Building Automation Knowledge Base

Issue After setting up a UNC as a BBMD the other UNC's on that subnet are getting "network mismatch" errors in their standard output windows. Product Line TAC IA Series Environment R2 UNC's with BACnet jar's earlier than 535a. Cause This type of error is indicative of a routing loop on the BACnet internetwork.  UNC's with a BACnet jar prior to 535a when set up as a BBMD will broadcast on their own subnet as well as to the other BBMD's and this can cause that also. Resolution Install BACnet jar 535a or later in the UNC that is the BBMD to see if this corrects this issue.  If the issue persists investigate the BACnet internetwork for a router loop. 

Issue How to configure a G3 station as a BACnet BBMD Product Line TAC IA Series Environment I/A Series G3 version 3.3.22 and later Cause Documentation Resolution In the G3 Station, expand the BacnetNetwork, expand Bacnet Comm, expand Network, and then double-click IP Port.  The IP Port Property Sheet opens (see below)   In the IP Device Type drop-down, select Bbmd.  Click Save The station is now configured as a BACnet BBMD

Issue On a large system with more than 32 subnets, is their a workaround for the 32 BBMD limit on a Continuum system? Product Line Andover Continuum Environment BBMD Foreign Device Registration. Subnets Cause It is not possible to configure more than 32 BBMDs within a BCX40xx. Resolution While foreign device registration can be used as a workaround when controllers must be installed on a subnet but no additional BBMDs can be configured, foreign devices present their own limitations. Let's say we have two subnets.  Subnet A has a BCX4040 that is configured as a BBMD.  On subnet B there are two BCX4040 controllers.  One controller on subnet B is registered as a Foreign device with the BBMD on subnet A Foreign Device Table registration supports up to 1024 devices on the workstation (v1.94 and later).

Issue How many BBMDs and Foreign Devices can be registered on a CyberStation and a bCX4040?   Product Line Andover Continuum Environment BACnet bCX4040 CyberStation BBMD Foreign Device  Cause Concerned about large BACnet sites and exceeding a limitation on the number of BBMDs and Foreign Devices that can be registered.  Resolution Maximum number of BBMDs The maximum number of BBMDs that can be registered on a bCX4040 or on a CyberStation is 32.   Maximum number of Foreign Devices (FDs) Continuum v1.93 or earlier The maximum number of Foreign Devices (FDs) that can be registered on a bCX4040 or on a CyberStation is 64. Continuum v1.94 or later The maximum number of Foreign Devices (FDs) that can be registered on a CyberStation is 1024 The maximum number of Foreign Devices (FDs) that can be registered on a bCX4040 is 64.

Issue Configuration of multiple BACnet BBMD's over several subnets Product Line Andover Continuum, EcoStruxure Building Operation Environment Bacnet BBMD Cause Bacnet Broadcast communication through IP router Resolution Configuration of multiple BBMD's over several subnets: INCORRECT METHOD For example if there are 4 subnets on the same B/IP (BACnet IP Network): Subnet 1: Workstation setup as a BBMD with an entry in the Broadcast Distribution Table for each BCX Subnet 2: BCX setup as a BBMD with an entry in the Broadcast Distribution Table for itself and the Workstations BBMD Only. Subnet 3: BCX setup as a BBMD with an entry in the Broadcast Distribution Table for itself and the Workstations BBMD Only. Subnet 4: BCX setup as a BBMD with an entry in the Broadcast Distribution Table for itself and the Workstations BBMD Only. The BCX's do not reference any points from each other. Since there is no data sharing between the BBMD's, one might think that in each BCX's Broadcast Distribution Table that there would only be a need to add one entry for the Workstations BBMD. Sounds like with this setup we'll reduce traffic by only adding the one entry in each BCX's BBMD Broadcast Distribution Table. The above scenario could in fact slow down a network and cause unnecessary traffic.   The supported method is that every BBMD on a B/IP network have the same routing tables(better known as the Broadcast Distribution Table) as in the below configuration. CORRECT METHOD For example there are 4 subnets on the same B/IP (BACnet IP Network). Subnet 1: WS BBMD with an entry in the Broadcast Distribution Table for itself and all BCX's setup as BBMD on the same B/IP. Subnet 2: BBMD with an entry for all BBMD's(including self) in the Broadcast Distribution Table on the same B/IP. Subnet 3: BBMD with an entry for all BBMD's(including self) in the Broadcast Distribution Table on the same B/IP. Subnet 4: BBMD with an entry for all BBMD's(including self) in the Broadcast Distribution Table on the same B/IP. The entry below can be found in the BACnet Standards 2004:   J.4.3 BBMD Concept Each IP subnet that is part of a B/IP network comprised of two or more subnets shall have one, and only one, BBMD. Each BBMD shall possess a table called a Broadcast Distribution Table (BDT) which shall be the same in every BBMD in a given B/IP network.

