMX fixes as listed below
 Electra fixes, see Electra Encoder release notes for full list of issues resolved. Below are
two major issues that are resolved in this patch:
 PR 83935 - Electra 7000 and Electra 5400: Fixed additional software counters in video
compression firmware to prevent service disruptions after 414 days of continuous
service.
 PR 78330- Electra 8000: Fixed software counters in video compression to prevent
service disruptions after 207 days of continuous service.
NMX 5.5.2.0.112 includes the following new features:

 Device Support
 PVR7000 limited device redundancy support – NMX able to manage PVR7000 devices
in Nx1, NxM and 1:1 redundancy configurations for ASI inputs
 Support for new features on the Ellipse 1000/2000 products. Please see release notes
for specific new features added to the product.
 Support for CAS-related alarms on NSG9000-6G

NMX 5.5.1.0.103 is a maintenance patch release

Page 3
 NMX 5.5.4.0 Release Notes

NMX 5.5.0.0.101 includes the following new features:

 NMX
 Compression improvements of the data between server and GUI functions including
connect services to network, switch state and link dialogs
 EAS Support:
 IGMP v3 Support – configuring ProStreams with Source Specific Multicast Address
for EAS Communication
 Service definitions – propagating EAS preferences from super server to proxy
servers and handling EAS definition changes
 Checkbox to enable Control Word synchronization between primary and backup
ProStream scramblers
 Edge-Based Descrambling Unit (EDU) Support
 Daemon that spools configuration through the control socket of the EDU
 Implemented single button to disable encryption on all services
 PID Based Filtering support for EDU

 Device Support:
 DSCP self marking – added additional QoS support for the IP output transports to help
identify traffic within the network
 Support LRV provisioning of Electra8000 device using the Ancillary Channel/Program
group in 96x96, 128x96 & 192x192 resolutions
 ACE VBR support – allow ACE streams in DiviTrack pool
 Third Party Device Support:
 Control of T-VIPS CP541 Seamless ASI A/B Switch Control

 Support for Electra 8000

Ancillary Video Encoding: support for encoding an ancillary video in the Microsoft PiP
formats

 Support for new features on the following Harmonic Products. Please see release notes for
the products listed below for full feature set:

 Support for new features on the ProStream 1000 product. Please see release notes for the
products listed below for full feature set:
 ACE – video transcoding:
 HD MP2 => HD MP2 – CBR, VBR
 HD H264 => HD MP2 – CBR, VBR
 Audio – pass-through
 EDU – Edge-Based Descrambling Unit
 AES Descrambler configuration
 Integrate with VMviewRight (Verimatrix) – Embedded VM Client that receives ECM
and convert it to CW
 Support Differentiated Services Code Point (DSCP)

Page 4
 NMX 5.5.4.0 Release Notes

 Socket redundancy from different ports (same GbE board).

 SNMP alarms forwarding.

1.3 Release Details

This NMX release includes:

 An NMX CD containing:
 NMX application software (Server, Client, Consolidated Alarm Viewer, Stream Editor
and Automation Server)
 Embedded software for Harmonic encoder and multiplexer products:
o DiviCom™ MV12, MV30, MV40, MV45, MV50, MV100, MV400, MV450, MV500,
and embedded software for the MN20 re-multiplexer.
 Embedded code for all Harmonic BNG (Broadcast Network Gateway) products
 Embedded code for Harmonic‟s ProStream 1000
 Embedded code for Harmonic‟s Proview 7000
 Embedded code for Harmonic‟s Ellipse 1000 / 2000
 Streamliner Cluster Workstation software
 Third party software: Adobe Acrobat Reader and MG-Soft MIB browser.
 Release notes for NMX and all embedded products.
 Integrated NMX online help for hardware and service configuration.
 The basic suite of Harmonic product hardware guides in PDF format
o Standard Definition DiviCom MPEG-2 Encoder user guide and Quick Reference
card
o High Definition DiviCom MPEG-2 Encoder user guide and Quick Reference card
o DiviCom Audio Encoder Guide
o MediaNode Remultiplexer Hardware Guide
o Broadcast Network Gateway Hardware Guide
o ProStream 1000 Hardware Users Guide
o NMX Digital Service Manager Installation and Startup guide
o NMX Digital Service Manager Getting Started guide
o ProView 7000 Quick Setup Guide
o Proview 7000 User Guide
o Ellipse Installation Guide
 Platform recovery media for the NMX platform included in your NMX order
 The following documents are available but provided with a separate part number (and
separate CD) from the NMX product CD.
o XML Service API (SAPI) Reference Guide
o EIS-MUXCONFIG Reference Guide
o EIS-MUXCONFIG SDK Guide

1.4 NMX Network Security

Harmonic recommends that users install network security patches from Microsoft as well as
installing virus detection software. NMX 5.5.4.0.127 includes all security patches provided by
Microsoft as of service pack 2 for Microsoft Windows 2003 operating system as well as service
pack 2 for Microsoft SQL Server 2008 Standard Edition. Security patches released by Microsoft

Page 5
 NMX 5.5.4.0 Release Notes

after these should be installed to provide optimum security for NMX. For a complete list of
available security patches visit the Microsoft web site (http://www.microsoft.com/windows).
These patches can be downloaded directly to your NMX server or client platform from this site.

1.5 Accessing Online Help

NMX online help files provide detailed information on how to use the NMX Digital Service
Manager. You can access the online help files by selecting Help from the menu bar. The help
files are located on the product CD and are automatically installed with the application.

1.6 Viewing Manuals in PDF

® ®
The software installation CD includes applicable product manuals as Adobe Acrobat PDF files.
During the software installation these files will be loaded on your system at C:/Program
Files/Harmonic/User_Manuals. You must have Acrobat Reader version 5.0.1 or greater to view
the PDF files. Acrobat Reader 7.0 is installed for you during the software installation if you select
this option.

Page 6
 NMX 5.5.4.0 Release Notes

2 Issues and Limitations

This section lists the improvements and significant issues resolved since release of NMX 5.4.x It
also describes issues that remain in the release of NMX 5.5.4.0.127 with suggested workarounds
in the event you encounter these issues in your Harmonic system.

2.1 Resolved Issues from Previous Releases

2.1.1 NMX Problem Reports Resolved in Release 5.5.0.0

Problem Report Description

Number / Release
61882 While changing power level on Sencore port property page user needs
to enter the power level value by multiple of 10. The default value of
the power level is -200 dBm. This means the actual default value for
device is -20dBm.
52491 SNMP Error alarm when AAA card is provisioned MPEG-2 HE AAC

82017 Unable to add FLEX cards to Electra 8100 in offline map.

82218 Bit rate for FLEX Data stream is wrongly changed to 5M by NMX if the
number entered by use is less than 5 in edit service properties

82321 NMX GUI doesn't save audio profile Dolby "Multichannel 3+2" on AHC
561
83089 NMX stat .pdf reports not showing up in c:\statistics\reports

80760 bNSG should extract video as "generic", not as the old specific types
(HD, SD, etc)
81090 When restarting the domain manager the Gbe interface changed from
RJ45 to SFP
83070 Offline/Online or Powercycle the backup ProStream 1000 in a 3G
system and output on primary ProStream 1000 goes to NULL PID
84114 Gateway does not unsubscribe the client when Domain Manger goes
5.5.1.0.103 down
84221 Split PMT regeneration from table regeneration (EIT)
5.5.2.0.112

