General

For the article “Using FME Flow with Esri ArcGIS Software” at https://support.safe.com/hc/en-us/articles/25407403131917-Using-FME-Flow-with-Esri-ArcGIS-Software, it would be helpful to know whether users who install ArcGIS Server on the server with FME Flow can turn off the ArcGIS Server service without interfering with the FME Flow services running on the server.  Turning off the ArcGIS Server service if it is not being used by FME Flow would improve our security posture by reducing the surface area for attacks when the ArcGIS Server service is disabled on the machine.

Hi all, I have a workspace that was running fine on server, but now the user is unable to upload an input file as a temporary resource. The upload says completed but the OK button is greyed out. This seems to be the case with all workspaces, not just the one in question. Any ideas whats going on? Thanks,  Owen 

Hello Forum, In FME Flow's App builder I would like the word in the OK button to be 'Run' instead of 'OK'. This is because OK, is often thought of as also meaning close (yes, I know the users could just close the browser tab).  But of course, in the case of FME Flow Apps OK most definitely means 'run it'. So, it would be better if the word OK was replaced by Run.  I've looked in Flow Apps>Manage Workspace Apps>(click the App in question)>Customise, but there doesn't seem to be an option to change the wording of the OK button. Any ideas? Thanks,Stuart 

Hi everyone,I’m seeing a reproducible crash in FME Server when using the ili2fme Writer.Environment: FME Flow 2024.2.3 (Windows Server) also happening with FME Flow 2025.2. ArcGIS Server 11.4 on the same machine Nightly scheduled jobs (~50) using ili2fme Writer to validate and export INTERLIS data Problem: A job that fails with model validation errors finishes correctly. The next job in the same FME Engine process crashes with EXCEPTION_ACCESS_VIOLATION (native code), engine restarts automatically. 2Thu-27-Nov-2025 02:02:24.622 AM   INFORM   Thread-20   localhost_Engine1   Emptying factory pipelineThu-27-Nov-2025 02:02:25.271 AM   INFORM   Thread-20   localhost_Engine1   FeatureReader_18 (QueryFactory): 1 input initiators processedThu-27-Nov-2025 02:02:26.032 AM   INFORM   Thread-20   localhost_Engine1   FeatureReader_16 (QueryFactory): 1 input initiators processedThu-27-Nov-2025 02:02:26.837 AM   INFORM   Thread-20   localhost_Engine1   FeatureReader_15 (QueryFactory): 1 input i

We have a job running as a stream, it is not using any specific stream transformers, but it calls an API endpoint and just runs over and over (~10s) as fast as it can poll the API and write the data. The problem is the engine does not release memory between runs, so it slowly grows and grows. And if left for weeks unchecked the server starts having low memory issues. We can stop and start the stream and it will release the memory back.  There is the server config setting MAX_TRANSACTION_RESULT_SUCCESSES which we have set to the default 100, but this does not seem to restart the engine and release the memory. Not sure if a stream just counts as 1 success so it never gets to 100 or what. So, wanted to reach out for any possible solutions for this? Only idea we have is to make and schedule a job that can call the v4 API to stop and start the stream every night. Any better ideas? Maybe some configuration we are missing? Thanks a lot!  

We sometimes have a problem with our MQTT-connector that is disconnecting (always running on our FME Server), but it only shows in the log as a warning. 21318 2021-5-17 11:08:21 | MQTTConnector: Unexpectedly disconnected from broker. Will attempt to reconnect automatically... Error: Connection Refused: unacceptable protocol version. Restarting it works just fine, but it didn't try to reconnect, or at least no log message about it. It was haulted like that for 10 days until I manually stopped & started it … When this happenes, the job seems to be running like normal, and you manually has to go into the log (when running) to see that the job has stopped receiving data from the MQTT-provider. I had to implement a separate test on a different FME Server job to check whether sensordata came in or not, but the optimal would be to get a notification (eg. by email) if the connection is not reestablished within a few seconds. Looking forward to suggestions & future updates. Kenneth.

Hi All,Does anyone have any experience and examples for calling FME server from a QGIS processing script? I have published some (data download) workspaces in FME server into workspace apps for my end users within the company. They really like this, but they would prefer not to leave QGIS. I would like the script to place the zip file returned bij FME server in a folder and unpack it. Thank you.

