Last revision: 5/23/06
This document contains important information about Wonderware® Industrial Application Server 2.1 and 2.1 Patch 01. Review this document for known issues in these versions and changes to the Industrial Application Server documentation included on this CD.
Release Summary - Industrial Application Server 2.1 Patch 01
Installation Information - Industrial Application Server 2.1 Patch 01
Resolved Issues - Industrial Application Server 2.1 Patch 01
Known Issues - Industrial Application Server 2.1 Patch 01
New Features in Industrial Application Server 2.1
System Requirements - Industrial Application Server 2.1
Installation Information - Industrial Application Server 2.1
Before You Begin - Industrial Application Server 2.1
Known Issues – Industrial Application Server 2.1
Known Issues – UserDefined Object 3.1
Known Issues – Industrial Application Server Documentation
Industrial Application Server Documentation
Customers running Microsoft SQL Server 2005, Microsoft SQL Server 2005 SP1, or .NET Framework 2.0, must install this patch.
The Industrial Application Server
uses the .NET Framework 1.1 SP1. This patch only allows for coexistence with
the .NET Framework 2.0.
Before installing this patch or upgrading Microsoft SQL Server, we strongly recommend that you exit all Wonderware programs, including executable (.exe) files and services. If you do not exit all Wonderware programs and services, you may see errors in the logger. Rebooting resolves these errors. After upgrading, restart the computer even if you are not prompted to do so.
This patch was tested on:
· Windows XP SP2, Windows 2003 SP1, and Windows 2000 SP4
· .NET Framework 1.1 SP 1 and .NET Framework 2.0
· Microsoft SQL Server 2000 SP4, 2005, and 2005 SP1
Important! The Galaxy Repository locks the SQL Server maximum memory usage to 65 percent of the physical memory. For optimal performance, we recommend that the Galaxy Repository and SQL Server node has at least 2 GB of physical RAM. Also, you cannot install and use Industrial Application Server on a computer that has both Microsoft SQL Server 2000 and Microsoft SQL Server 2005 or 2005 SP1 installed.
Under SQL Server 2005 or 2005 SP1, some degradation in the performance of certain operations associated with the GR may occur. Wonderware continues working on optimizing the performance of the system under SQL Server 2005 or 2005 SP1 as a major goal in the evolution of our products.
To determine which Wonderware products are supported with this patch, see the Compatibility Matrix at: http://www.wonderware.com/support/mmi/esupport/matrix/index.asp
If Industrial Application Server 2.1 is already installed, install only Patch 01 (setup.msp). Otherwise, install the full remastered version of Industrial Application Server 2.1 with Patch 01 (setup.exe). The version number for both the patch and the remastered version is 2.1.001.
The
following defects were fixed for this patch release.
L00044902: If the .NET Framework 2.0 is installed on the
Galaxy Repository node, some objects fail to deploy to remote nodes on which
.NET Framework 1.1 is installed. A communication error occurs.
L00045106: When security is enabled for the Galaxy, the AckMsg attribute does not work.
L00045255: After you import a .NET Framework 1.1
script library, the IDE sometimes shows an error and automatically exits.
L00045330: If the .NET Framework 2.0 is installed on the
Galaxy Repository node, and the .NET Framework 1.1 is installed in a remote
node, the remote engine does not respond after an object is deployed that runs
a script referencing the remote node.
L00046320: If GRAccess is used
with Microsoft SQL Server 2005, some operations fail.
L00046422: You cannot save an object built using the ApplicationObject toolkit.
L00046462: The InputOutput
extension performs two write operations instead of one write operation.
After upgrading to Microsoft SQL Server 2005 or 2005 SP1, restart the computer even if you are not prompted to do so.
If you disable ping on the OPC server node, FSGateway and OPCClient cannot not connect using the OPC interface. Both clients must receive a successful ping before continuing the connection; otherwise, the connection cannot succeed. To correct this, enable ping on the OPC server node.
McAfee Desktop Firewall 8.0 may prevent SQL TCP/IP connections, causing problems connecting and deploying galaxies. Turn off the firewall to solve these problems.
Industrial Application Server (IAS) 2.1 provides key features requested by users.
A new software license file is REQUIRED to enable this version.
