General

I have a 3D dataset of lines several lines contain large spikes - like the one in the screen shot and attached template. I tried using the SpikeRemover but it doesn't seem to work - Does the Spike remover just not work on spike in the Z direction - the documentation doesn't really help identify why it doens't work?  Failing that is there another way to remove points like the (z points which are about 100 meters off from the expected position). I started going down the line of extracting coordinates and trying to identify them that way but it seems like there must be a better way

Hi, We have recently upgraded our FME Server 2021.* to FME Flow 2023.1.1.1 Build 23636. Many of our workspaces work fine but some have started to take much more time. 1h30m copying 619855 features from SQL server to PostGIS in FME Flow with 15828934!!!  warnings vs. 4m in FME Form with 530 warnings.  The warnings are of this type (both Flow 2023 and Form 2023 - none in FME 2021):2023-11-14 14:01:50| 186.1|  0.0|WARN  |POSTGIS Writer: Renamed user attribute 'DW_TimeStamp' to 'dw_timestamp' based on format constraints for invalid characters, length, and caseIs this behaviour due to the fact that our workspaces in FME flow 2023 still are in version 2021? Regards Jakob

In Manage Schedules, there is a column named "Total Runs"; this tells you how many times a Schedule has been run; is there a FME Flow Setting to have a particular Schedule "auto-refresh" the Total Runs at a specific interval? Currently, to view the total runs a user needs to refresh the page or step out of and back into Manage Schedules.  

FIXEDWe have encountered a strange problem on FME Server 2020.0.2: all of a sudden the Web Application Interface has stopped working.This happened right after I terminated a job that had run all weekend. In the logging of FME Server we can see all jobs are still run as usual, so it is just a problem with the Web Application Interface. This appears to be a Tomcat problem. The server runs Apache Tomcat 9.0.24. So we checked the latest catalina.<date>.log file. It contains the following error messages:06-Nov-2023 09:28:32.913 SEVERE [main] org.apache.catalina.core.StandardServer.await Failed to create server shutdown socket on address [localhost] and port [8005] (base port [8005] and offset [0])java.net.BindException: Address already in use: JVM_Bindat java.net.DualStackPlainSocketImpl.bind0(Native Method)at java.net.DualStackPlainSocketImpl.socketBind(DualStackPlainSocketImpl.java:106

Hello, we would like to authenticate FME Flow users via „KeyCloak“ application. Our idea is that FME Flow would somehow be connected to the KeyCloak application. The moment a user wants to log in to FME Flow, he would be redirected to KeyCloak, authenticated. KeyCloak passes the information back to the FME Flow application, and user will be logged in. Do you have any experience with FME Flow-KeyCloak collaboration?Keycloak supports protocols OpenID Connect, OAuth 2.0 and SAML 2.0.

I have added a new team member into FME Server, with fmeuser role. Basically this user only care read and Run workspaces under Parks repository so in FME Server under  Repositories slide-down I set this configurationbut when I checked the user in the Server end point I am seeing this:As you can see two Samples and Tools repositories also listed there! When I check back in Server side, looks like there are some circled-checks on read and run options for these two! Now can you guys pleas let ne know how I can force this user to ONLY have access into Parks repository using

Hello,We have updated FME Server from 2020 to 2022.2. Then there is something wrong with the upload files : We publish into FMEServer a workbench containing a "File/URL" User parameterThen we RUN the workspace 2 times : The file we upload has the same name, but the content of these 2 files is different--> FME Server doesn't update the file located here : FME Server\\SystemShare\\resources\\system\\temp\\upload\\ if we don't purge the folder (In case we change the file name, it works well.)We have not encountered this problem with FME 2020.2 Can anyone help?Thank youRonan

I'm trying to prep a road network for use in the ShortestPathFinder. I found a flow on FME's documentation that looks like it does what I need (it uses a road network to prep for routing).  The first three steps appear to have run fine, but I think the Snapper step keeps repeating. The following line has reached 100% completion 3 times but every time it does, it starts over doing the same thing: "Intersection operation processing one line against many others... 99.34% doneIntersection operation processing one line against many others... 99.87% doneIntersection operation processing one line against many others... 100% doneIntersection operation processing one line against many others... 0.52% doneIntersection operation processing one line against many others... 1.05% doneIntersection operation processing one line against many others... 1.58% done" Below is a screenshot of the flow I'm running.

