FME Flow

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?

In order to pull data from one of our clients’ data repositories, we are using a custom Python SDK that they have provided.  Installation and testing went smoothly for use with FME Form, but no so well with FME Flow.   For FME Form, I used fme.exe python -m pip install.  For FME Flow, I’ve tried navigating to Resouces>Engine>Plugins>Python>python311 and uploading the directories there, I’ve tried running fme.exe python -m pip install with the target directory set the FME Flow Resouces\Engine\Plugins\Python\python311.   No matter what I do, my transformer fails with the following errors:Python Exception <ModuleNotFoundError>: No module named xxxx Any suggestions?

Hello, I am new to administering FLOW. We actually have just purchased the product and are in the process of setting everything up, and learning.I would like to integrate SAML with FLOW. It is straightforward enough with the ability to use the metadata URL or import from a xml file. All of that and actually enabling SAML under User Management was quite straightforward. We are a huge IT shop and I have to submit tickets to another department to have the applications integrated with ADFS.The software was installed as port 8080, and is accessed through an F5 (load balancer). After install, I had the F5 team set up listening on port 8080 and the software is accessible via the F5 VIP and internally. All that said, there have been a few issues. Initially the software was passing communication for ADFS over http, because that is how it was setup. So this was expected. The problem there, is the ADFS console would not allow http to be used, it only allows https. 

Hello Forum,I am currently working on a task that involves copying files from a source server (Server 1) to a destination server (Server 2). I am new to FME and am unsure how to approach this.Could this process be achieved using an FME transformer, or would it require writing a Python script? I would greatly appreciate any guidance or suggestions on where to begin.Thank you in advance for your support!

Where can one find the system cleanup log files on FME Flow (2023.2; build 23764)? I’m testing cleaning up a folder under Resources/Temp. I don't see any corresponding logs under Jobs or System Configuration → System Cleanup. Or, is it possible to test a specific cleanup task on FME Flow? I’m currently at the mercy of whenever system cleanup runs, which is vaguely set to once a day at no set time so not ideal for testing.

hello all, I have 2 sets of data with no fields which I can join on (for Feature Merger transformer)I would like to keep the 1 set of data (ieqm2023) with its geometry and attributes and add to it the attributes ONLY from the 2nd dataset (ieqm2025) where geometry overlaps with the (ieqm2023) geometry.  so for instance I want the same numbers of records (ieqm2023 - 1,880,638 ) afterwards only with more fields (from 2nd dataset)    (see attached) I was thinking an Aggregator or an area on areaOverlayer transformer might work? but I have had no luck getting the correct result so far. thanks and much appreciated for any help!! 

Hello and happy 2025!Was speculating on feature types selection possibilities when reading as a single merged feature type with a wildcard on flow.My scenario being that a user will drag and drop a DWG to a Flow tool and then that said tool would read it in and then display all the feature types (layers) that said DWG consists of  in a list. Where the user could then check a box in front of each and every layer they need and afterwards select “Run” to start the main tool (which is a conversion into another file format).Thus the tool not converting all features/layers within a file but only a selected few which are selected like one does for example with the “Feature types to read” option in the FeatureReader.Anyone here done this before that wishes to share?

Hi, I am trying to send EMAIL on Port 587 from FME Flow application (V 2023.1.1) and I get below error. Can someone please help on how to resolve this error? Do I need to specify port 587 on any configuration file? 

We are trying to create a read-only user within the FME Form PostgreSQL DB using pgAdmin logged in as the fmeflow user. We are using the out-of-the box PostgreSQL DB which is installed on the core machine. However, we can’t seem to create a new user. Items (such as right-click menus, etc.) are grayed out. Is there documentation somewhere on how to do this? We want to query the FME Flow DB directly. Or is this frowned upon?Version: FME Flow 2024.1.1.0 (Build 24619)Thanks,Mark

Hi everyone,I just noticed in the dockerfile (version 2024.2.1-b24801) line 130 a new engine with the type UTILITY. Does anyone know what this is for? Is it a mandatory component in the architecture?Thanks a lot!

