IoT & Connectivity

Dear Members, I am new to KEPServerEX 6 and request expert guidance to configure Modbus RTU (RS-232) data in KEPServer, which I have been unable to achieve so far. Current Challenge At present, to bring RS-232 data into Modbus TCP/IP, I am using a protocol converter. However, this setup works only after running a Python script on the historian server that continuously sends ASCII commands (video attached) to read serial data and then pushes it back to the historian. Only then does the converter provide usable values for our Digital Data Visualization. Pain Points This approach is becoming unsustainable due to the following reasons: The Python script must run 24×7 Any script failure results in complete data loss Maintenance and troubleshooting are time-consuming The solution is not reliable for long-term plant operations What I Am Looking For I want to directly configure the RS-232 / Modbus RTU device in KEPServerEX and expose the data as Modbus TCP/IP (or directly to the hist

Hello,I am experiencing an issue when browsing/importing OPC UA server tags into Kepware. Several tags that are present on the OPC UA server do not appear in Kepware after the browse/import process.Details:The missing tags are not visible in the Kepware browse tree, so I cannot import/add them to my datalogger.I have verified the tags exist and are accessible from the OPC UA server side.I also confirmed using UAExpert that the missing tags are readable, and their data type is Float (so they do not appear to be unsupported types).Could you please advise what could cause Kepware to omit certain OPC UA nodes during browsing/import, and which Kepware settings or troubleshooting steps I should review to ensure all tags are discovered and available for import?Thank you. However, when running Tag Generation, it gives warning message of data type unknownWhen try to import, the tags are shown in imported itemsOPC-UA Server has tags under folder AHU/01 such as RA_RH, RA_T, SA_RH, etc.The fi

Is there any possible way to update the "isConnected" Property (in the "Connectable" Thing Shape) from within ThingWorx?  This property/TS is part of the "RemoteThing" Template and is great for visualizing if a Thing is connected, like in Composer -> Monitoring -> Remote Things.  I'd like to extend this capability to API connections we make in ThingWorx that do not have a native ThingWorx connection or remote agent.  Essentially I have defined a "connector" Thing in ThingWorx that makes REST calls on an API via scheduler/timer; I want to utilize the "isConnected" property as an indicator of whether the API connection is "healthy."  If an API call is successful, "isConnected" would be set to 'true'; if an API call fails or has a particular error code, "isConnected" would be set to 'false.'   Is it possible to define this logic within ThingWorx, or must it be done via SDK, or some other method?

In the data logger feature, by default "Name" column is being considered as "Item ID" making difficult in database. How can i change this to "Log item Name" instead of default "Item ID"

Hi everyone,I’m sending MQTT data from Kepware IoT Gateway to ThingsBoard. Everything works fine when I publish around 250 tags in a single JSON payload. But when I increase it to 500 tags, the MQTT client in Kepware starts connecting and disconnecting repeatedly, and the message never publishes.I’m trying to figure out whether this limitation is coming from:Kepware IoT Gateway’s MQTT Client Agent, orThingsBoard’s MQTT broker / MQTT transport,or simply an MQTT payload size limit on either side.What I’m seeing:~250 tags → OK~500 tags → Kepware repeatedly connects/disconnectsNo clear error in Kepware UIThingsBoard logs show only connect/disconnectBehavior suggests a payload or packet size limitMy questions:Does ThingsBoard MQTT have a max payload or packet size limit?Does Kepware IoT Gateway restrict MQTT message size?Is this connect/disconnect loop typical when the payload is too large?Is splitting tags into smaller groups the recommended approach?Any insight from users who have handled

 Hi everyone,   I’m working on a use case where our SAP team wants to send data to ThingWorx via an HTTP API call. They will post the data as XML, and ThingWorx should receive it in a service, parse it, and store the information in a DataTable or Dataset.   However, I’m not sure how to receive the XML payload properly inside a ThingWorx service. When SAP calls my service endpoint, I don’t see any way to directly access the raw XML content — it’s either not recognized or comes as an empty input parameter.   What’s the correct approach to handle this in ThingWorx?   Specifically: How can I receive XML from an external system in a ThingWorx service? Do I need to configure a specific Content-Type header (like application/xml)?     If anyone has done SAP → ThingWorx integration with XML payloads, I’d really appreciate an example or best-practice recommendation.   Thanks in advance!   