IAS 2.1 is the next in a series of software enhancements that allows you, the Application Developer, System Integrator or IT Professional, to build powerful HMI, Supervisory Control, SCADA or Production and Performance Management solutions with the lowest total life-cycle costs of any system.
Industrial Application Server 2.1 provides the following new
functionality:
This configuration allows you to start implementing pilots or small systems in a single node and later grow your Galaxy as needed. See the Compatibility Matrix located on the Wonderware Technical Support Web Site for additional information on combining products on a single computer.
· Scaling of Input and Output values
· History
· HiHi, Hi, Lo and LoLo Limit Alarms
· Rate of Change Alarms
· Target Deviation Alarms
· Bad Value Alarm
· Statistics
· State Labels
· History
· State Alarm
· Bad Value Alarm
· Statistics
Important! Industrial Application Server 2.1 requires installing Microsoft SQL Server 2000 with Service Pack 3a. You must ensure that the proper service pack is installed for proper functionality. You must also have .NET 1.1 SP1 installed.
Note: Refer to Microsoft documentation on details about implementing Microsoft SQL Server 2000 with Service Pack 3a.
To run Industrial Application Server, we recommend at least the following software and hardware configurations:
Software Requirements
FactorySuite A² Development seat - IDE with Galaxy Repository (Project Database)
MicrosoftÒ SQL Server 2000 with Service Pack 3a and
MicrosoftÒ WindowsÒ Server 2003 with Service Pack 1 or
MicrosoftÒ WindowsÒ 2000 Server with Service Pack 4 or
MicrosoftÒ WindowsÒ 2000 Advanced Server with Service Pack 4
Important! The Microsoft SQL Server login for BUILTIN\Administrators group must be present and enabled.
FactorySuite A² Development seat - IDE with no Galaxy Repository (Project Database)
MicrosoftÒ WindowsÒ Server 2003 with Service Pack 1 or
MicrosoftÒ WindowsÒ 2000 Professional with Service Pack 4 or
MicrosoftÒ WindowsÒ 2000 Server with Service Pack 4 or
MicrosoftÒ WindowsÒ 2000 Advanced Server with Service Pack 4 or
MicrosoftÒ WindowsÒ XP Professional with Service Pack 2
FactorySuite A² Application Server Runtime
MicrosoftÒ WindowsÒ Server 2003 with Service Pack 1 or
MicrosoftÒ WindowsÒ 2000 Professional with Service Pack 4 or
MicrosoftÒ WindowsÒ 2000 Server with Service Pack 4 or
MicrosoftÒ WindowsÒ 2000 Advanced Server with Service Pack 4 or
MicrosoftÒ WindowsÒ XP Professional with Service Pack 2
Hardware Requirements
PC with 2 gigahertz (GHz) or higher processor clock speed
1 gigabyte (GB) or more of RAM (512 MB minimum supported; may limit performance and some features)
8 gigabytes (GB) of available hard disk space
Super VGA (1024 × 768) or higher-resolution video adapter and monitor
CD-ROM or DVD drive
Keyboard and Microsoft Mouse or compatible pointing device
Bootstrap, IDE and Galaxy Repository are supported on the following language versions of MS Operating Systems: English, Japanese, Chinese, German and French. The Galaxy Repository is also supported in English, Japanese, Chinese, German and French version of MSSQL Server 2000.
Read the Industrial Application Server Deployment Guide found at www.wonderware.com/support/mmi under the Documentation link.
The Microsoft SQL Server key is required to install the MS SQL Server. When installing the MS SQL Server, you are asked to enter the 25 digit key number. The Microsoft SQL Server key number is GJDD4-GMYXW-GQR4C-KT4P9-QDQGB.
You must have Microsoft SQL Server with Service Pack 3a installed before you install Industrial Application Server. Download this service pack from Microsoft at http://www.microsoft.com/.
Note: During the installation process you may see an error message that you need Microsoft SQL Server 2000 with Service Pack 3a to continue. If you have Microsoft SQL Server 2000 with Service Pack 4 on your computer, ignore this message and continue the installation by clicking Next. Everything will work normally.
You must also have .NET 1.1 SP1 installed on your computer.
