FME Flow

We have a server running many jobs and one of them normally runs successfully and a log can be reviewed but at other times the job fails and when i access it i get the message Error: The log file for job ID "######" does not exist. I’ve seen the post in FME Server 2019 Error: The log file for job ID "2613" does not exist and we have plenty of disk space, the job frequently works successfully, other jobs are still creating log files. I’ve checked all the suggestions in that post. The failed job is recorded as running for a couple of seconds but without the log I have no insight into why it is failing

The trigger is an incoming IMAP email with potentially more than one attachment. I wasn’t able to use the “Send Event As JSON” option because the email body is being delivered as HTML which fouls up the JSON parser in FME Flow. So I’m sending individual components of the email to a workbench +From:{email.from}+Subject:{email.subject}+Sent:{email.sent}+BodyType:{email.contentType}+Attachments:{email.attachments}+Body:{email.body} I’m splitting up the individual sections of the delivered data in the workbench using StringSearchers with Regex. I wind up with a string with the location that email attachments have been saved to in Flow, which is something like this: $(FME_SHAREDRESOURCE_TEMP)20250121124721-IMAP_Email_Testing-FW_Attachment_test-98e024d4-2316-4cd5-9905-1eb756d02a09 I’ve been struggling with what to do with this string in order to access the attachments to save them somewhere on the network. The ultimate goal is to have the inbound email trigger a workbench that automatically

FME Flow 2024.1 Build 24612 - distributed setup I’m raising with Safe but if anyone has any bright ideas before Vancouver wakes up :-) Recently upgraded our pre-production environment to the above version. Noticed an issue however with some jobs getting stuck in a queued state and never moving on. The jobs are triggered by automation that runs every 10 minutes and assigns the jobs to a specific queue which in turn has two engines available. Sometimes the job gets submitted and runs successfully, sometimes the job gets put in a queue where it remains forever. e.g. the job triggered at 11:20 was processed successfully, the job triggered at 11:30 ended up in the queue, the job triggered at 11:40 was processed as expected The automation log (running in debug mode) doesn’t show any errors or warnings, the queued jobs show as submitted but with no other information Other perhaps relevant information, the jobs should also expire once in the queue beyond a certain length of time and that is no

We currently run all our FME jobs as an AD service account. Unless a workspace uses SQL Authentication for DB access, all DB access occurs as this service account. This has turned into a security concern and I’m sure others have dealt with this. Which of the following is the best solution or what have I missed? It sounds like the new FME licensing allows unlimited engines. Could we create a new engine for each workspace and run it as the service account with DB access, and access to network shares? Kindly point me to steps to do this. Group Managed Service Accounts - can a FME workspace run as a gMSA? Use SQL Authentication instead of Windows Authentication for DB access. I understand FME encrypts passwords but we’d still use the FME service account to access a ton of resources. How do others handle FME jobs and access to resources?

I have an workbench published in FME Flow which gathers parameters then passes them in an automation to workbenches which use them. I am using an Automations writer in it. When I try to run it in FLOW it gives the error: AUTOMATIONS writer: FME Flow error: `FME Flow error: `UNPROCESSABLE_REQUEST(Automations) Route "NewFeatureType" does not exist'' I have never used this writer type before. Does it need to be registered on the server or something?

I’m using an IMAP email trigger for an automation where I’m using the content of the email in a workbench with the goal of delivering a JSON payload to Cityworks to automatically create a new Work Order. I’ve set up a User Parameter in the workbench to receive data from the the email. When I use the “Event as JSON” option for the parameter in FME Flow, the message body is being delivered to the workbench as HTML. This crashes the JSON being delivered. Everything up to the body is fine, but the message body doesn’t make it over, and everything that follows the body is also lost. I’ve been able to get the message content into a workbench by using the individual headings such as “Email Body”, “Email Subject”, “Sent from Address” headings in Flow as the parameters to pass, but doing it this way does not allow you to use the JSON transformers in the workbench. Essentially what I am needing to do is receive JSON formatted data in a workbench from the IMAP email trigger, create attributes fro

Hi, is it possible to upload a file to the FME Flow shared resource folder through the Flow REST API? I tried to setup a POST request for fmerest/v3/resources/connections /<resource>/filesys/<path> but get a 415 Unsupported Media Type Response. I used the following parameters in Postman: In this example I want to upload the text file “test.txt” to the shared resource folder “Test”. Is there a problem with the request, or is this the wrong approach for uploading a file via the REST API? The goal would be to provide a means for external applications to upload files that are then processed by FME Flow. Thank you for your help!

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. I have since used a FME articl

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

Hi We 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 anyone have experience with this kind of

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 edition Seems 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 appreciated.

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 applications I'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 time. Does the "Optimizing Memory Usage" message have to do with the warning a