Is there any updates on when the License Transfer functionality will be fixed on my.kepware.com? We're currently unable to Transfer Licenses due to the error and all the Support Engineer can do is tell me to reach out to sales@kepware.com ?

I need to set up an external REST client to read AuditLog and EventLog over API hopefylly anonymouslyIs it possible?

Hello everyone, I’m trying to configure a tag of type LLong (Signed 64-bit Integer) in KEPServerEX 6.18 for a Modbus TCP/IP device, but I’m running into an addressing issue. My setup:Channel: Modbus TCP/IP EthernetDevice: Custom Modbus server running on Raspberry Pi (C#)The server exposes a 64-bit signed integer, stored correctly on the Modbus side.When I attempt to add a tag with data type LLong, KEPServerEX refuses any register address I enter. I’ve tried:4000740000730007300007But I always get the same error:“Data type 'LLong' is not valid for address '400007'.” It looks like KEPServerEX does not accept any Modbus register address for this data type, and I’m not sure whether the issue is related to:register type selection (3x/4x),addressing mode (1-based / 0-based),strict syntax rules,or a limitation of the Modbus TCP driver regarding 64-bit values. My questions.1) Has anyone successfully configured an LLong (64-bit) tag in a Modbus TCP/IP channel?2) What is the c

Can someone point me to where these data are stored in Operator Advisor's equipment configuration UI? I need to find it so I can migrate it to a different location in a DB. Same for the attachments and other data on the general information page once equipment is clicked, but we can start here.  

Hi everyone, I have a checklist mashup that consists of around 6 pages, each page containing a mix of text fields, checkboxes, and dropdowns — in total, roughly 100+ input widgets whose values need to be saved in the database. My question is about best practices for handling this kind of large form in ThingWorx: Should I bind every single widget value directly to a service input parameter and call a “save” service? Or is there a better way to collect all values dynamically (for example, in JSON) and store them in one field (like metaData)? Ideally, I would like to: Save all the entered values in the database as a JSON object (for flexibility and version tracking). Load the same data back into the mashup later, automatically populating all corresponding widgets. What’s the most efficient and maintainable approach to achieve this in ThingWorx?Would you recommend using MashupFunctions / Service calls / JavaScript / JSON encode-decode logic for this? Any examples, tips, or perf

We are preparing to migrate our current Thingworx environment from version 9.7 to version 10.x. Before proceeding, I would like to gather some insights regarding the stability of the 10.x version. Has anyone encountered issues or challenges during their migration process to this version? Any specific details or experiences you can share would be greatly appreciated, as they will help us plan our migration more effectively. Thank you!

Hello Everyone,I have a ThingWorx instance hosted on an OCI (Oracle Cloud Infrastructure) server, and I would like to enable Single Sign-On (SSO) for the platform.I am not sure about the correct steps or prerequisites for configuring SSO in this setup. Could you please guide me on the following:What are the requirements for enabling SSO on a ThingWorx server hosted in OCI?Which SSO protocols are supported (e.g., SAML, OAuth, OpenID Connect)?Is there any documentation or step-by-step guide available for configuring SSO in this environment?Do I need to make any additional configurations on the OCI side (networking, certificates, identity provider integration, etc.)?Any best practices or common issues to be aware of.can we have both native login and sso enabledAny help, references, or examples would be greatly appreciated.Thank you!

Hi,  Are advance tags only in read only mode or can we write to them from the OPC client ..? 

Hi,I am using the Application Key as of now to access services and Mashup for the end user. (Adding APP key in URL)If I use the SSO moving forward, how can I access those services and Mashups now?Thanks,

Hi I upgraded ThingWorx from 9.3.x to 9.7.3 yesterday. And after the upgrade the JDBC connecting Thing to MSSQL is not connecting.I have even upgraded and installed the new JDBC extension 2.1.0 as well.  