Hi All, im having the issue of my ArcGIS readers and writers on my FME server (cloud 2021) expiring after a couple of weeks and needing manual re authentication. What is the suggested work flow to automate authentication in FME Server to AGOL? I have read all the forum posts about expiring tokens, watched a range of youtube videos, added client IDs, client secrets and FME server redirects but no luck, i have tried the FME desktop - Arcgis_Token_Getter Transformer. i can get a token but not sure where to put it....  My requirements are - From FME server 2021 read or write a ESRI arcgis online feature service and not to have to manually re-authenticate every two weeks? All my Desktop workspaces work fine (because I'm manually authenticating when I add the AGOL reader or writer) once I publish the workspace to server I have to wait 2 weeks to 'test' if it fails when the token expires so it takes me two weeks it iterate over my failure...  Does anyone have a workflow they co

I have a Flow/Server running, where users access it via SAML authentication. Each user is then assigned to a role with appropriate permissions for a team user belongs to.While users have fmeauthor permissions on Test environment, we have a requirement that users can get read only access on Prod env. With this jobs are run by a system account on the Prod. The challenge here is to find a way to let users be able to view "their" job logs, review configuration or perhaps validate generated data in FME_SHARED_RESOURCE_DIR . Is there a way to configure role in a way so a user can view all jobs, job logs, automations, resources, etc.? For example: speaking about jobs - from what I can see I can either grant Jobs/Access permission (so user can see jobs ran by the user only) or Jobs/Manage (which grants too wide permissions for read only access). Version: FME Server 2022.2.4, Build 22792 - linux-x64

I am new to FME server and I have been trialing creating a few apps for people to use around my organisation.  One of the apps is a simple task where the user drags in an excel file and some analysis is done on it before a number of different files are written.  The first writer is an excel dataset that is fanned out based on one of the input fields, the second is a set of CSV's that just summarise a few duplicates that appear in the original dataset.  When running on FME desktop both files are created in the relevant folders and all appear to be complete. However when running the output zip file from the "Data Download" app only contains the outputs from the excel writer within the workspace. Do I need to set something up within FME server to tell it to expect outputs from multiple writers? I'm currently running FME server 2020.0 Thanks in advance for the help.  

Is there any way to display the job id when a workspace fails when it is executed through a FME Server app ? Currently when a FME server app fails, it just displays a Failure message. I know I can probably do an automation and send an email with that info somehow, but it would nice to just have the job id displayed with the Failure message.

Hi Everyone,I have created a Server App (as admin) and gave an active directory account a right to run it (please, see attached)Then, using app's url I log in as myself (dshatilov) and run the app. I thought it would run the job (the workspace) as dshatilov, but yet it runs it as admin. Is there a way to fix it? What I am trying to achieve is to be able to run a server app as an active directory user and automatically extract user's name, so I could send the results back to a user without asking him or her to type in an email address.When I log in as myself to the FME server and run the workbench from there, it runs it as dshatilov and sends the results back to my address, but when I run the server app, it doesn't. 

Example of idea: 

in a FME FORM i have made some user parameter, that takes in input; one of them have a conditional visibility based on whether another user parameter has an input of > 100. When i run the creator in FORM, and input a number abobe 100,  it works, but when i publish it to FLOW the conditional visibility doesn’t work.then, if I change the condition to “is 100” instead og “> 100” and republish, it works.is this a bug, or is there a work around?  

Hi all, I am trying to modify a workbench and re-upload it to the server (2020.1 to 2020.1) and am getting back an error that: The following readers or writers are included in an enabled service, but are not present in the workspace: [SourceDataset_JSON]Request to xxx returned status code 422. This has only happened since we updated the server from 2020.0 to 2020.1. The workbench takes incoming JSON from a notification service using a topic and subscription and runs it through the workbench and dumps it into an Esri Portal FS.  This error is odd because there is only one reader and one writer in the service.  If anyone else has seen this problem please let me know. Thanks!-Adrian

Hi allI have installed the FME Flow 2025.0.1 Build 25220 - win64, and the server is running but some texts displaying as hexadecimal. Does anyone have any idea why it's happening?