Note the following important issues before you continue:
If you are upgrading from Industrial Application Server 2.0 or Industrial Application Server 2.0 Patch 01, you must upgrade all the nodes as well.
Trying to import a large
Galaxy with multiple levels of template containment can take a long time.
Actual time is dependent on the size of the Galaxy and number of template
containment associations that must be handled during the import process.
If Microsoft Windows
Terminal Services is configured in Application Server mode, creating, importing
and deploying Galaxies are significantly slower.
Anti-virus software should not process files in the following ArchestrA folders:
C:\Program Files\ArchestrA\Framework\Bin\CheckPointer
C:\Program Files\ArchestrA\Framework\Bin\GalaxyData
C:\Program Files\ArchestrA\Framework\Bin\GlobalDataCache
C:\Program Files\ArchestrA\Framework\Bin\Cache
C:\Documents and Settings\All Users\Application Data\ArchestrA (default setting, specified on WinPlatform editor’s General page, History store forward directory option)
If the Anti-virus software does so, this may result in slow performance as deploys are performed.
If you try to install Industrial Application Server 2.1 while ArchestrA applications are running, the Files in Use dialog box appears. On a Windows Server 2003 computer, the applications shown in this dialog box may be incorrect. For example, it may list “ArchestrA IDE” three times. You can disregard the specific information provided by this Microsoft operating system dialog, but it does indicate that you must quit at least one ArchestrA application before installation can proceed.
If you remotely upgrade Industrial Application Server from a previous version to 2.1 using Terminal Services, an MSI application error message appears at the end of the installation process. Also, errors are sent to the Event Logger. These messages can be safely ignored. Installation completed successfully.
When
setting up SQL Server on the Galaxy Repository node, "Dictionary order,
case-insensitive, for use with 1252 Character Set" must be selected.
Otherwise, lockups in the ArchestrA IDE may occur.
If you uninstall and then reinstall ArchestrA, starting the IDE logs Warning messages in the Logger. These Warning messages do not mean a system problem happened. Everything should function properly.
The ArchestrA user account that enables node-to-node communication (created during ArchestrA installation in the User Name and Password dialog box) must have read/write permissions to a checkpoint directory location if that directory is on a network share.
During the Installation Wizard, a button becomes available to open the Installation Guide online help. The online help opens and then reports that the page cannot be found. Security patch 896358 from Microsoft prevents chm files from opening from a remote location. For more information, see Microsoft’s website.
Industrial Application Server on an InSQL node may corrupt the registry entry for the "InSQLConfigPS.dll" object. To correct this problem, re-register this DLL located in C:\Program Files\Wonderware\InSQL\Server.
If Microsoft Windows
Terminal Services is configured in Application Server mode, creating, importing
and deploying Galaxies is significantly slower.
Trying to unassign as many as 700 templates (or 4,000 instances) at the same time may result in only some objects being unassigned and an IDispatch error being appearing in the progress dialog box.
During a typical Galaxy Load, a set attribute operation does not cause Error messages to be logged when both the set attribute value and the target attribute’s value are the same. This is not true for Switch and FieldReference objects. The Galaxy Load operation is completed successfully, but Error messages are generated during the operation.
Terminating the IDE application with Task Manager when a bulk operation is in progress causes Errors to be logged to the Logger, impacts the functionality of other IDE sessions, and requires restart of aaGR service on the Galaxy Repository node. You can restart the aaGR service by rebooting the computer or shutting down and then restarting the service through Control Panel.
In a multi-user environment, right-clicking an object, clicking Assign To in the shortcut menu, and trying to use the list in the Assign To dialog box may not show all available objects. This problem may occur even when you created the object expected in the list. If you type the name of the object in the Assign To box, the object appears in the list.
Opening IDE Help fails if the default locale of the Galaxy Repository/IDE node is set to a non-English language.
Creating objects using InTouch WindowMaker’s SmartSymbol capability causes an internal IDE session to be created and held open. Consequently, certain actions on other IDE sessions that require exclusive access to the Galaxy Repository may be prevented. For example, if you try to configure security in the IDE, the following message appears: “Configure Security : Cannot configure security as more than one session of the IDE is running. Open in read-only mode?” To close the internal IDE session created by WindowMaker, you must quit WindowMaker and WindowViewer (if running).