I am trying to add 2 separate SLC 5/05 plc's in our Kepware server using Allen Bradley Ethernet driver because both of the PLCs is networked. We have also used the DF1 driver with the ethernet encp. but that did not work either. Nothing sees the PLC. We then tried adding a test SCL 5/05 PLC in our training room with a small network for a controlled environment and it still will not see the PLC. Kepware EX6-6.16.203.0

Hi everyone,   I have a DataTable with several entries. When I click an entry, a service runs and loads the images from the imagePath stored in that entry. Everything works fine normally — the Collection widget displays the correct images.   However, when I switch between entries very very fast, sometimes the images from a previous entry stay visible and do not update, even though the selected entry has changed. It looks like the Collection widget doesn’t refresh properly when service calls overlap or happen too quickly.   Has anyone experienced this behavior or found a reliable way to force-refresh the Collection widget when rapidly changing entries?   Thanks,Jamal

I'm having a problem getting Flow 10 to get configured with Foundation 10.  All the services start and I can connect to Foundation via the Flow nginx port but I get the following error when running WorkflowSubsystem.ValidateFlowSetup: Found 1 issues. [FLOW-API-01] - Failed to connect to Flow: Internal Server Error For more information, refer to the 'Troubleshooting Your Installation' topic in the Help Center. The ApplicationLog has these errors: 2025-11-21 14:29:06.092+0000 [L: ERROR] [O: c.t.s.s.f.WorkflowSubsystem] [I: ] [U: Administrator] [S: ] [P: ] [T: https-openssl-nio-8443-exec-7] Failed to get Delegated UserOAuths : Internal Server Error 2025-11-21 14:29:06.092+0000 [L: ERROR] [O: c.t.f.f.FlowDelegatedOAuthFilter] [I: ] [U: Administrator] [S: ] [P: ] [T: https-openssl-nio-8443-exec-7] Unable to initialize session attributes I don't have SSO currently enabled although I did have it on in previous attempts so I'm not sure with there is an OAuth error.

The KepServerEx is configured as OPC DA Client to connect the OPC DA Server on another machine. It get callback failure and bad data. While using OPC Quick Client connected the OPC DA Server to get data directly. It get good data. Two servers are on the same domain but running with different Log On account.

Hi everyone, I am currently designing a simplified deployment setup for our ThingWorx environment based on the standard multi-stage pipeline.     Our planned flow looks like this:   👉 Local Instances (Developers) → DEV Environment → Git (Build Pipeline) → Artifact Repository → TEST Server → Production Servers (multiple locations like China, Italy, etc.)   This structure seems efficient for our team and helps streamline deployments, but I’m unsure how to handle bug fixes in this setup. For example, if a bug is found in production, what’s the best practice for fixing and validating it —should it go directly from DEV → TEST → PROD again, or is a temporary branch/environment recommended? We are deploying our projects as extensions.   Would appreciate any insights or examples from those who’ve implemented a similar lightweight pipeline. Thanks in advance! — Jamal

We copied the virtual machine from old physical host to new one. The kepware licenses are broken. All the operation system is same as before. Why the license don't work? This issue can be fixed?Thanks.

Resources["ContentLoaderFunctions"].PostJSON调用大模型接口，短回复没有问题，长回复报以下错：javascript service. Message ::Read timed out。请问是不是系统又限制？在哪里可以更改配置？

Hi, I have this code I can compile with C-SDK I want to make a .snap from.I'm using the version 3.2.0.3436.I used cmake and everything worked fine between thingworx and the agent I coded.The architecture is as such : > c-sdk     > build     > examples          > my code     > lib     > src     > test cmake is done in c-sdk>build The thing is, I now need to make a snap from this project and I don't really know where to start. If I make my snap from my project, I have the same errors I used to have before I did cmake in c-sdk>build (twApi.h: No such file or directory). I even tried to correct it by adding this in my CMakeLists.txt : include_directories(    ${TW_SDK_PATH}/include    include    ${TW_SDK_PATH}/src    ${TW_SDK_PATH}/src/api    ${TW_SDK_PATH}/src/config    ${TW_S

Hello, We run Kewpare 6.16 on a VM Hyper-V for long time. The old system was Windows 2012 R2 and we use Kepware to feed our own Software using OPC DA communication. This system was running quite fine. After we migrate the VM to New Hardware our software is dying on resources on the system. After troubleshouting we found out that the request from our Software (Multi Clients) via Kepware to the Data Server are not responded in time. But looking at the Task manager of the system we could not find any issues. Processors, Memory, Disk and Network are showing no issues. the new hardware is comparable to the old hardware.When we disable the Kepware Service on the computer the system is running fine and is responding timely, but starting the communication with Kepware let the system dying of resources. It seem the requested are not answered timely. When starting only one or two clients the system is reacting correctly but with more clients started on the system the system getting slo