84405
5.5.2.0.112 Spooler failed to spool EDU's after deleting 1 EDU from 1000

84475 ProStream 1000 ACE: Need to add "Wrong video format (SD/HD)"
5.5.2.0.112 alarm to NMX.

83934 Not seeing alarms from NSG9K6G

5.5.2.0.112

Page 7
 NMX 5.5.4.0 Release Notes

84596 While adding 1000 EDU's to spooler carousel bit rate jump to 90Mbps
5.5.2.0.112 for 30 seconds

85646 TxId="0" attribute can cause mismatch between the NMX configuration
5.5.3.0.116 and the device object HW IDs
86322 Get XML option on encoder Icons does not save the license
5.5.3.0.116 information
86325 Network Group Level option "IP Encapsulation Mode" is broken
5.5.3.0.116
86333 BMS set Pool bit rate but video normal bitrate does not update
5.5.3.0.116

83881 Domain crash after restoring a specific AS catalog

5.5.4.0.127
86815
5.5.4.0.127 NMX redundancy not failing over to backup NMX server properly
(with DPI Server)
87323
5.5.4.0.127 Program Redundancy Issues
87633
5.5.4.0.127 HD Pools cannot be edited within edit service configuration
spreadsheet view
87690
5.5.4.0.127 Upgrade from NMX 4.6 & 4.8.2 and Pools are not available in the
new map anymore.
87830 ProView 7000 Redundancy doesn‟t work.
5.5.4.0.127
85775 NMX redundancy can change the Gbe pro card properties on Electra
5.5.4.0.127 8000
88319 The alarm "Primary Pid missing" isn‟t available to be customized
5.5.4.0.127
88700 Nightly backup of catalogs is not working
5.5.4.0.127
88623 Enable Nightly Build in Tools - Options and get an error popup related
5.5.4.0.127 to DMS
88817 Require the capability of adding VBI circuits to HD FLEX services
5.5.4.0.127
89459 For FLEX services the AC3 parameter RF Modulation (AC3 Advanced
5.5.4.0.127 Tab) is checked for decode/encode audios. The parameter is grayed
out, so users can't change this setting.
88704 DPI Server Crash after upgrade
5.5.4.0.127
86618 Server goes to busy state and device not getting created while
5.5.4.0.127 creating Cue Trigger device
89818 Cannot receive the Switch thrown notifications thru SAPI
5.5.4.0.127

Page 8
 NMX 5.5.4.0 Release Notes

2.2 Known Issues in this Release

2.2.1 3rd Party Devices

PR 79915 - For Newtec Modulator 2277(ASI Only) changing Output Frequency and FEC-
rate could cause macro blocks, especially when Baseband processing is Stuffing.

Workaround: power cycle the device

PR 80229 When the Newtec Modulator (2280_ASI, 2280_IP, 2277_IP) modulation mode is
DVB S2, and DFL mode is manual, changes in the Baseband DFL value and Sync distance
value could cause loss of service and may not recover.

Workaround: power cycle the device to get service back.

PR 62001 – 3rd Party EM managed devices: device timeout value is less than expected
compare to Option settings
The timeout period for ICMP ping is not taken from user defined values in NMX options and is
hard coded to 3 retries with 1st timeout value as 100 ms, 2nd timeout value as 50 ms and 3rd
timeout value as 50 ms, making total timeout period in case of ICMP ping to 200 ms. ICMP ping
timed out values and retries count are same for all 3rd party devices.

PR 61983 -- Metrotek: standalone application does not work on windows 2000 server. It
works only on windows 2003. If NMX GUI tries to launch standalone GUI on 2000 server,
client will not be displayed.

PR 53865 -- PVR6000 IRD may not reply to NMX ping for a period of time (30s to 45s)
which will result in the device in timeout state
This may happen when NMX Domain manager restart. The workaround is to increase the ICMP
timeout value to 60s in NMX options --> change the "ICMP polling interval from 5s to 10s" under
"Switch/TCP" tab.

PR 40947 – Newtec 2277 and 2280 do not support coding rate 9/10, if FEC Frame Type
“Short” is selected for DVB S2.

PR 40950 - Newtec 2277 and 2280 device only supports "External 188" as MPEG framing
mode in DVB S2 Mode.

PR 43198 – "Launch Stand Alone Manager" menu may not be available

Entering the URL at any stage other than at first icon creation will require a GUI refresh to show
the "Launch stand alone manager" GUI menu item.

PR 43539 – Restart NMX or Re-status Backup PVR6000 is service affecting for couple of
seconds if the backup PVR6000 is carrying the service
This happens only when NMX is using serial port to push the configuration file into PVR6000 IRD.

PR 43474 - Newtec modulator s/w version v1.15 doesn't reply "SNMP error" if the value
sent by NMX is out of the range

Page 9
 NMX 5.5.4.0 Release Notes

PR 36932 - Sencore modulator ASM988A does not keep the data rate settings when power-
cycling the device.
If you Power cycle the Modulator the device will raise "loss of input sync" alarm and will not
maintain the original rate prior to loss of power. The workaround is to Change the symbol rate
and then change back to the desired rate.

2.2.2 Automation Server and Automation Admin GUI

PR 40227 – AS2 only supports Program Redundancy in Macro Mode not Service Plan Diff
mode
In order to automate Program redundancy, the operator will currently be required to use batches
in Macro mode.

PR 40657 – Alarm may not remit from Sencore modulator

"Loss of Input Sync" alarm may not be remitted when off-lining a Sencore Modulator. You can
work around this by removing the input cable, then on-lining the device and reconnecting the
cable. When you re-status the modulator, the device remits the alarm correctly.

2.2.3 Element Managers

PR 80238 - FLEX Input Extraction does not display all unknown type of pids as data pids
and shows extracted descriptor information.

Data pids corresponding to ancillary services typically come and go in transports - NMX does not
display 0 bps streams. To address this, device will report a minimum bitrate of 25 Kbps for data
pids to NMX - this minimum is reported ONLY if the pid is not found to be flowing during
extraction.

For all pass thru streams in general and data pass thru specifically, the user must check the
bitrate returned by input extraction, compare with the maximum bitrate expected to be seen for
the stream, and provision the stream with the larger of the two.

PR 74173 -- "HTTP Error: Provisioning parse failure" stuck alarm raised when setting
H.264 encoding (and changing back to MPEG-2) on Electra 5000 MPEG-2

Workaround: Re-status the encoder (Right click on the encoder icon) or deactivate
/activate service configuration will remit this alarm.
PR 76604 -- Starting the Domain Manager for the first time with NMX 5.2.2 may cause
overflow in bandwidth at the ASI out of the ProStream for a few seconds

This may happen when starting NMX server the first time after upgrading. This condition will
recovery in a few seconds.

PR 67719 – In rare cases, encoder may raise an alarm "Duplicate output pid 0". Once the
alarm is asserted it will not remit.

Workaround: Restart Domain Manager will remit this alarm.

Page 10
 NMX 5.5.4.0 Release Notes

PR59630 -- The input discovery done by the Flex encoder will display the audio PIDs in
random order, not sorted based on the PID value

PR60199 -- Change the FLEX audio pass thru format will causing the audio type field to be
enabled on the audio encoding Source tab
For Flex pass-thru audio, user only need configure the total bit rate setting on the encoding‟s
format tab, device will ignore all the other encoding parameters.