An “invalid data” message does not appear if you enter invalid data for an element of an array on the UDAs page of an object’s editor and then move to the UDA name list. “Invalid data” messages appear if you change focus to any other control on the page.
If you extend an object with a UDA that contains double-byte characters, trying to use the Edit Filter function of the Attribute Browser on that attribute does not perform any filtering at all.
If you select OS Group Based security as your ArchestrA model, the following scenario may happen. When you add a Role on in the Configure Security dialog box, the Select Groups dialog box appears. The Available OS Groups list is populated only with those groups that belong to the domain shown in the Select In box. This occurs because the domain controller does not provide a list of its groups to non-domain users who do not have permission to browse groups on other domains.
If you set the IDE to non-maximized mode, dock the Application View or Template Toolbox windows outside of the IDE client area, and then change the screen resolution on your monitor, the Application View/Template Toolbox windows may disappear from view. Reset the screen resolution to its original value to see these windows again. Re-dock them inside the IDE client area before changing the screen resolution again.
UDAs of type "OBJECT Writeable" can be configured in the IDE to have a user specified initial value. After doing a Galaxy Dump and Galaxy Load, "OBJECT Writeable" UDAs resets to a default initial value.
Occasionally, deleting a template object results in an hourglass that won’t go away after the object is successfully deleted. Move the mouse to remove the hourglass.
When an object is imported that brings in a newer component that is common to the WinPlatform object, no other object can be deployed or redeployed in the Galaxy until its WinPlatforms are redeployed. This problem would apply to importing objects like DIObjects, which contain components like the DAServer Manager that are installed when WinPlatforms are deployed.
Disable hyper-threading on all ArchestrA nodes. Some computer vendors (for example, Dell) may enable hyper-threading by default. To disable this feature:
If you reassign objects from one host to another, occasionally the IDE may not refresh the screen. If this occurs, the reassignment function was successful, but you should quit and then restart the IDE to refresh the screen.
Sometimes after working in the Security settings for a while, your changes may not appear in the InTouch Security Logon. For example:
Node A:
1) Change the security mode from None to OS-Group Based and add some Group (such as Wonderware_hq\Domain Users) to the Roles and click OK.
Node B:
2) Create new IT application, select ArchestrA as security mode and Switch to View.
3) Attempt to log on.
If this happens, redeploy the client platform and re-start View. Just rebooting the client machine does not help.
If you are running IAS on Japanese Windows system, don’t mix Half-width Katakana or Full-width Katakana when naming Security groups. If you do mix them, you may see duplicate objects.
If you change users and the new user has security restrictions, you may not see a prompt reminding of these restrictions in Windows 2000 if you leave the password blank in OS and OS groups Mode. If you are running under Windows 2003, you see the prompt.
If you edit the HTML help file for an object in Word 2000, the IDE crashes when you try to open the edited help file in the IDE. To prevent this from happening, use Word 2003 or an HTML editor to edit the file.
Imported script function
libraries written in Managed (.NET) C++ do not appear in the Script Function
Browser.
To manually modify the XML file
in an .aaSLIB file to designate a script function
library as a COM object requiring registration, follow the procedures in the
“Script Functions” section (“Enhancing an Object’s Functionality” chapter) of
the IDE documentation. Script library re-import registers the COM DLL into the
Galaxy file repository. Note that if objects that have scripts referring to the
.aaSLIB are already deployed, then nodes where those
objects are deployed do not get the latest script library unless you redeploy
the platform. After you cascade redeploy the platform,
objects use the new script library.
If you are importing a large .aaPKG file (.aaPKG file size of more than 100MB and object count more than 30,000), at the end of the import, you may see errors in the log that look like:
“Error CRLinkServer From Microsoft OLE DB Provider for SQL Server. Transaction (Process ID 55) was deadlocked on lock resources with another process and has been chosen as the deadlock victim. Rerun the transaction.. Access is denied. (80070005) raised.”
If you see this error, reboot the GR node. After you reboot the GR node, the galaxy will function normally.
If you try to upgrade a GR node after one of the remote nodes is upgraded while objects are deployed, you can't upgrade the GR node. When trying to undeploy the remote platform the Mark as Undeploy option, you see a version mismatch error. The platform cannot be undeployed using the Marked as undeployed option because there is a version mismatch.