Issue Cannot open Calendars on several BACnet Networks. Product Line Andover Continuum Environment Continuum 1.9X BCX 4040 B38XX controllers Cause BBMD Sub Net Mask Configuration. Resolution There are two configuration methods allowed in a BBMD Subnet mask. One hop configuration and a Two hop configuration. See Understanding BACnet BBMD Distribution Mask for further details

Issue When configuring a BBMD what is the Broadcast Distribution Mask and what are the valid settings for the field. Continuum EcoStruxure Building Operation Product Line EcoStruxure Building Operation, Continuum Environment Building Operation Automation Server Building Operation Enterprise Server Continuum CyberStation bCX4xxx Cause Documentation Resolution When configuring a BBMD the BACnet protocol defines two methods of propagating the broadcasts messages to other networks. The two methods are referred to as One-Hop also known as a Directed Broadcast and Two-Hop.   Note: In this article the subnet mask is shown using Classless Inter-Domain Routing(CIDR) notation. An IPV4 address is 32 bits long, CIDR notation defines the number of bits used for the subnet mask. The most commonly used are: /8 or 255.0.0.0 /16 or 255.255.0.0 /24 or 255.255.255.0 /32 or 255.255.255.255 One-Hop With the One-Hop method the local BBMD transfers the message directly to the remote devices via the remote router. In this case the IP routers have to support the transfer of IP broadcast messages to remote IP subnets Figure 1 Figure 1 above illustrates the one-hop method, when the bCX1 which has been configured to take the role of BBMD for network 172.16.0.0/16 detects broadcast messages such as WHO-IS, I-AM etc, it sends a message directly to the broadcast address of the 192.168.1.0/24 subnet, it is then up to Router2 to send the directed broadcast to all devices on the subnet. Any device on the subnet that is acting as a router to other networks will then also re-broadcast the message on those networks. (i.e a bCX hosting an MSTP network on COMM2)   Two-Hop With the Two-Hop procedure a BBMD device transfers broadcast messages to a further BBMD in the remote IP subnet, which then distributes the messages in its local network as IP broadcast messages for its IP subnet. Any device on the subnet that is acting as a router to other networks will then also re-broadcast the message on those networks.(i.e a bCX hosting an MSTP network on COMM2) Figure 2 Figure 2  above illustrates the two-hop method, when the bCX1 which has been configured to take the role of BBMD for network 172.16.0.0/16 detects broadcast messages such as WHO-IS, I-AM etc, it sends a unicast message to each BBMD device on its table, it is then up to each BBMD device on each subnet to send a broadcast that will be heard by all devices on the subnet. Which distribution mask should be used? As mentioned earlier, in order for the one-hop method to work, the router on the remote subnet MUST be configured to process incoming messages sent to the broadcast IP address, the two-hop method on the other hand is guaranteed to work, so when in doubt use the two-hop method. Most sites use two-hop distribution for the reason stated above,, the only obvious disadvantage of the two-hop method is that it is a little more verbose than the one-hop method since the message appears twice on the subnet, once as a unicast sent to the BBMD device and then again as a broadcast sent by the BBMD device. In Building Operation and Continuum the distribution mask defaults to the two-hop method (/32 or 255.255.255.255) How to configure the distribution mask For one-hop enter the subnet mask of the remote network in the distribution mask, the system will apply the mask to the IP address to obtain the broadcast address for the remote network. For two-hop enter 255.255.255.255, this tells the system not to do a direct broadcast and instead do a unicast to the remote BBMD device. Sample distribution mask In the BBMD table of a bCX shown in the screen shot above, BDT #1 and #3 use the two-hop method while BDT entry #2 uses the one-hop method.