PR60372 -- It is recommended to disable the service and change the Flex encoding group,
as on activated map the changes may cause the encoder to raise „video wrong format‟ or
„loss of signal‟ alarms and the service may not output from the encoder.

PR55253 -- VBR Pass Through for services with PCR not on ES is not supported in this
release.

PR53852 -- When switching between "VBR Pass Through" to "Rate Change", services
with PCR not on ES may stop flowing out the ProStream 1000
Work around: Disable/Enable service

PR52167 -- If users removes more than 4 blades from a bNSG platform not all alarms on
the platform will be displayed

PR 46617 - Alarms are not raised for re-encoded reference service.

Currently the ProStream 1000 does not report this alarm. Please refer to ProStream 1000
release notes for details.

PR 41037 - In DPI multicasting configuration DPI Daemon always displays the service
nd
name for 1st TS even if the cue trigger is sending service for 2 TS.
This is not a service affecting PR, though there is an error in the information being shown

PR41075 - VBI circuit name with over 32 characters causes the multi-channel encoder to
issue HTTP error message
If a VBI circuit is created and selected with one line of teletext greater than 32 characters, the
multi-channel encoder will issue an HTTP error. This will not happen if the circuit name is 32
characters or less.

PR41180 - Re-status may be required in case that the GbE card has SFP inserted and SFP
parameters are not displayed.
On a ProStream 1000, a change in the SFP may require the user to re-status the device in order
to show its new parameters. This will happen only if a GbE Card with SFP inserted will be
assigned.

PR51485 - One RAC card gets stuck with loss of input alarm after the DA is powered back
Video and audio are provisioned (embedded audio). Failure of signal to the encoder will cause
Loss of signal on encoder.
If the „Correlate Cable Disconnect on First Device‟ is selected, NMX will correlate the service
affecting alarm on the video port and de-provision the video stream but the audio stream is left on
the device in 1:1 encoder configuration.
As the embedded audio stream is provisioned without the video stream, the device will complain
the provision sent.
To recover the situation disable the channel and re-enable after the signal is recovered

Page 11
 NMX 5.5.4.0 Release Notes

2.2.4 Graphical User Interface (GUI)

PR 80310 - When checking license alarms using NMX, user should re status the encoder
first to see the actual alarms.

PR 81846 – To avoid encoder inserting HTTP error or partially provisioned. User should
change LRV H264 encoding profile to Baseline or Main and the GOP M (Frames) should be
changed to 2 before activating the service configuration

PR60352 -- On a fully provisioned encoder, the Input extraction will take little while (up to
10sec). Defining additional ranges for extraction before the previous extraction is not
completed may crash DSM GUI. Please restart DSM GUI to continue working.

PR55551 -- Audio pass-thru input audio encoding always display "Compression" if

"MPEG-L2" format is selected.
For Flex pass-thru audio, user only need configure the total bit rate setting on the encoding‟s
format tab, device will ignore all the other encoding parameters.

PR59921 -- EAS: no validation for defining Two EAS - same IP and UDP, one without
Source Address and one with Source Address

PR55750 -- User needs to press on "Refresh" button to add notification for ProStar
On "IMS and CAS Data Extraction" window: Add notification that user needs to press on
"Refresh" button in order to see extraction for ProStar.

PR 56054 – After off-lining a Proview IRD device the icon may show red instead of
magenta.
The work around is to re-status the device.

PR52766 -- DiviTrackIP FEC, On the cross-connector 'TS Min svr rate' can be set to 1 mbps
but the ProStream 1000 can only support a min of 2 mbps.
Currently ProStream doesn‟t support TS Min Service Rate less than 2 mbps and there is no
warning from NMX.

PR52658 -- Virtual port icon in the switch GUI changes to as same as regular port GUI
Sometimes the NMX GUI shows the virtual port as regular ports on the Switch State dialog.
Closing and re-opening will show correctly.

PR52370 -- In Manual EAS mode, after IRD Signal for Test MSG, all EAS toolbars switch
automatically to EAS
NMX will always show EAS status as activated for the Test and Real EAS activation, though the
alarms are different.

PR48992 – When using IGMPv3, the property of “Use Same Source IP on Encoder” should
be set to “Disable” in Cross Connect
In this case we want the provision layer to send both Primary and backup port IP address on the
SSM info to the PS1K. You need to set the property of “Use Same Source IP on Encoder” to
“Disable”.

Page 12
 NMX 5.5.4.0 Release Notes

PR48770 – Auto-switch format: 720p format and resolution don't show in GUI when using
720p source
“Video Format Auto Switch” means device auto-detects the format, and switches to that format.
NMX GUI is not updated because NMX does not poll device to get format information when input
video format changes.

PR47551 - No sync between device and NMX regarding "backup service activated" alarm
In the service redundancy mode a switch from a primary to a backup service is triggered by the
device itself. Once the backup service is activated, the only mechanism to switch back to primary
is by the user taking action. Until that time, NMX continues to keep the service on the backup
service, whether the device asserts an alarm or it remits the alarm. Note that the primary and
backup devices are provisioned identically. This is the behavior of the program redundancy
feature. However, the NMX and the device can become out-of-sync. For example, if the "backup
activated "alarm is asserted on the primary, you may encounter the situation where you reset
primary device (when it is in an alarmed condition) and still see an NMX alarm even though the
device remitted the alarm. Likewise, if you reset the backup device, it may still have the alarm
however NMX remits the alarm. This issue is relevant for single remultiplexer, or 1+1 and N+M
remultiplexer configurations as well.

PR 46722 / 46747 - Spreadsheet View Does Not Refresh Shows Blank Window
If this happens, close the Digital Service Manager GUI and then reopen it.

PR 41206 - Profiles will not be deleted from Manage Service Profile

Deleting the profiles will not actually remove them from the hard disk or from the drop-down list
until a Domain manager restart is performed or the profile is manually deleted from the hard disk.

PR 42823 - Output service view window scroll bar may get into a state where you cannot
see all the Transport Stream information.
To recover from this, either restart the DSM or in some cases, you can go to the site level and
press the “+” button to expand all the directories but this does not always work.

2.2.5 Installation

PR67058 -- User needs to login to Windows with administrative privileges to run NMX
server functions. NMX is currently not supported with user that has no administrative
privileges because you do not have access to the registry.

PR43499 – Upon upgrade to NMX version 4.2 above users may encounter a windows error
message “HLDCGDaemon Process could not lock service database”
This is an erroneous error message which is not service affecting. Clicking OK will continue the
upgrade process as normal without interruption.

2.2.6 Stream Editor

PR 82937 – Currently the extraction of AC3 audio is not supported fully, and is recognized
as data stream.

Workaround: Delete the stream and create manually.

Page 13
 NMX 5.5.4.0 Release Notes

PR 62970 – NMX does not support delay adjustment on Flex audio pass-through yet, even
though this appears on the NMX GUI. Changing this parameter will have no effect on the
device.

PR 82115 – NMX does not support LRV streams to be in a DiviTrack pool. NMX will not
prevent in the user from adding LRV streams to the pool

Workaround: Do not add LRV streams to the Pool.

PR 67348 – Flex audio decode Electra 8211: Audio decode stream raises "Feature not
supported: The audio port is not supported with current software version. Stream
provisioning is not allowed."

"Feature not supported: The audio port is not supported with current software version. Stream
provisioning is not allowed." alarm can be raised for unknown reason when changing some
properties (like output transport stream bitrates‟, audio stream bitrates etc). (non flex encoder and
flex encoder)

Workaround: Disable the program where the alarm is raised, Restatus the encoder, Reenable the
program should make this alarm remitted

PR75293 -- The max bit rate for Electra 8000 encoder is 24MB/s but NMX does not limit this
in the Stream Editor.