The workaround to this is:
Both nodes are upgraded and you can now redeploy.
Validating an object with a script that uses an array indirect causes an error. An example of an array indirect is the following:
Dim index[5] as indirect;
index[].bindto(“test.float[]”);
Array indirects do not work in this release. Objects using them are not validate and therefore cannot be used.
The following problem occurs when running scripts that use .NET libraries, imported dlls and WDF. Opening a SqlConnection for a long time can cause a memory leak. Microsoft recommends against keeping a SqlConnection open for a long time. If you want to do so, though, you can eliminate the leak by adding “Min Pool Size=2” to the connection string in the script.
The following quality-related script functions do not work with an alias that references a quality function in the hosting object: SetBad(), SetGood(), SetUncertain(), SetInitializing(). When you save the object configuration that contains the script function, a warning appears about not being able to resolve the reference. If you deploy an instance of the object, quality is not set as expected (for example, Bad, Good, Uncertain, Initializing).
If the syntax of a script
expression is invalid, an error message appears in the Errors/Warnings
page of the object’s Properties dialog box. In this release, the message
may be wrong. For example, if you use a semi-colon at the end of a script
expression, the error message states “Expected ‘;’ ” when it should state that
it is invalid.
If you undeploy the WinPlatform from the InSQL 8.x node, the InSQL console stops working and the following Error message appears:
“Error on server xxxx - interface not registered.”
To recover from this problem, deploy the WinPlatform again to the InSQL 8.x node. To safely undeploy the WinPlatform from an InSQL 8.x node, use the remove platform functionality in the System Management Console.
ArchestrA requires that the DCOMCNFG Default Impersonation Level be set to "Identity" on a Windows 2000 Server node. If the Default Impersonation Level is set to Anonymous, deploying a WinPlatform to the computer fails. To set this option, run DCOMCNFG from a command prompt on the Windows 2000 Server computer. Select the Default Properties tab and set the Default Impersonation Level to Identity. Deployment of a WinPlatform to the node now succeeds.
When redeploying objects to a remote node on a slow network, multiple error messages are logged to the Logger if the computer is disconnected from the network. Even if you reconnect the computer to the network, redeployment continues to fail because Microsoft DCOM times out. About 15 minutes after reconnecting the computer to the network, you can redeploy again.
Assume a deployed object that references another object, and the latter object is subsequently undeployed and deleted. The state of the referring deployed object changes from Good to Warning. But if you manually validate this object, its state changes to Pending Update. The difference is due to the fact that as soon as the object is deleted, the Galaxy Repository marks all of the referring objects in Warning state whether they are deployed or not. However, the validate operation could mark the objects in Error or Warning states because of other reasons besides a broken reference. That is why the validate operation creates a new version of the object when changes are encountered. In this particular scenario, the issue is the different state of the object caused by the deletion of the referred object.
Deployment in the following procedure fails, assuming two sets of objects hosted by separate AppEngines that are already deployed to the runtime:
The delta configurations of the affected objects are not deployed. To do so, deploy only those hosted by a single AppEngine in one operation, and then deploy the second set of objects.
The InTouchAlarmHistoryBufferSize key in the Registry can be used to configure the alarm records available in the Historical mode of the AlarmViewer control. If you change the value of this key from its default and then undeploy the WinPlatform object, that change is not persisted. In other words, when you deploy the WinPlatform again, the key’s value is the default. You must manually reset this key value each time you deploy the WinPlatform.
Create an instance of $AppEngine and set its CheckpointPath to a node that does not exist. When you deploy, you see several deploy error messages in the Deploy dialog box and in the Logger.
The errors in the logger are:
aaEngine - CApplication::InitializeCheckpointer - AppEngine_002. Failed to open checkpointfile. FsEngine exiting ...
aaEngine - WinMain exiting error code 80004005
The GR marks the object as undeployed and the SMC indicates the engine exist in the runtime and is shutdown (startup failed).
Correct the $AppEngine to specify a CheckpointPath that does exist and redeploy.
If you rename an area without redeploying all its child areas and objects, the child areas and objects will have incorrect Area value and hiearchical names. To solve this problem, undeploy and redeploy all the child areas and objects of the renamed area.