Issue Communications failure due to transaction timeout when reloading b3 device. b3's version number as seen from the Infinity side reads 0.0000 Product Line Andover Continuum Environment CyberStation bCX4040 b3 Cause A b3 BACnet controller is found to exist in the BACnet folder under an MSTP Network which is different from the Network specified in the bCX's comm2 Port. Resolution Delete the MSTP network under the BACnet folder and perform a 'Find new BACnet devices'.  A new MSTP network should be created with the right network ID.

Issue What is the BACnet attribute COV Increment and how to use it Product Line Andover Continuum Environment BACnet Cause The COV Increment attribute on a BACnet point is equivalent to the Threshold attribute of standard Continuum Infinity inputs. When a Continuum BACnet point is subscribed to by other BACnet devices using COV (Change of Value), its value must change by at least that COV Increment value before a new Value Notification will be sent out. Using this attribute will decrease network traffic especially on the MSTP network when subscribing to frequently changing values.  Resolution This CovIncrement attribute can be set in the BACnet point's Editor on the Basic Alarms Tab.   If you have a CyberStation prior to Version 1.8, setting this value from the point editor will not stay persistent in the controller after the controller is reloaded, it will default to 0. In order to keep the value after a reload, you will need to write a PE program to set the desired value after a reload. The program should be Autostart and Fallthru. A sample of the program is below: Line SetCov Writeproperty("Full Bacnet Path to Point", CovIncrement, "Desired Value") Stop If you have several points that need the CovIncrement set, you would just add the additional lines into this program.   In 1.8 or greater, this program is not needed as these attributes get sent to controller on a reload. Suitable initial CovIncrements still need to be added to each BACnet point however as they default to 0.

Issue There is a need to know if Continuum and/or EcoStruxure BACnet support NAT. Product Line Andover Continuum, EcoStruxure Building Operation Environment Continuum EcoStruxure BACnet BBMD NAT Cause Many sites use NAT for security purposes.  Resolution NAT traversal is not supported in BBMD implementations (Continuum or EcoStruxure). It is an optional feature that has not been implemented. Requests to add this feature should be directed to Product Marketing to include on the backlog for prioritization. Related Continuum article: Can Continuum operate over a NAT Ethernet connection? Ashrae information: here

Issue SmartStruxure comes preloaded with LonMark's resource files. It is not documented what version of the LonMark Resource Files are build into each version of SmartStruxure. Product Line EcoStruxure Building Operation Environment LonWorks LonMark Cause This is missing from the documentation Resolution SmartStruxure Version Resource Files Version 1.2 13.02 1.3 13.02 1.4 13.12 1.5 14.00 1.6 and later 14.00

Issue Site implements Entry/Egress anti passback access sequence and uses a program that runs overnight to 'forgive' all anti-pass back violations. Product Line Andover Continuum Environment Continuum CyberStation Cause Documentation Resolution The following PE code is provided as a sample of how to have a PE program set the current area in ALL personnel objects. NOTE: The code is given for illustration ONLY and is not intended to be taken as a complete solution, engineers using this sample code must be sufficiently proficient in Plain English so as to make the necessary changes and modifications to the code to adapt it to their particular application.                                         'This WorkStation program sets the current area of ALL personnel objects found in 'theFolder' to "theArea" 'Program's flow type is LOOPING. Number RESULT Object thePerson, theArea, theFolder Line INIT 'ENTER THE FULL PATH TO THE FOLDER THAT CONTAINS THE PERSONNEL OBJECTS   theFolder = Root\CredentialHolders 'ENTER THE FULL PATH TO THE AREA   theArea = Root\MyAreas\TestArea2   Count = 0 'OPEN THE LIST OF PERSONNEL OBJECTS IN THE FOLDER   RESULT = OpenList("Personnel", thePerson, theFolder)   If RESULT = Success then Goto COUNT_THEM Else Goto CLOSE_LIST 'COUNT THE OBJECTS SO WE KNOW WHEN WE HAVE GONE THRU THE LIST EVEN IF WE ENCOUNTER AN ERROR Line COUNT_THEM   While GetObject(thePerson) is Success     Count = Count + 1   Endwhile   CloseList(thePerson)   Goto RE_OPEN   Line RE_OPEN   RESULT = OpenList("Personnel", thePerson, theFolder)   If RESULT = Success then Goto SET_AREA Else Goto CLOSE_LIST Line SET_AREA 'GET THE NEXT PERSONNEL OBJECT IN THE LIST  While GetObject(thePerson) is Success 'SET THE CURRENT AREA   thePerson Value = theArea   'thePerson Value = UnKnown   'to set all to UnKnown comment the line above and un-comment this line   Count = Count - 1  Endwhile  Goto CLOSE_LIST   Line CLOSE_LIST   CloseList(thePerson)   Stop Line E   If Count > 0 then 'WE ARE NOT AT THEN END OF THE LIST, HERE PROBABLY DUE TO SOME ERROR SETTING THE AREA, SKIP THIS OBJECT AND GO TO THE NEXT     Goto SET_AREA   Else 'END OF LIST, WE ARE DONE     CloseList(thePerson)     Stop   Endif