Please refer to Electra 8000 specification when configuring it from NMX.

1
PR 61134 – Dolby Digital Plus descriptor tag for DVB systems is missing from descriptor
list. If this descriptor is required please contact Harmonic Technical Support to get a
custom descriptor.

PR 51365 – A new AAC Descriptor has been added 0x7C

NMX now supports two descriptor tags "DVB AAC descriptor [tag 0x79]" (per DVB/ ETSI TS 101
154 V1.7.1) and "DVB AAC descriptor [tag 0x7c]" (per DVB/ ETSI EN 300 468 V1.7.1 and DVB/
ETSI TS 101 154 V1.8.1). Tag 0x7c is the correct, current descriptor value for DVB compliant
systems. Legacy systems may continue to use tag 0x79.

PR 50338 - Problems in spreadsheet view

When using spreadsheet function in Stream Editor you may encounter two issues occasionally.
The workaround is to exit and open application again.

Blank spreadsheet screen with both pull downs grayed-out.

Spreadsheet does not refresh, stuck in the last state before making changes in the tree
view and does not reflect those changes (after going back).

PR 41206 – Deleting a profile from Manage Service Profile dialog will not immediately
delete it from the available profile list in the Edit service configuration dialog.

1
Dolby, Dolby Digital, Dolby Digital Plus, Dolby E, and AC-3 are all trademarks of Dolby Laboratories.

Page 14
 NMX 5.5.4.0 Release Notes

In order to delete this profile, the user must delete the actual file from the hard-drive. These files
are stored under c:\Program files\Harmonic\NMX\GUI\Graphics\Stream Editor Profiles

2.2.7 Redundancy

PR82421 – (NMX Chassis redundancy) Ocaasionaly provisioning cannot recover from

NMX Path selection failure after primary and backup encoders time out then recover
around the same time.

Workaround:
1. Re-status the mux clearing up the socket alarm.
2. Reset the encoder which has the master back-off alarm.
Upon coming out of reset the conditions for HHP will be different and hence the alarm will not be
raised.

PR61174 – NMX alarms “Loss of Stream Error” for mirroring TS on 2nd MUX when
switching between devices thru first A/B switch
This issue may happen in ProStar map with 3 MUXes in chain (MN20 => PS1K => ProStar) with
"CAS Data" socket injection at "Scrambling only" and output through this socket has one EMM
PID and Mirror TS with injection at Scrambling/Edge on ProStream device (second MUX in
chain).
Workaround: the stub switches for the TS mirroring only transports should be in automatic mode
only. Use the main transport stream switches to manually throw the switches.

PR55211 -- Dialog box prompted "Redundancy already set for Backup NMX server" during
setup, but Domain Manager GUI didn't displayed state of backup pane
NMX redundancy setup should be initiated between 2 PC‟s which have successfully been fully
configured for standalone use.
The message “Redundancy already set for the Backup NMX server” will show up in the following
cases.
1. NMX redundancy was removed on Primary PC while backup was not powered on or was
off the network. Subsequent NMX redundancy setup between these 2 PC‟s will show up
with this error.
2. A machine was pulled from another NMX redundancy setup without removing NMX
redundancy. NMX redundancy setup to this new PC will show up with this error message.

PR49540 -- Dynamic DPI CBR rate does not get passed to backup encoder during
redundancy switch
If the encoder redundancy happens during the splicing in progress with the service is configured
as CBR, the new backup encoder will not be configured to be CBR and transport at the
downstream device may oversubscribe as the Ad Server will be streaming the ad at the CBR,
where the encoder is allocating the VBR rates

PR49636 -- Encoder redundancy switch at the end of an inserted ad (DPI) potentially can
lose the end splice message
If the encoder redundancy happens during the splicing in progress with the DPI event is
configured to send the end of splice message for the downstream splicer to switch back to the
network, the end splice message will not be sent as the new backup encoder will be unaware a
splice was in progress, and there is no way to send the end splice message from the NMX
application. The Ad Server will typically detect the end of its playlist and generate an End splice
message to return to network.

Page 15
 NMX 5.5.4.0 Release Notes

PR48394 – Encoder redundancy causes service disruption and alarms on the modulators.

In the configuration of a ProStream 1000 and a modulator, a redundancy switch on encoders may
cause service disruption and the following alarms may become active on the modulator:

“Upstream cable disconnected”, “Clock PLL error” and “Loss of Input Sync”

The workaround is to add alarm latency on the modulator platform and ports by doing the
following.

1) Add 3 seconds latency for "Loss of Input Sync" alarm for ASI port in both primary and
backup modulator.
2) Add 3 seconds latency for "FIFO Underflow" alarm for ProStream 1000 port in both
primary and backup modulator..
3) Add 3 seconds latency for "Synthesis Clock PLL Lock Error" alarm for platform of
both primary and backup modulator.

PR 46593 - Encoder redundancy is only supported if both encoders (primary and backup)
have either FEC enabled or FEC not enabled.
We currently DO NOT support the configuration with FEC enabled on one device and not enabled
on the other device.

PR 46727 - NMX may cross connect the backup encoder on a real redundancy switch.
This does not affect the output. There is an issue with the Electra 7000 encoders where should
the format change from 1080i to 720p or the other way, the format change will take around 30
seconds to take effect. For more details please see the Electra 7000 release notes.

PR 44081 – Redundancy with Feeds Priority enabled.

When re-enabling the primary encoder after a redundancy to a backup encoder, NMX will switch
the feed back to the encoder with the highest priority. This will happen only if feed priority is
enabled. A work around is to set the backup encoder to the highest priority first, then re-enable
the primary encoder. At the operator‟s convenience, switch the source back to the primary
encoder.

PR 43590 – In Hot redundancy set up, the Ion IP output port may get a leftover GPF for
audio stream and the device will show no alarm.
The problem seems to happen only on the primary encoder. To resolve this issue, deactivate and
activate the stream to remit the GPF alarm.

2.2.8 Performance

PR42886 - HHP - GigE Switch Hard Reset Using Large Map may cause Multiple Service
Interruptions and Longer Stabilization time
Resetting a network switch that has 2 ProStream 1000 connected in 1+1 Hot configuration with
HHP may result in longer than usual service interruption. All services should resume correctly.

2.2.9 Provisioning

Page 16
 NMX 5.5.4.0 Release Notes

PR 73284 – If using Flex audio passthrough, user must set the Bit Rate fields. This is
located in the Source Tab under the audio encoder property page of NMX. The Bit Rate
value should be equal to: ES Bitrate + 10 % of the ES Bitrate.

Not doing so can lead to issues such as GBE buffer overflow and other audio passthrough
issues.

PR73942 – NMX only supports IP SPTS input for FLEX encoder with PS2K splicer in the
same network map

Workaround: Set up separate network map, one with FLEX encoder only map with MPTS output,
another map to take FLEX encoder outputs as an input to PS2K for splicing.

PR 72044 – Adding a MV500 channel to a DiviTrack pool may cause momentary CC errors
on the stream

PR 73272 – on the MV100, variable closed GOP Value of M=4 is not valid setting. If user
sets this value to M=4 the MV100 will send an SNMP Error, bad SNMP Value.

Workaround: Set the value between 1 and 3.