An OPCClient object on NodeA is connected to an OPC Server on NodeB and then NodeB is shutdown. Undeploying the OPCClient object hangs the AppEngine for 1 minute before the AppEngine recovers. Not all OPC servers hang the OPCClient on the AppEngine.
We don't recommend that you deploy to:
Always deploy to nodes that are not involved in a fail over event.
PV.Quality changes between Bad and Initializing are not sent to the InSQL historian for instances of AnalogDevice and FieldReference $Float and $Double objects. InSQL stores “NULL” for Bad quality and interprets Initializing as a form of Bad quality. Since there is no change (from “NULL” to “NULL”), nothing is stored in the historian when PV.Quality changes from Bad to Initializing.
When an ArchestrA node goes into StoreForward mode (it has been disconnected from the InSQL node), eventually it reaches its storage threshold. After this point, it begins deleting the oldest blocks of StoreForward data. This operations functions properly, but logged messages tracking this deletion process indicate an incorrect order of deletions.
With a memory usage utility like Performance Monitor, you may notice a gradually increasing consumption of private bytes by the Microsoft module, LSASS.exe, on a node sending data to InSQL. According to Microsoft, memory consumption levels off at an amount proportional to the total memory on the computer. Also, the memory manager releases memory when other applications require it. This issue is not a memory leak, and it should not affect your system’s functionality.
When a redundancy failover occurs, the quality of a dynamic attribute in a hosted object is set to Initializing until the Standby AppEngine becomes the Active AppEngine.
Trying to
undeploy hosted objects during a Redundancy Failover
operation results in the objects not being undeployed
but marked as undeployed. Note that the failover is successful and everything
is running properly on the new Active AppEngine,
including the objects you tried to undeploy. To
successfully undeploy them, you must deploy them,
selecting the Mark as Deployed check box in the Deploy dialog
box, and then undeploy them.
During a failover, data
quality for attributes from DIObjects may change
briefly to Initializing and then to Bad before returning to Good. This behavior
is particularly important in the scripting environment. In your scripts, you
should always test for data quality especially before doing calculations.
During the failover
transition period that a Standby AppEngine becomes
Active, the value and quality for attributes subscribed for by InTouch views
may appear as “Comm?” and “Bad”, respectively. This
condition is resolved when subscriptions are rebuilt on the new Active AppEngine and the failover process is complete. This
problem typically occurs on setups in which the DIObject
data source is located on a remote computer from the original Active AppEngine.
An INVALID HRESULT LINE 76 FILE error message is logged to the ArchestrA Logger when you do the following steps:
The product functions normally
when you successfully deploy the Backup WinPlatform and place the Primary
WinPlatform onscan.
Deploying a Primary AppEngine when the Backup AppEngine’s
node is unavailable (for example, the computer is powered down) results in an
incomplete deployment. To ensure successful deployment of the Primary AppEngine, do the following:
1. Verify that the Engine Failure Timeout option
on the Primary AppEngine’s editor is at least 10,000
ms (the default value).
2. Unassign the Backup AppEngine from
its WinPlatform.
3. Deploy the Primary AppEngine.
If you have already tried
to deploy the Primary AppEngine, do the following first
before doing the procedure above:
1. Redeploy the Primary AppEngine,
selecting the Mark as Deploy check box on the Deploy dialog box.
2. Undeploy the Primary AppEngine.
You cannot redeploy a Primary AppEngine with pending updates if the Backup AppEngine is undeployed and you do not include it in the Primary AppEngine redeployment. If you try to do so, redeployment fails. The workaround is to undeploy the Primary AppEngine first and then deploy it again.
On computers with
multiple NICs (3 or more) and hosting a redundant
pair of AppEngines, if you shut down the computers,
unplug the primary and RMC network cables between the two computers, and
restart the computers, the two AppEngines are unable
to arbitrate their Active/Standby status and result in an Active/Active scenario.
To avoid this problem, do not use DNS to assign flexible IP addressing on
multiple NIC computers; rather, use fixed IP addresses.