Issue Changes to COV and ReadProperty in bCX from v4.500048 Product Line Andover Continuum Environment bCX b3 Controllers Cause New feature to allow user flexibility for using Forced Polling and not COV Resolution From bCX version v4.500048, all Controllers will have the option of forcing a poll of points from other Continuum and third party BACnet controllers. This new feature: allows programmers to decide which data exchange method to use by the syntax in which they write their data-sharing programs. provides a method for forcing point-to-point polling from a device that does not support COV and sends an invalid error message when the subscription is attempted. This previously caused Continuum to reject polling, resulting in a failure to share data. allows programmers to decide to poll a point only when needed, even if the device supports COV subscriptions. This ensures that data from a device is only sent when necessary instead of always being sent on a “change of value”. The following Plain English formats force polling: "Local point = Infinity Remote Point Value" "Local point = BACnet Remote point Value" "Local point = ReadProperty(BACnet Remote point Value)" The second argument, Value, ensures that polling is commanded – now for BACnet, as well as Continuum/Infinity. (The third example shows the use of Plain English’s ReadProperty system function.) If the Value argument, shown in these examples, is left out, Continuum BACnet will operate as it always has – always attempting a COV subscription and only falling back to the polling method if the target device sends the appropriate COV subscription failure message. Caution: If you have a device that does not support COV subscriptions or you have chosen to use the polling syntax, you must be careful how you write your Plain English programs. A program that runs every scan and polls a point from another controller will cause major problems. For instance, if your scan is running at 100 milliseconds and you try to poll any BACnet device at that rate, you will most likely knock the device offline. It will also cause the program to disable and seriously slow down the scan in your controller that is requesting the data (since the scanner waits for an answer to a poll before continuing). You should also use a line E in your programs, since any communication problem to the target device will cause the polling program to disable.

Issue An application will not open in WorkPlace Tech Editor with the following error:   WorkPlace Tech Editor returned errorOperation aborted (Exception from HRESULT: 0x80004004 (E_ABORT))   Product Line EcoStruxure Building Operation Environment Building Operation Workstation (WorkPlace Tech Editor) Visio 2013, 2016, and 2019   Cause Visio crash causes the Add-In for WorkPlace Tech Editors fluent ribbon to become disabled Resolution To re-enable the WorkPlace Tech Editor fluent ribbon bar, perform the following steps: Open Visio.  NOTE: Open Visio, not WorkPlace Tech Editor Select Options from the File menu. From the menu in the Visio Options window, select Add-ins At the bottom of the window, you will see a field that says 'Manage:' with a selection box after it, select 'Disabled Items' and then click the [Go] button. In the 'Disabled Items' window, select 'SE.WorkStation.IA.VisioRibbon' then click the [Enable] button. Close the open Visio windows and sub-windows. After closing Visio, the WorkPlace Tech Editor fluent ribbon bar will be available in WorkPlace Tech Editor