PR 71420 – DiviTrack needs a block of 6 ports for each intermediate Transport Stream.

Occasionally NMX may assign a invalid port number for the intermediate Transport Stream and
will not alarm.

Workaround; If NMX asserts alarms of " NMX will alarm "Loss of stream -intermediate TS
multicast address port configuration error". Disable the service and correct the IP cross
connection settings.

PR 61215 - If you have a map with a multiplexing level beyond a ProStream, there is a
possibility that, if you set up an IP mirror transport on the ProStream in the middle of the
chain, the mirror transport will be enabled on both the primary and redundant ProStreams,
causing an IP conflict on the mirror.

This configuration is very uncommon. The workaround is to put the IP mirror transport on the
final multiplexing level in the chain only, rather than on the intermediate muxing level.

PR 59666 – When off lining the VIA5 port on encoder to trigger encoder redundancy, the
MN20 will have an oversubscription for a couple of seconds.

It is recommend to offline the encoder to trigger encoder redundancy instead of off lining the VIA5
port.

PR52953 -- Slate: in IP - same service from input and from slate will not work
NMX will not validate the same IP service defined as Slate service in global options and as a
regular input transport and configuring in this way will not work as expected and will result
undesirable/unpredictable behavior.

Page 17
 NMX 5.5.4.0 Release Notes

PR51153 - Disabling the PCR stream or any single program disables the entire PCR on all
the programs,
When the PCR stream is added to multiple programs under a single transport as virtual streams,
disabling the PCR stream or any single program disables the entire PCR on all the programs.
This action will cause the other programs provisioned in the transport to not have a PCR. This
affects the service for the programs.

PR51683 – ProStream 1000 may alarm with "TableStreamOut File General Error " when
restarting the Domain Manager
If this should occur please rebuild PSI or restart the Domain Manager

PR48550 – While using a ProStream 1000, if a user mistakenly configures the EAS port/slot
that is already configured as an output port, a "Loss of Stream Error" alarm will NOT be
asserted, but actually the service configuration will not be provisioned on the device,
because the port X will be configured as Input port.

The workaround is to reset and ZAP the device

PR47573 – The ProStream 1000 web GUI will alarm GbE Input Socket Timestamp Error.
This will occur when the ProStream is unable to flow streams from an input transport.

When injecting data from an IP input transport into ProStream 1000 through a cross-connect or
directly received by the ProStream 1000 the user needs to specify the correct source IP address
or just enter “0.0.0.0” to disable the source filtering function in ProStream 1000. Otherwise the IP
data will not flow out of the ProStream 1000. This could happen when the input transport is
created from another (previously-saved) output transport in which source addresses had already
been configured.

PR 46899 – A False oversubscription alarms may appear on the encoder ASI port
The encoder ASI post may show an oversubscription alarm if the transport bandwidth is set close
to total minimum rate. The Alarm "Provision ERROR", "Hardware Oversubscribe" will show if the
NMX application is open. There is no impact to the CLEARcut captures based on this alarm.

PR 44812 - Power cycle Electra 7000 during file downloading, device may come back with
Card Busy alarm.
If this should happen, zap the encoder to recover.

PR40104 – While starting up the NMX, due to an NMX redundancy, the DPI Daemon will
stay on "Initializing" state when sending cue triggers.
If this happens the DCG EM will stay on "Managing" state and DM cannot complete startup. The
work around is to restart the NMX .

PR41874 - Loss of stream error asserted on PSI tables

Path finding for IP transports that have no other stream but PSI is not supported, these TS must
include other stream types.

PR 44752 – A primary encoder fails over via GbE cable disconnect and the alarm on
primary asserts, but it remits after about 20 seconds.

Page 18
 NMX 5.5.4.0 Release Notes

After a GbE port redundancy occurrence due to cable disconnect, the alarm on the primary port
may remit. This can give a false positive indication that the port connection is fine. Users should
physically double check that the primary port is connected before switching back to the primary
GbE port.

2.3 Legacy Known Issues in NMX Released Versions

rd
2.3.1 3 Party Devices – CherryPicker

Please note that the CherryPicker is no longer supported in NMX after NMX 4.3 release.

PR 27993 – NMX CherryPicker Alarm Display shows alarms that were remitted by device.

In a few cases the CherryPicker alarm display on NMX shows alarms that were remitted by the
device. This status does not update on the CherryPicker for a very long time (at least hours). For
an example of a specific case, the 'lost signal on input' alarm may be shown on one of the
CherryPicker inputs, even though it is active, and you can even extract inputs from it. A “Re-
status” of the device remits the alarm.

PR 30969 – CherryPicker does not remit some alarms.

Some alarms on the CherryPicker will still show, even if the alarm is no longer relevant or was
remitted (but the device still waits for acknowledgement). For example if a cable was
disconnected on either input or output - the alarm will remain until cleared, even if the user does
not intend to have that cable connected any longer (i.e., user logically disconnects the input or
output). The same thing happens for loss of signal alarm.

PR 31140 – The CherryPicker extraction function will not identify HD video streams.
This is currently not supported in the CherryPicker MIB and the extraction will show SD video
instead. The user needs to correct this manually using extraction for the definition of the input
transport streams

PR 31843 - CherryPicker output cable disconnect will put NMX into a 'toggle detected
state'.
After the toggle is detected and stopped, the system will still lose the disconnected streams. The
service will recover when the cable issue is resolved.

PR 32017 - NMX may show a (false) warning about exceeding bandwidth capability of
devices downstream from CherryPicker.
NMX may show a (false) warning about exceeding the bandwidth capability of devices
downstream from the CherryPicker. However, since the CherryPicker limits the bandwidth of the
output programs, as long as the downstream devices can handle the bandwidth set at the
CherryPicker output the NMX warning is irrelevant. When this alarm occurs it suggests that the
rate shaping requested on the CherryPicker is beyond the capabilities of the device. Try
increasing the bandwidth allocated to the alarming video streams, or have all video streams rate-
shaped (and not only some of them).

PR 32482 - CherryPicker requires ZAP to clear its states between redundancy and
provision changes.
If this is not performed, many times the NMX provisioning fails due to hidden configuration
information (captured PIDs) accumulated in the CherryPicker device.

Page 19
 NMX 5.5.4.0 Release Notes

PR 34673 - CherryPicker reserved bandwidth may not be sent to the device in some cases.
This problem is a result of having the reserved bandwidth parameter associated in NMX to the
CherryPicker pool, while on the device it is an output port property. In some cases the reserved
bandwidth will not be sent to an output port, such as:
 When there are multiple output ports, but smaller number of pools
 When input programs are being multicast to different outputs
Verifying with a sniffer shows that in these cases no command is sent from NMX to the
CherryPicker. This problem can have service quality effect on CherryPicker-BNG maps.
The workaround is to have all CherryPicker configuration files (configFile.txt) set with the
maximum required reserved bandwidth for BNG PSI output.

PR 35714 - DPI splicing will fail on NMX maps where encoder is connected directly to a
CherryPicker.
DPI splicing will fail on NMX maps where the encoder is connected directly to a CherryPicker.
Therefore, the DPI feature cannot support configurations where encoders connect directly to a
CherryPicker.