If you re-import a
new version of a script function library used by a redundant pair of AppEngines, you must restart the AppEngines
for the new script function library to take effect. When you shut down and then
restart the redundant pair of AppEngines, though, the
new version of script function library is not loaded into system memory on the
Backup computer. To work around this problem, make the Primary AppEngine the Active partner and then redeploy the Backup AppEngine.
Failover by
shutting down a WinPlatform or AppEngine from the SMC
triggers a new alarm in the Alarm Summary. This issue does not occur when using
the ForceFailoverCmd attribute or by disconnecting
the primary network cable.
Consider a scenario in which one or more WinPlatforms host objects with subscriptions to attributes on another WinPlatform, P1. For example, P1 may have been enabled as an InTouch Alarm Provider or FieldReference objects on the other WinPlatform(s) read the ScanState of an object on P1. If WinPlatform P1 is shutdown, network traffic increases due to additional communications from the other WinPlatform(s) trying to reestablish lost subscription data. The additional communication continues until P1 is started up again.
All nodes in your Galaxy must be able to communicate with each other by using both IP address and node name in the Network Address option of the WinPlatform’s editor. This is true no matter which type of Network Address you choose to use. For example, assume two nodes in your Galaxy (host name: NodeA, IP address: 10.2.69.1; host name: NodeB, IP address: 10.2.69.2). NodeA must be able to ping NodeB with both “NodeB” and “10.2.69.2”. The reverse must also be true for NodeB pinging NodeA. Failure in either case, for example, may result in the following: you may not be able to connect to a remote Galaxy Repository node from the IDE or deployment operations may fail.
If you are using Industrial Application Server on a network with slow links between your Operator View nodes and your domain controller, and you use OS Group based security in ArchestrA, you may experience long log in times. For example, an operator logging in to an InTouch View may wait 30 seconds or more before being able to operate the View. To mitigate this problem, either provide additional domain controllers or configure the Login Timeout and Role Update interval options in ArchestrA security.
Anti-virus Software
Issues
Anti-virus software should not process files in the following ArchestrA folders:
C:\Program Files\ArchestrA\Framework\Bin\CheckPointer
C:\Program Files\ArchestrA\Framework\Bin\GalaxyData
C:\Program Files\ArchestrA\Framework\Bin\GlobalDataCache
C:\Program Files\ArchestrA\Framework\Bin\Cache
C:\Documents and Settings\All Users\Application Data\ArchestrA (default setting, specified on WinPlatform editor’s General page, History store forward directory option)
If the Anti-virus software does so, this can result in slow performance as deploys are performed.
ArchestrA relies on Regional Settings when dealing with numerical attributes that contain decimals (a period in English language locales, a comma in some non-English locales). InTouch uses only periods as decimal markers. Therefore, writing numerical attributes from InTouch to an ArchestrA Galaxy using a non-English Regional Setting fails. Nodes in your application must be configured with the Decimal Symbol for Numbers in the Regional Settings set to a period rather than a comma.
In order to show the
security classification of an object attribute, a client must parse the
classification from an “MxSecurity” prefix. The
classification that is received is in the following format: MxSecurityXXXX,
in which XXXX is the security classification.
ArchestrA mistakenly
validates as OK an Alarm Extension that contains a string array in the Alarm
Message Attribute option (attribute: DescAttrName).
This option does not support string arrays. If you do enter a string array in
this option, the Alarm Extension does not function properly in the runtime
environment.
Ensure that you do not inadvertently add leading or trailing spaces when typing any reference in an object’s editor. References with such spaces fail to bind and their quality is Bad.
If any nodes in your ArchestrA environment have multiple NICs, be aware that proper configuration of those computers is essential to successful communication between ArchestrA nodes. Information about configuring multiple NIC computers is included in the Introduction and ArchestrA Redundancy chapters of the IDE documentation (IDE.pdf). See the “Multiple NIC Computers” section of the Introduction chapter and the Important! notes in the “Configuring Redundancy in the WinPlatform” section of the ArchestrA Redundancy chapter.
If two InTouch WindowViewer users simultaneously log in on two different ArchestrA nodes, attribute writes to either node may not succeed until after the nodes synchronize user information. This problem most likely occurs on slow and intermittent networks.