Issue Too much LON communications traffic can cause LON communications errors. Product Line TAC IA Series, EcoStruxure Building Operation Environment I/A Series MNL 800 Building Operation Automation Server Premium LON Network  I/A Series R2, AX, N4 Jace LON Network Cause Sending output LON SNVT values too frequently or more often than reasonably needed. Resolution There are a series of NCI (configuration) objects that control the transmission and reception of these data values on the LON network. One of four NCI objects (nciSndHeartBeatA, B, C, or D) is assigned to each NVO object in the application program. This value regulates the periodic LON transmission queuing of the NVO object values for each binding. One of four NCI  objects (nciRcvHeartBeatA, B, C, or D) is assigned to each NVI object in the application program. This value regulates the fallback timeout for each NVI object value. If a value is not received via the binding in this time period, the NVI object value will revert to its application-defined default value. In an MNL-800 controller, there are two additional NCI objects that are used to regulate the flow of data onto the LON network. The nciMinOutTime object is used to throttle the transmission of all data values queued to be sent via bindings over the LON network. This NCI object sets the transmission time interval between each NVO value being sent on the network. The important fact here is that the product of the number of bound NVO values times this interval (nciMinOutTime) must be LESS than the shortest nciSndHeartBeatX (X=A, B, C or D) interval. For example, if the Send HeartBeat is set to the default of 120 seconds and there are 60 NVO SNVTs bound from this controller, the nciMinOutTime must be set to a value of fewer than 2 seconds.  The recommendation for setting this SNVT value one half the shortest nciSntHrtBr divided by the number of NVO SNVTs bound from the controller.   The nciMinPropTm object is used to regulate the transmission queuing of changing bound NVO object values. This timer acts as a time-based filter to prevent a rapidly changing NVO value, such as a constantly changing temperature, from overwhelming the communication capacity of the LON network. Typically this value is set to a value of 1 to 5 seconds. Additional information can be found in Chapter 6 of the WorkPlace Tech Tool 4.0 Engineering Guide (F-27254). BACKGROUND: This section is provided to give the reader a basic understanding of bound LON communications, as applied to LON controllers in general. Data communications across a LON network is accomplished by using a LON Network management tool to create a binding between a device with a data source (NVO object) and one or more devices with data receivers (NVI objects) in other LON controllers. The device containing the data source will send the bound data value at a periodic rate, as defined by the associated SendHeartBeat NCI object, and whenever the value of that object changes. If the NVI object does not receive a value from the sender (bound NVO object) within a specified time period, as defined by the associated ReceiveHeartBeat NCI object, the NVI object will use a local fallback value until such time as it receives a new value via the binding.

Issue The Universal Input real-time values on a MNL-800 (or Microzone II) are fluctuating unrealistically.  They rise, peak, drop, valley and then start over again very quickly.  It is similar to an AC wave pattern. It is happening with all the inputs no matter their configuration. Product Line TAC IA Series Environment MNL-800 Microzone II (Network 8000) Cause When an actuator is interfaced with a MNL-800 (or MicroZone II), and the UI cable shields are connected to earth ground (the SHLD terminal), the shield has a common mode AC potential that couples to the UI input. Only a very small signal (UI shield to signal common) is needed to result in objectionable UI signal fluctuation. Resolution Connect all drain shields on UI inputs to UI common (C terminal). Read additional solutions and a more descriptive cause in Technical Bulletin 01-36.

Issue How to search for points, names, or descriptions in EcoStruxure Building Operation Product Line EcoStruxure Building Operation Environment Building Operation WorkStation Cause The EcoStruxure Building Operation search feature can be performed on an entire database or on individual controllers / folders. This can help reduce the amount of time it takes to locate multiple points when creating mirrored points. Resolution How to search for specific names in EcoStruxure Building Operation Right click on the folder, object, or controller you would like to search in and click "Search. Note: When the search is narrowed down to specific areas of the database it will be able to complete the search faster than performing a full database search." In the search field type in the name that you are looking for. In this example the search is for snvt*. This will locate all points underneath the controller that begin with snvt in the point name. After a few seconds, depending on the size of the search area, the results will be displayed in the window to the right. In order to filter the search further, you can set different search parameters under the search field.