PR 36594 – The CherryPicker GUI does not show PMT flowing for virtual channels
CherryPicker GUI does not show PMT flowing for virtual channels but if you capture the Cherry
Picker output flow on an analyzer you will see that it is present. This is because the CherryPicker
GUI does not get updated with the additional PMT that is created by NMX for the virtual channels

PR 36672 – A CherryPicker reboot from the front panel, via telnet or using the
CherryPicker GUI does not get registered on NMX
A CherryPicker reboot does not get registered on NMX. NMX will still show a green icon for the
CherryPicker even through is not configured. If the CherryPicker crashes or has a power recycle
within 3 minutes of the last configuration change, that configuration will not be saved by the
CherryPicker. In any of these cases, NMX should take control of the CherryPicker once it has
rebooted and will reconfigure it as necessary. This may cause a slightly longer outage, but is very
uncommon.

PR 36863 - On the fly changes to CherryPicker TS bit rate do not take effect on GBE
output.
Changes on-the-fly to the transport bit rate does not take effect. The recommended approach is
to disable and re-enable the transport, or you can de-activate and re-active the services. This is a
limitation on CherryPicker. Once the GbE transport is created, transport properties cannot be
changed on the fly. This is the same case with CherryPicker GUI. The transport has to be deleted
and a new transport, with new properties, has to be created. From NMX, to change the GBE
Transport properties, the transport has to be disabled first. Then you can change the properties
and enable the transport.

PR 40130 – On a DM Network CherryPicker, a switch back to the primary program when

the primary is still failed will not switch to the backup DM Network CherryPicker.
This is a DM Network CherryPicker limitation. The device allows NMX to provision a program
even when there is no feed coming in (the presence of PSI is enough).

PR 40133 – Alarms for Program Redundancy can take up to 1 minute to reconcile.

If the primary program's input transport is disrupted upstream on a DM Network CherryPicker, the
switchover is immediate but the alarms can take up to one minute to fully be reconciled

Page 20
 NMX 5.5.4.0 Release Notes

PR 41299 - Mirroring port of CherryPicker ASI I/O card does not work on NMX (old 2 ASI 2
DHEI works)
There is no setting for port mirroring on the new ASI I/O card. The previous ASI/DHEI card did
support this feature.

PR 41999 - No alarm is sent by CherryPicker if the custom grooming has failed due to
missing streams in the input
If services are provisioned using custom (PID) grooming and there are missing streams in he
input, the CherryPicker will not alarm and no indication will be given to the user that streams are
not flowing except by looking at the CP stand alone GUI.

PR 33884 – When you zap the CherryPicker it will cause the GUI to hang for 30-60 seconds
When you zap the DM Cherry Picker it will cause the GUI to hang for 30-60 seconds depending
on the number of input and output ports that have been provisioned on the device. There is no
known work around for this condition.

PR 34137 - CherryPicker and MG-Soft MIB Browser Configuration are not compatible.
MG-Soft MIB browser requires “contact” to a device before it is possible to browse the MIB or set
any parameters. This browser '”contact” command involves performing “get” or “get-next” on a
specific MIB variable. By default the MG-Soft MIB browser will try to perform get-next on MIB-2
SysUpTime. The problem with this is that CherryPicker devices respond only to the Terayon
private MIB. When trying to get other MIB variables, the device will respond with an error
message. When trying to “contact” CherryPicker from the MG-Soft browser, the device will
respond with “no such name” error, and the MIB browser will consider the “contact” as failed, thus
not allowing any further commands to the device.

The Work-Around is to set the MIB browser preferences as follows:

1. Go to the MG-Soft -> View -> MIB Browser Preferences -> General -> Contact OID
2. Check the Contact OID box
3. Browse the MIB to ISO / org / dod / internet / mgmt / private / terayon / video /
networkCherryPicker, and select 'agentState' or enter directly 1.3.6.1.4.1.1456.20.2.1.0
4. Change Operation to "Get Request" (not mandatory)
5. Press OK.

Note that with this work-around, after changing the preferences, the same problem will happen
with all DiviCom and other SNMP devices.

PR 39635 - All services from DM Network CherryPicker are lost after upgrading from NMX
4.0
If a customer is upgrading a map from NMX 4.0 to NMX 4.0.0.1 and the map has a DM Network
CherryPicker device, the following steps need to be performed in the 4.0 map before upgrade.
1. Create a "CherryPickerPool" per output transport. (Name each pool as "Tr-1-Pool" for
convenience, where "Tr-1" is name of transport).
2. Add the video streams from the output transport to the respective pool created. DO
NOT add streams from different transport to a different pool. (The GUI does not
restrict this.) From the video stream property page, select "No recorder" for the newly
added streams.
3. If the service configuration already has a "CherryPickerPool", then make sure steps
1, 2 are implemented.
4. Finally, there should be as many "CherryPickerPools" as output transports and all
video streams should be under the "CherryPickerPools".
The Map is now ready for upgrade.

Page 21
 NMX 5.5.4.0 Release Notes

PR 31727 – CherryPicker transport bandwidth over subscription alarm does not take into
account the reserve bandwidth pool.
For example, the CherryPicker transport (output) can be set to 40Mbps. If a user is to provisions
38Mbps total output with a CherryPicker pool bandwidth reserve of 3 Mbps, there will not be a
warning that the user has provisioned more than the bandwidth allocated for the transport. The
40Mbps is reduced by 3Mbps to have 37Mbps of allocated bandwidth.

2.3.2 Third-Party Devices – Other devices

PR 43520 – (MV3550) ISO 639 Language descriptor is not updated at the output until
disable/enable transport or deactivate/activate is performed
The MIB and standalone GUI gets set correctly, but the audio language descriptor at the output
does not get updated.

PR 36282 - Alarm not remitted on NMX for Thales Granite when removing input cable and
connecting back.
The un-remitted alarm is "MPEG Analyzer PSI/SI/ATSC Table Syntax Error [ ASI Card - Input -
001/SA - 001 ] [ MPEG Analyzer PSI/SI/ATSC Table Syntax Error ]". This will be remitted if you
re-status the device.

2.3.3 Automation Server and Automation Admin GUI

PR 34068 – The auto pre-roll will not set in the Automation Admin GUI for parallel batch if
the batch is scheduled more than 5 minutes in the future

When you send a parallel batch to NMX from the Automation Server for changing encoding
properties, the auto pre-roll will be calculated immediately. However, if the batch is scheduled
more than 5 minutes in the future, when the batch is sent to NMX, the auto pre-roll is not set in
the Automation Admin GUI. At 5 minutes before the batch execution time, the auto pre-roll will be
properly shown in the Automation Admin GUI.

PR 38847 - The CA Descriptor Location option “Both” will not set descriptors in both the
program and component levels.
CA descriptor location can be managed by Automation Server at component or program level.
But you cannot use the “Both” option shown in the GUI pull down menu, this will result in only the
component level descriptors being set in the output.

2.3.4 Element Managers

PR 30824 – CAS CAT extraction isn't supported on BNGs or ProStream 1000.

Therefore EMMs appear as Ghosts in the extraction table. The following parameters should
appear as result of CAT extraction: EMM PID, client ID and Private data

PR 36622: Null packet insertion behavior is identical in NMX GUI and the Multi-channel
encoder stand alone GUI (Ion, Electra or Audio encoders).
When enabling null packet insertion on a Multi-channel encoder, the device will add null packets
to the transport in a rate that is designed to create a smooth transport flow at a rate which is

Page 22
 NMX 5.5.4.0 Release Notes

equal to the sum of all the elementary streams in the transport. NMX GUI allows defining a rate
for the output transport, but this is not reflected at the actual output of the encoder. The encoder
will still output at the minimum smoothed out bit rate.
In cases that a Multi-channel encoder was configured mistakenly to a transport rate which is
lower than the sum of all streams, NMX will generate an alarm.