Hey. I need some directions, please. I currently have several workspaces that I need to run chronologically. They are workspaces that process data, and they include WS that perform operations on the database. What is important is that each subsequent workspace is run based on the previous ws running correctly. My idea would be a sort of queue where each succesfly status of the previous ws would trigger the ws following it. This whole thing should then run on the server. Maybe a workspacerunner? Or something else?Thanks

HiWe are about to configure Azure AD (Entra ID) Integration to authenticate users to our FME Flow. There are some workspace applications (flow apps) that are embedded into an ArcGIS Portal Experience builder app. The authentication to the ArcGIS Portal is also done with Azure AD accounts. The idea is, that the end user will open the ArcGIS Portal web application and access the FME Flow workspace application from that platform, rather than re-navigate to the Flow app. If SSO is enabled, will the FME Workspace/Flow app application authenticate automatically without extra login to the workspace application, when it is accessed from the ArcGIS Portal Experience application, as it is embedded to the application? Current setup with basic user authentication and FME Flow user access control fully inside FME Flow naturally requires a log-in to the app OR requires the Flow app to be published without authentication, but we want to ensure security. Does anyo

Hi All, I want to extract detailed log messages from FME job running on FME Flow (Ver 2021.2.3). What options are available to extract this information along with basic job information like Workspace, Repository, user, source, start and end time. etc. Appreciate any feedback. Jay    

After upgrading FME Flow to 2024.2 some of our workspaces fail with the following error:WARN  |Library 'F:\FMEFlow\Server\fme\plugins/realdwg/REALDWG.dll' was found but could not be loaded. Ensure that all the dependent modules exist for this library WARN  |Could not load module 'F:\FMEFlow\Server\fme\plugins/realdwg/REALDWG.dll' : 'The specified module could not be found' WARN  |Could not load module 'F:\FMEFlow\Server\fme\realdwg/REALDWG.dll' : 'The specified module could not be found' ERROR |Module 'realdwg/REALDWG' is unavailable for use with this FME editionSeems like RealDWG and AUTOCAD_OD readers/writers do not work anymore. -Is this a known issue/bug?-Is RealDwg not supported anymore? It does work in FME Form 2024.2

Effective August 1, 2025, Legacy products (all editions of FME Desktop and FME Server) will be deprecated and updated to FME Form and FME Flow. Annual Maintenance will be based on the list price of FME Form and FME Flow.https://fme.safe.com/pricing/annual-maintenance/I am interested to know how these updates (upgrades) will be done. Are these updates mandatory and will they be done automatically? If so, is there a risk of any errors?Our company has several Production servers with FME Server v2020.2 installed and we also have old versions of FME Desktop installed.

From FME Script I’m using FMEFlowNotifier and sending one attribute value to a specific topic. Then I have a subscription that listens on that topic. So far so good. The problem is that when email is sent, the email body that I specifified on the server is not included in the mail.  Anyone who can think of a reason for this?   

I am trying to create an automation using a trigger that watches a directory. If a file in the directory is added/created it launches an FME Workspace. However I keep getting the following error:2024-12-14T20:10:08Z | 410109 : (Automations) Creating link from trigger...3 2024-12-14T20:10:08Z | 410108 : (Automations) Creating dirwatch trigger...4 2024-12-14T20:10:08Z | 410112 : (Automations) Unable to assemble automation. There was an issue with component "Trigger". Publisher dirwatch does not exist.The same thing happens when I try add a trigger that watches an FTP? and I get the same error when trying to add a Publication under the Notification.What am I missing? Do I need to add something special for a trigger to work? is it possible that I am missing some part of FME?I am working with a FME FLOW 2024.1that is in a Azure Kubernetes cluster.Any help or hint that could explain the error is highly a