Issue What image formats are supported in a TGML graphic? Product Line EcoStruxure Building Operation Environment EcoStruxure ES, AS Vista Graphic Editor Cause Clarification of supported graphic formats required. Resolution When inserting a picture in a TGML graphic you can select the image file from the following file types: Bitmap, GIF, ICO, JPEG, Metafiles, PNG, TIFF  

Issue There are a number of reasons that RPC error occur on Cyberstation and troubleshooting steps are needed Product Line Andover Continuum Environment Continuum Webclient Windows XP Window Vista Windows 7 Windows Server 2003 Windows Server 2008 Cause RPC (Remote Procedure Call) errors happen on Cyberstation when the connection to the SQL database is not available Resolution There are many reasons that RPC errors can occur. A Google search of RPC errors will indicate that many applications have similar issues with these errors. An RPC error on a Cyberstation means that it has lost connection to the database for some reason and you will need to restart Cyberstation in order to recover. Below is a list of things to check when dealing with RPC errors on CyberStation.   Multiple Cybers Does the issue occur on multiple Cyberstations? Database Does the issue follow the database? i.e. load the database onto a test server or try a test blank database. Were some new objects recently added to the database when this started happening? Does reverting to a backup stop the errors? Increase Virtual Memory Right Click on My Computer> Properties> Advanced System Settings> Under Performance select the Settings button> Advanced tab> Under Virtual Memory click the Change button Recommended is 4096 Network disruption or slowness Networking  -  Verify all aspects of your networked environment.  This includes, but not limited to, slow networks, possible problems with your network interface card (NIC), cabling issues, switch/router issues, busy networks.  As you can see from the above definition of what RPC is it can be heavily reliant on the stability of the network and all it’s interfaces. Continuum does not support DHCP Multiple NIC cards – ensure there is only 1 NIC card enabled. No all NIC cards are created equal. Connections issue can be caused by an integrated NIC card of low quality. There can be settings within the NIC card itself that could affect consistent communication. Check the manufacturers website for details. Verify that there is no duplication of IP addresses on the network. Pull a cable and ping the address, if you still get a response there are 2 devices with the same IP address. WINS -  Verify your WINS settings and make sure there are no issues with the WINS Server.  This is known by Microsoft to cause RPC errors. DNS - Verify your DNS settings and make surer there are no issues with the DNS server.  This also is known by Microsoft to cause RPC errors. Verify that  the "TCP/IP NetBIOS Helper" service is running. Check and see if “File and Print Sharing” is enabled.  There are references in a lot of online articles, on RPC, problems that turning on “File and Print sharing” resolves RPC errors. Ping and SQL Is the SQL database still available to Cyberstation? From a DOS prompt> Ping IP (SQL Server) –t (continuous) Ex. c:\> ping 192.168.1.5 –t Note the number of responses as well as the time in milliseconds to respond. A local SQL server would usually be under 100ms. Check resources on the SQL Server (CPU, I/O Read/Write, memory usage, etc.) Windows Server 2003 Network card setting (This setting isn’t available in Windows Server 2008) Shut Continuum down on the machine having the problem. Start>Settings>Network Connections>Local Area Connections>Right Click and select Properties> Click on File and Printer Sharing for Microsoft Networks>Click Properties button> Select 'Maximize data throughput for network applications'. (Default is Maximize data throughput for File Sharing) Ok through all the screens and start Continuum and see if this makes a difference. XP/Vista/Windows 7 Start> Settings> Network Connections> Local Area Connection> Properties> Advanced tab> Windows Firewall - Click the Setting button> Click the exceptions tab Verify that File and Print Sharing is selected. File and printer sharing are disabled by default with XP SP2. There is no network throughput option in XP, Vista or Windows 7 like there is on Windows Server 2003. Windows Event Viewer Take a look at the Windows Event Viewer Application Log (Start> Run type ‘eventvwr’ or Launch Windows Control Panel> Administrative Tools> Event Viewer Examine the events and look for RPC error indications in the details. There may be a pattern of which executable is causing the errors. You may also see informational events that indicated momentary disconnects. A large number of these can indicate an issue with the Cyberstation being able to connect to the database consistently in a timely manner. Monitoring tools Run an acctrace.exe (C:\Program Files\Continuum) and a wireshark capture at the time of the RPC error to see the Continuum and network processing going on at the Cyberstation. Windows Process Monitor may also be helpful in seeing which process is having trouble. http://technet.microsoft.com/en-us/sysinternals/bb896645 Check the system for SpyWare and/or Viruses. Workgroup versus Domain connection An excerpt from the Continuum System Admin manual In a LAN Configuration, the Continuum CyberStation must be in a “Microsoft Domain “ type installation. This is because the Continuum CyberStation will use Named Pipes Interprocess Communication System (IPC). This is accomplished with Windows remote procedure call (RPC). In a Microsoft Domain model installation, once the session is established between the Continuum Cyberstations Client(s) and the SQL server it is a persistent connection once authentication is established.   