2.3.5 Graphical User Interface (GUI)

PR 29843 – Error messages are generated in GUI when you close the GUI window.
If you leave several proxy windows open on a running system for a long time, when you close the
GUI window, you may get an error message that is generated by the OS such as “xxxxx has
generated an error and will be closed by Windows. You will need to restart the program.” When
this occurs reboot the server to insure that all processes are properly reestablished.

PR 29954 - Random Proxy site network connection problems may cause network
connection error messages.
This is potentially caused by loss of system event messages. The state of the proxy site is
typically updated on events from the server. If the event is lost, state will not be updated and you
will not be able to open the proxy site. The problem goes away if you refreshed the view by
clicking F5

PR 32470 – No indication is provided on the NMX GUI to indicate that the second Ethernet
port (CAS port) on the BNG is active or not.
With the NMX GUI you can configure the second Ethernet port via NMX or the BNG HTTP GUI
without even knowing that actually the port is in disable mode. If you telnet to the device and
trying to ping the IP address that is located on the CAS subnet you will get the following error
"Send to error 5041".

PR 33132 –Domain Manager "Option Settings" can get out of sync with other standalone
applications, such as Standalone Stream Editor and Automation Server Admin GUI.
You will encounter this issue if, from the Domain Manager options setting, you change some
preference for service when Automation Server Admin GUI or Standalone Stream Editor are also
open. Automation Server and Standalone SE will keep the original settings. If you then change
the service settings from Automation Server preferences and go back to Domain manager Option
Settings, they will be out of sync. The work-around is whenever you change preference settings
from one interface, always re-starting the other applications and they will then be in sync with the
Domain Manger.

PR 36995 – NMX allows ECMG names up to 80 characters however the BNG extraction
only supports lengths of 20 characters.
If you set ECMG names in the NMX GUI longer than 20 characters these are not saved in the
BNG extraction.

PR 37432 – If you configure the second output port on a BNG 6202 and do not close the
Properties page, NMX does not disable the ability to switch from slave mode.
If you configure the secondary output port on a BNG 6202 via the properties page, and you leave
this page open while connecting this output to a downstream object, NMX does not disable the
ability to switch from slave mode. The property window still allows you to configure the port as
slaved or not slaved, this should not be allowed. Always close the port properties page before
connecting the port to a downstream object.

Page 23
 NMX 5.5.4.0 Release Notes

PR 37813 – Disable SDT/NIT tables for one transport and the tables in another transport
may still contain that transport‟s data.
If you disable the SDT/NIT tables for transport A on the multiplexer, the SDT/NIT tables in
transport B, for example, will still contain data of transport A. However, if SDT/NIT tables are
deleted from transport A (instead of disabling), the SDT/NIT of transport B will not point to
transport A.

PR 38346 – NMX Spreadsheet View allows software change to monitored devices but
should not.
In spreadsheet view, you can attempt a software version change (pull-down menu selection) for a
device being used in the NMX “monitor only” mode and the NMX GUI will pop-up a message
saying that the device is going to be zapped. No zapping actually happens since NMX does not
control monitor-only devices.

PR 38808 – NMX main GUI will hang when deleting multiple cue trigger devices rapidly
from the network tree view.
This problem arises when you try to delete multiple devices rapidly from tree view before the tree
view has been refreshed for previous deletes. Waiting for the tree view to refresh before deleting
the next device will prevent this problem.

2.3.6 Installation

PR 34429 - after an NMX upgrade or re-install, alarms are no longer emailed to the desired
server or distribution list.
The desired nmx_notify.ini file is overwritten (removed/replaced) during the upgrade or re-install.
You must either remember to preserve this nmx_notify.ini file or re-create the email list after every
upgrade or re-install of NMX.

2.3.7 NMX Server Data Base

PR 30951 – Provisioning transports over two GPI monitored devices.

If you have two or more transports from a single re-multiplexer device that needs to be
provisioned over a series of two GPI monitored devices, then each transport must be modeled to
go through a separate unmanaged device. However, doing this may result in some issues related
to alarm distribution from a single physical GPI monitored device.

PR 31138 – If you disconnect an ASI cable between the primary multiplexer and an NxM
switch, the BNG asserts an upstream cable alarm, but no redundancy action takes place.

This is because NMX cannot determine whether the cable is disconnected between the switch
and upstream device or between the switch and the downstream device. Currently, NMX always
assumes this is on the downstream device and asserting on the downstream devices. Hence,
NMX does not have any other paths available to provision the services. If you encounter the
cable disconnect alarm, be sure to check the connections to the NxM switch as part of the
troubleshooting process.

PR 31786 – NMX BNG redundancy operation will go into the anti-toggle state when a single
input source to the switch preceding the BNGs is lost.
This occurs when redundant units are unable to find the source and continue to switch back and
forth until the toggle limit is reached and the toggle alarm is asserted. The current NMX
redundancy scheme cannot detect the lost source condition into the switch since it is not possible

Page 24
 NMX 5.5.4.0 Release Notes

to determine if it‟s the source, the switch or the BNG that has failed. Recovering the lost source
will enable the system to return to normal operation.

PR 32849 - EIS will corrupt the SCG information if changing different service
configurations over the same network when the SCG network IDs are same.
The SCGs created using the Automation Server GUI are not tied to any service configuration but
are tied to the network and transport IDs. When a service configuration is deactivated in NMX, the
SCG is deleted from the BNG, but it will still exist in the Automation Server SCG DB.
Subsequently, when the 2nd service configuration, that has the same network and transport ID is
activated, EIS tries to configure the SCG to the new BNG in new service configuration. If the new
service configuration doesn't have all the components in the SCG, then you will encounter
problems (such as the SCS rejects the SCG or shows alarms etc.). If the service configurations
are switched, then user has to make sure the Automation Server is in sync with new service
configuration or rebuild new SCG database in the new service configuration.

PR 34629 – A BNG 6202 configured to output 2 transport streams, 1 scrambled, 1 not

scrambled, will lose circuits from one of the transports.
You will encounter this problem if, after configuring the BNG with 2 output transports you
subsequently add circuits (either audio or video) to the second unscrambled transport, the added
stream will appear in the unscrambled stream but will not appear on the scrambled stream. This
can be resolved by rebuilding the second transport as scrambled even though is isn‟t.

2.3.8 VOD Statistics

PR30317 – When no statistics data is retrieved from an NSG device this will not be treated
as zero data in the history for this time period.
The VOD utilization chart presentation type is 'linear chart'. As a result of this, a straight line will
always connect two consecutive data points. Specifically, if data is missing for a period of time
for the queried object's statistics - a line on the graph will connect between the last point of data
and the first point of continued data, and will not show zero utilization for the period of time
between these two points. This may cause the graphs to look unusual when there is missing data
for a long period of time.

PR 30510 - The VOD statistics web server can get randomly stuck while trying to access a
specific report query page
After this occurs no query pages can be opened even though the report tree still loads. If this
happens, reload the web server by doing the following: go to Control Panel -> Administrative
Tools-> Services and stop the World Wide Web service and the IIS publishing service. Also
restarting your PC will correct the problem.

2.3.9 Stream Editor

PR 31640 - Extraction of sources with virtual channels isn't supported.