For the AnalogDevice, validation for alarm sub-states (like LoLo, Lo, Hi, HiHi) does not occur during configuration. For example, assume you configure the LoLo alarm limit of an object to be 100 and do not enable the Lo alarm. The default value of the Lo alarm (zero) is used, but this invalid relationship between the LoLo and Lo alarm limits is not validated when you save the object’s configuration. In the runtime environment, though, an attempt to change the 100 limit for the LoLo alarm causes the following Error to appear:
“Operational error: Attempting to set invalid limit value. The limit value must maintain the condition: HiHi.Limit > Hi.Limit > Lo.Limit > LoLo.Limit.”
Ensure that these relationships are maintained when configuring alarm sub-states in objects.
Assume a scenario in which an instance is created from a derived template of the $UserDefined base template. If you extend the derived template’s functionality with a non-array UDA, the change is propagated to the instance object. If you reverse the configuration of the derived template by setting the UDA to be an array, the change is not propagated to the instance.
This error message may occur in one of the two following scenarios:
· If the time settings on the Galaxy node and a remote node are significantly different (several hours). In this case, if a user on the remote node tries to connect to the Galaxy Repository through the IDE, the error message appears.
Extending a CustomStruct attribute with an output extension that writes to another CustomStruct attribute does not write to the output CustomStruct attribute.
The PV of the $String FieldReference object has a data type of string, and therefore does not read strings greater than 1024 characters. If an input value exceeds 1024 characters, the PV keeps its current value, quality remains Good, and the PV.Input.ReadStatus property does not show any problem.
In this release of Industrial Application Server, a string UDA can read values greater than 1024 characters. It should not do so. This will be corrected in a future release. Do not use string UDAs for this purpose.
The InTouch security logon panel does not always reflect the new security mode (for example, OS Group Based), which appears in the title bar. This occurs if the user changes to several different security settings. This behavior occurs on computers that run different operating systems.
An IAS node on a separate network from the other IAS nodes and InSQL 9.x does not time sync with the InSQL node.
Create or edit a UDA, specifying that it is an array. Deploy the object. Now change the UDA to a non-array. Redeploy. You see errors in the Redeploy dialog box for the UDA. The UDA is still shown as an array. No errors about this appear in the error Logger.
To avoid this situation, if you need to change an existing array UDA to a non-array UDA, undeploy the object, make the changes, and then redeploy the object.
Deploying objects in a system that has a large number of Areas (>1000) can cause the SQL Server tempdb database on the InSQL node to grow to consume a sizable amount of disk space on the historian system. This disk space can be released through a manual procedure on the InSQL node. Please contact Wonderware Technical Support for details.
Importing the new $UserDefined.3 object into an IAS 2.0 or IAS 2.0 Patch 01 galaxy that has deployed instances of the UserDefined object fails with errors in the logger. To work around this issue, undeploy all instances of the UserDefined object before importing the $UserDefined.3 object.
During configuration time, if you select Enable I/O scaling, you cannot configure Enable limit alarms and Enable target deviation alarms out of EUMin and EUMax range. This validation check does not occur during runtime.
In an Analog Field Attribute with Enable rate of change alarms selected, the ROC.IncreasingHi alarm is intermittently raised on Deploy or Redeploy.
State labels do not appear in the ActiveFactory Trend.
The Field Attribute name can be up to 200 characters.
Multiple bitwise Internal sets that are sent within a scan to
an Output only Integer Field Attribute, will be bitwise set into the Field
Attribute even if the value is outside the Scaling EngUnits
Min and Max limits. Although the value set to the Field Attribute may be out of
the EngUnits range, the value written out to the
field is clamped to the RawMin or RawMax.
When multiple bitwise Supervisory or User sets are sent within a scan to an InputOutput Integer Field Attribute, only the last bitwise set is written out.
When bitwise sets are applied to an Integer Field Attribute
with an Output configured, the value set to the Output will be scaled to the Raw value range. If the scaled value is out of the Raw value range, the value written out to the field is
clamped to the RawMin or RawMax.
In IAS 2.0 and IAS 2.0 Patch 01, the Deviation, ROC, and Boolean alarms are re-raised after an engine failover. To work around this problem, contact Wonderware Technical Support.
The following table describes changes to the documentation.
Item |
Location of Change |
Existing Documentation |