Hi All,I have a FME workspace published with the Data streaming service of FME Server 2016.The workspace creates a HTML file for output.When I run the workspace my browser allows me to download the file (or run it).I want the HTML contents of the file to replace the FME Server page (or automatically open in a new tab).What do I need to change to open the HTML file without user interaction?Thanks for any help,Erik Jan

I wanted to make a simple app that reads from database and writes as either a csv or txt file. I created parameter to either choose csv or txt. I put that user parameter on the csv writer under dataset extension.When I try the app, I’m able to successfully download the csv but when I choose txt - it does not download but shows the full table right on the internet browser. I was wondering how to fix this. Also I am unsure as to how to use the txt writer and put my data as csv there instead. Maybe that would be a good workaround. 

There is one workbench file that only has a SQLCreator within it that executes some update queries in the database tables for Oracle DB.This workbench was previously scheduled to run from FME server but  we recently migrated our FME environment from FME Server (on-prem) to FME Flow 2023 (azure environment). We are seeing issues when this process is run from FME Flow , the SQL queries defined to be executed within the SQLCreator are not getting executed. However, when run from FME Form (2023) it is working fine.Can you please confirm if there is a known issue with SQLCreator from FME Flow.We have observed that when SQLCreator is replaced with Creator+SQLExecutor , the process seems to be working fine from both FME Form and FME Flow.Hence need confirmation to proceed with the change

When you create a Gallery App you have the option to add URL's. By default these will open in a new tab when clicked. It would be nice to have an option to choose between a new tab or to open in the current tab.I now have a Gallery App with URL's to other Gallery Apps. In each Gallery App I put one URL to return to the main Gallery app. After navigation in this a little bit I'm stuck with multiple open tabs. 

We have enabled Version Control to an external repository on FME Flow. Are there any best practices for managing Workbenches in development in FME Form?   My current thought, which I haven't spent much time developing, is to recommend that once a Workbench is uploaded to Flow, the desktop version is archived and any future work on the Workbench must be done by Downloading from Flow and Republishing. Basically, to insist that all work must be done from Workbenches from Flow.  The downside I can see from this is that there is no way to force developers to follow that. Any suggestions?

My FME Server dashboards are not working. When I try to run JobHistoryStatisticsGathering.fmw, it fails with:Terminator (TestFactory): Terminator: Termination Message: 'Translation Terminated: There was an error getting job history.Error Message: HTTP/1.1 404 - http://v01w-app-p0427/fmeserver/fmerest/v3/transformations/jobs/completed?limit=1000&offset=0'What file is it looking for? 

I have published a workspace to FME Flow. When I run it in FME Flow, I get this error in the log file:  Path Reader: Error opening the PATH Reader on folder '\\\\myservername\\T_DFshared\\FME Workflow\\All Reports to Folder5'. Could not find folder.​I checked and the server where the folder resides can communicate with the server that FME Flow is installed on.  The folder does indeed exist and has the files used in the workspace in it.  The workspace published from FME Form starts with a Directory and Pathnames Reader and runs with no errors in FME Form. ​Could this error be caused by the spaces in the folder names in the UNC path shown above or could something else be causing this error?

Hi, This is just for esthetics: I geht houndreds of warnings from FME_Server runnung an Job with some SystemCallers that copy files with robocopy. Although everything works fine and the the files are copied, "Failed to Execute" does not look nice. SystemCaller_12: Failed to Execute `robocopy \\\\RZ2-ZDA-01\\Data$\\Aktuell\\INSPIRE \\\\RZ2-ZDA-01\\Data$\\Historisch\\INSPIRE CP_CadastralParcel.gpkg' If I use the commandline "cmd" from Win2019 on the FME-Server then I dont get any warning.   

I am trying to convert an old script that loads 100+ tables in Postgres to a FME Server automation. One of the things I am doing is a INSERT INTO schema1.table SELECT * FROM schema2.table with a SQLExecutor where schema2 is a Oracle foreign table. This works fine for some smaller tables, but for a large table around 33gb I eventually receive a 'could not receive data from server: Connection timed out' error whether I run it on desktop or server. This worked fine when run from the script. I'm at a loss here, what could be causing the problem?