FME shows the following warning during the translation:2023-09-06 10:24:29|5015.8|  0.3|WARN  |Failed to free sufficient memory to reach the process usage limit. To improve stability and performance please increase the memory available to FME. Available memory can be increased by adding physical RAM, increasing swap space, or closing other applicationsI'm using FME(R) 2022.2.1.0 (20221202 - Build 22776 - WIN64) on Windows 10 22H2 Build 19045.2006. I have 64 GB of memory. FME uses about 10 GB during that translation and 50% of RAM are free. Why does FME show this warning when there should be enough free space? FME says it's trying to free memory because it doesn't have enough memory, right? Why does FME fail to free memory? How can FME achieve that goal? What does it do to free memory? And what is the process usage limit? Also I get these log messages and I feel like optimizing memory usage takes up very much ti

Hi everyone,I’ve created a python script that has the docx library and arcpy library imported in it that takes field values from a GIS feature class and writes it to a docx document successfully. I would like to trigger the python script to run every time that feature class in an enterprise geodatabase is updated.I am working in the 11.3 Enterprise Environment and have come across this support page regarding implementing webhooks and FMEhttps://support.safe.com/hc/en-us/articles/25407414850061-Integrating-ArcGIS-Enterprise-Feature-Service-Webhooks-With-FMEI will attempt to do the above, but I have a simple question regarding the FME workspace itself.I have read up on the PythonCreator and PythonCaller transformers, with the PythonCreator not requiring an inpput

We got this message: On December 17, 2024, we will release the following updates which will include the fix:• New release builds of 2023.2, 2024.0, 2024.1, 2024.2 and newer• Manual patches for versions 2021.1 through 2024.1.2.1 So I want to install the patch for our 2023.1.2 (build 23641)But where can I download the manual patch for this release?

Hello,  I am using the 2024 versions of FME (Form & Flow).  I would like to create an FME Form workspace that will be deployed on FME Flow and later called via the FME REST API.  In this API call (POST), I would like to include structured information in the request body that the FME Workbench can process.  For example, the request body could look like this:  ```json{    "entities": [        {            "schemaName": "xxxxx",            "entityName": "xxxx",            "fieldNames": [                 "ID",                "NR_ATT"            ]        }    ],    "whereClause": "ID = 'xxxxx'",        "constraint": "FIRST"}```  I would also like the response of the execution to contain structured information, which could include standard execution feedback along with additional custom data.  For example, the response could look like this:  

Hi everyone,after opdating FME Flow from 2024.0.3 to 2024.2.1 I’ve noticed that there is a (new?) “Data collection” optione within the Security Options. We are using FME Flow via docker compose. Is it possible to disable the “Data collection” via a (Docker) environment variable so I don’t have to disable it manually after every update?GreetingsPeter

I have a process that writes a number of graphs to a folder on the network.  It works great in FME Form.  If I attempt to publish this workspace to FME Flow as a Job Submitter it doesn’t write the files to the folder at all like the same file does at the desktop level.  The output path is set up in an RCaller so there isn’t a Writer in the .fmw script like normal.  There are zero errors on the desktop side, and I have made sure the FME Flow portion no R packages are causing errors.  When I run this process, I receive zero errors in Flow, but none of the files write to the directory.  The end process would be that a person could upload a number of spreadsheets run this process from a webpage (Flow) and the output would show up in their department’s folder.  I read in the community about it possibly writing to another directory and I attempted to check that out without success.  Any help or guidance would be appreciated.  

I tried to find an alternative transformer for the system caller to do the pg_dump, but even that did not work out.

HiI am the admin in my organisation for FME Flow and I woke this mornig with this messagehttps://support.safe.com/hc/en-us/articles/31265482270349-Security-Update-FME-Flow-Privilege-Escalation-VulnerabilityThe problem is that I can’t access it even after it asked me to log in.​@safe can you point me out to the right issue so I can understand the vulnerability that it might implicate in out environment?Best regards,Felipe Verdú