In a Workgroup installation, the database connectivity session is re-authenticated and re-established for each database transaction. This results in much slower performance and an inconsistent connection. If you are running a LAN system using Windows 2003 server SP2/R2 and SQL 2005 try disabling the TCP Chimney Offload feature by following the below. Disable the TCPchimney setting on the SQL server and application server Windows 2003 SP2 comes inbuilt with a new network scalability pack which is supposed to accelerate the Windows networking stack. But this setting may not be compatible with all the network cards and it could pose connectivity issues as well. For more information you can refer to http://support.microsoft.com/kb/912222 http://support.microsoft.com/kb/942861   To disable this feature please take the following steps on both the SQL Server and application server.   Open up a command prompt on the machine. Run the following command: "Netsh int ip set chimney DISABLED" Change the values of the following registry keys to 0 a. HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters\ EnableRSS   HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters\ EnableTCPA Reboot the machine after the above changes. ==========================================================   Disable SynAttackProtect on SQL server: Windows 2003 Sp1 and Sp2 introduces this new feature which monitors processes getting high number of connections and if it detects that too many connections are being made, the connections will be rejected. Click Start, click Run, type regedit , and then click OK. Locate and then click the following registry key: HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters On the Edit menu, point to New, and then click DWORD Value. Type SynAttackProtect , and then press ENTER. On the Edit menu, click Modify. In the Value data box, type 00000000 . Click OK. Quit Registry Editor. Sound Card In some cases it was found that the built in sound card is causing the problem. If your not using WAV files for alarms then try disabling the integrated sound card in the BIOS.  If WAV files are being played then try disabling the integrated sound card and put in a PCI sound card. There have also been problems with OEM versions of the Operating system of the client machines.  A lot of vendors, like Dell, add additional applications to the systems they ship to assist them if they need to remote into the computer for troubleshooting purposes.  It is strongly recommended that these OEM versions of the Operating System is removed and install the system from a genuine Microsoft installation CD. Go to Control Panel-> Administrative Tools -> services, click with right mouse button Remote Procedure Call -> click properties. remote procedure call properties window is now open, go to recovery and you´ll see: First failure, second failure, subsequent failures -> choose : restart the service. In this setup the system will automaticaly restart RPC when something shuts it down. Check all your Firewall settings.  If possible turn off the firewall for testing and see if the problem goes away. Check to make sure there are no ports being blocked that the RPC Server uses.  See KB 908472 (http://support.microsoft.com/kb/908472) details the steps for specifying the RPC port range.  Also, be aware of the consequences of reserving the RPC ports. This has the potential to break other applications that assign ports in this range. KB 812873 discusses this, too. http://support.microsoft.com/kb/812873 Make sure all Peripheral devices like Network cards, Video cards, Sound Cards, etc. have the latest drivers for them installed. http://support.microsoft.com/kb/913384 Control Panel > AdminTools >Services and selected the Properties for Remote Procedure Call (RPC)  In the Log On tab, they changed the Log on as: to the Local System Account. It had been "This Account" NT Authority\Network Service. Is an EventNotification configured to send a page on alarm? If a PC running Continuum did not have a modem but on startup Continuum was trying to send an alarm with a page then accdataservice was crashing causing an RPC. Microsoft KB177446 article that describes in depth RPC troubleshooting procedures http://support.microsoft.com/kb/177446

Issue Continuum crashes with RPC error "The RPC server is unavailable" shortly after launching. RPC Server Error Product Line Andover Continuum Environment Cyberstation Cause Possible causes: Virtual memory, make sure virtual memory is set to recommended value. (see virtual memory requirement in Cyberstation installation document) Remote Procedure Call (RPC) locator service not started (Should be configured for automatic start) A resource needed by Continuum is not available. For example if an EventNotification is configured to page on alarm and there is no modem installed in the workstation then if an alarm is active on start up Continuum will encounter RPC error. Resolution Set workstation virtual memory to recommended value. Configure Remote Procedure Call (RPC) for automatic start. Make sure resource needed is available or reconfigure as needed.