Only one virtual channel with elementary streams appears on extraction window. For the
rest of the virtual channels only PMT PID appears under service.
In the case of common PMT only: all channels appear on extraction, but only one of them
contains PMT PID
Note that on Embedded Viewer the extraction appears correctly

Page 25
 NMX 5.5.4.0 Release Notes

PR 31979 – When using an MV45 encoder the device will report a GPF (general profile
fault), H (horizontal) filter alarm when video is provisioned.
When setting up device configuration using Stream Editor you can override settings, but keep the
H filter set to “Flat” and no GPF alarms will be reported.

PR 32564 - For all encoders, if you are using NMX 2.0 and later versions of the Audio GUI it
will show "Max Input Level".
This is only applicable to MV40/45. For MV50/100 “Max Input Level” is not applicable, instead
these encoders use "Gain" for the audio output level. While this is an inconsistency in the
GUI, all encoders under NMX 3.0 use "Reference Level" to set the actual audio volume
output. The "Gain"/"Max Input Level" field is strictly for operator information only. See the NMX
audio GUI dialog box for details.

PR 34093 – Some users logging into Stream Editor will not be able to see all menu items in
the short cut menus.
When certain users log in to Stream Editor, the Edit Service Configuration dialog box does not
include some menu items in shortcut menus, even though the user should have access to these
items. The missing menu items are Enable/Disable of streams and Rebuild PSI. These menu
items are available in the shortcut menus in the Service View tab of Tree View.

PR 34764 - Changing transport PCR settings (for rate shaping) will cause a rebuild PSI
alarm.
You can ignore this alarm since it is not service affecting (PCR has no connection with PSI) or
you can also clear this alarm by rebuilding PSI (which results in no PSI changes).

PR 37165 – Default for the ASI Monitor port shows as „-1‟ on BNG back panel.
Go to the back panel of BNG and edit ASI Monitor port, the default value of TS ID of ASI Monitor
port is '-1'. This should default to zero and it can also be set to a user-defined value.

PR 37818 - If you make an Access Criteria Change this will incorrectly raise a 'Rebuild PSI'
alarm.
Up to the current NMX release, as long as the property exists on the output side of the map and it
is being touched by some change in the service configuration, NMX raises the Stale PSI warning.
NMX does not keep a detailed mapping and does not check to see what properties really affect
which tables or which descriptors.

PR 38444 – The Framing Mode for QAM modulators controlled by NMX is fixed at MPEG
188.
You cannot use the Stream Editor to change the Framing Mode of a QAM modulator. For
modulators controlled through NMX the default QAM modulator Framing Mode setting is MPEG
188.

2.3.10 Redundancy

PR 33718 – If you are using a BNG N+1 redundancy scheme you cannot successfully swap
two primary BNGs with a backup BNG
If you are using a BNG N+1 redundancy scheme you cannot successfully swap two primary
BNGs with a backup BNG due to an incorrect configuration swap between Primary # 1 and
Primary # 2. This can only occur with IP input transport streams and the correct swap of devices
can only happen in the case where all the output transports can coexist at the time of path
calculation. In other words, if the BNGs are connected on the output with 4 RF ports, and
currently there are only two transports provisioned, the swap over can happen since the BNG can

Page 26
 NMX 5.5.4.0 Release Notes

have up to 4 RF ports, and during the path calculation the backup BNG can coexist with the
transports from the BNG2 and the transports from the BNG1 which is failed now.

2.3.11 Performance

PR 34228 – A multiplexer over-subscription can be caused by deletion of a DiviTrackXE™

pool.
When services from a pool 1 are deleted and added to a pool 2, and the pool 1 deletion is in the
same command buffer as the pool 2 addition, the multiplexer will oversubscribe. The problem is
service affecting due to the multiplexer over-subscription. The workaround is to separate the
deletion command from the addition command i.e., make these changes using separate
commands. In other words, perform the “Apply” action after the pool delete.

39358 – On a BNG with MPTS input you may get the following alarm: Get alarm IP Network
Packet Loss [ IP Input - 001/Gbe - 001 ] [ IP Network Packet Loss ]

This is a false alarm and has no impact to the BNG performance. If this should happen disable
the alarm in NMX

2.3.12 Provisioning

PR 37876 - PMT table is still spooled after disabling services with Virtual RSS
Scenario: Add service X with RSS and then add Service Y with the same RSS. If you disable the
PMT of service Y, then rebuild PSI, and then disable RSS of service Y, the PMT of service Y will
flow as a ghost. The same scenario on X does not cause the PMT to flow as a ghost.
Workaround: delete service Y instead of disabling its PMT and RSS

Note 1 – This happens only if RSS is output as virtual for a couple of services
Note 2 - In the case where you add RSS X first to service A then to service B - the bug will
happen only when disabling service B. when disabling service A BNG behaves as expected.

PR 38052 – NMX provisioning of ION may not work when channels or programs have
different stream configurations, or (audio) cards are connected using different ports.
This problem happens on systems where the encoding channels within the single ION unit are
configured differently in terms of number of streams and audio cards configurations. This is
likely to occur where path selection is available before the Ions (i.e., input switch). The problems
which have been seen are 1) failing to find path, 2) having 4 programs from a single input icon
provisioned to more than one Ion encoder, failing to complete redundancy actions. Using
balanced switch, or off-lining the input ports that are not connected solve the first two problems,
but not the third one.

PR 38603 – In a DiviTrackIP™ system configuration when all circuits are removed from a
statmux pool on a BNG, the encoders go into self-sync mode and can cause continuity
counter errors in the output.
Performing an operation where all circuits are removed from a pool results in 2 XML trees being
sent to the BNG. The first one deletes all the circuits from the pool. The second XML deletes the
pool. The BNG is responding to the delete pool command before all the circuits have completed
transitioning out of the pool. You will see CC errors dependent on the current mix of video
content (current video complexities). You may also encounter and over subscription at the BNG

Page 27
 NMX 5.5.4.0 Release Notes

output. Both of these conditions are transient. Workaround: delete all circuits except one, press
“Apply”; delete last circuit, delete pool, press “Apply”.

PR 39398 - Upon activation, the BNG asserts a critical alarm 'IP Network Packet Loss' too
soon, Electra 1000 provisioning is slower.
With Electra connected to a BNG 6202 multiplexer when the stream map is activated, NMX
provisions all the devices and the BNG asserts IP Network Packet Loss [ IP Input - 001/Gbe - 001
] [ IP Network Packet Loss ] too soon in the process. Eventually the Electra‟s will catch up and
the alarm will go away. Note this alarm is not service affecting.

PR 39716 - DiviTrackIP system: Following a BNG redundancy initiated by a BNG Reset, the
service alarm 'Generic Route Fault PID missing' will be raised when backup BNG comes
back online.
This condition can occur when two BNGs are in a 1+1 hot backup configuration and the primary
BNG is reset. A redundant switch to the backup BNG will occur correctly but when the primary
BNG comes back up and goes online, NMX will re-provision it. This causes the backup BNG to
assert the PID missing alarm on all streams for a short time period, this alarm will be shown on all
the stream circuits on NMX as “Generic Route Fault [ v14 ] [ Due to: Fault on [ v14 ] [ PID missing
440018:0x102B 5300018:0x1714 ] ]” alarms. The alarms are transient however if they persist for
a long time period, then NMX will correlate an upstream alarm on the Electra 1000 unit and may
result in an Electra redundancy.

Page 28