Hi,I would like to create a fme server app where a user can upload multiple ZIP files. For each zipfile I would like to call a FeatureReader for one file within the zip file. How can I accomplish that?I tried the directory and file path names reader but this gives an error in FME server saying it expects a folder and not a file.Thank you

I have a (FME Form 2023.0) workbench that generates three Excel files, where running locally via FME Form gets me a zip with this structure:my_data_$(TODAY).zipfoo_$(TODAY).xlsxbar_$(TODAY).xlsxbaz_$(TODAY).xlsxAll three XLSX files are generated by three separate writers and have very distinct structures and different template files. When I publish the workbench to FME Flow and try running it with the DataDownload or Data Streaming services, however, I get a completely mangled zip file with this structure insteadOutput8719176120317945288.zipXLSXW_1foo_.xlsxXLSXW_2bar_.xlsxXLSXW_4baz_.xlsxIt seems like both services ignore whatever path the readers are configured to write to, and instead use a

Hi, Following the documentation: https://docs.safe.com/fme/html/FME-Flow/AdminGuide/Starting-Stopping-FME-Flow-Manually.htm I tried to enable HTTPS for the server (FME Flow 2023.1 Build 23619 - win64). I can browse via HTTPS to my server without problem.When trying to run a job (Run Workspace / Samples / austinApartments.fmw / Data Download) I end up with an error 500. In the log (D:\\FMEFlow\\ProgramData\\resources\\logs\\tomcat\\NAME\\localhost***.log): SEVERE [https-jsse-nio-443-exec-8] org.apache.catalina.core.StandardWrapperValve.invoke Servlet.service() for servlet [JobRunnerForm] in context with path [/fmeserver] threw exception   javax.net.ssl.SSLException: Unexpected error: java.secur

The municipality of Eindhoven intends to change the current way of 'authentication' to 'SAML' authentication.The current way of authenticating is based on 'Windows Active Directory'.Within Eindhoven we use the publishing of FME Desktop scripts to FME server via the so-called 'FMEServerJobSubmitter'.This is based on an Active Directory account.FME version 2022.2.1 On the Safe website we find several articles that authentication within an FME flow via SAML does not support a number of transformers.One of them is the 'FMEServerJobSubmitter'. Two possible 'workarounds' are mentioned.We do have some questions about this. Hopefully someone can comment on this. 1.    Use a System AccountFME Server can consist of both SAML and system user accounts. Consider creating a system user account whose credentials can be shared with those users who auth

HelloI have checked the folder and file tree as well as the user rights, the proxy is disabled but nothing helps...Sincerely Error details :lun.-18-sept.-2023 12:23:20.749 PM  INFORM  fmeenginemgrnodeclient  402911 : Connecting to Job Queue localhost node host localhost port 6379.lun.-18-sept.-2023 12:23:21.253 PM  ERROR   fmeenginemgrnodeclient  402902 : Failed to connect to Job Queue. Please ensure Job Queue is started.lun.-18-sept.-2023 12:23:21.253 PM  ERROR   fmeenginemgrnodeclient  Failed to connect to any host resolved for DNS name.redis.clients.jedis.exceptions.JedisConnectionException: Failed to connect to any host resolved for DNS name.   at redis.clients.jedis.DefaultJedisSocketFactory.connectToFirstSuccessfulHost(DefaultJedisSocketFactory.java:63)   at redis.clients.jedis.DefaultJedisSocketFactory.createSocket(DefaultJedisSocketFactory.java:87)   at redis.clients.jedis.Connection.c

Hi all.Running 2022.2.6 on Windows.My server is stacked with a 100 Gb C drive, and a 200 Gb E drive.I would like to move the location where FME server stores the job log files from C to E, and have therefore edited the file "C:\\Program Files\\FMEServer\\Server\\config\\logger\\engine\\messagelogger.properties", changing the location in LOG_FILE_PATH to a path on the E drive.But after restarting FME server, it still writes the job log files to the old default location on the C drive !!!Is the logger property even used ?Cheers.Lars

Hello,I am trying to execute an job in datastreaming via HttpCaller.The message I get back is "HTTP/1.1 401" although I have my login data filled in.I tried to use my usernmae and pw on the browser, it works fine.Just the transformer tells me its not ok.I tried the two possibilities mentioned here for users without a token : https://community.safe.com/s/article/401-unauthorized-error-when-making-fme-server-requDoesn't work either way.... Maybe someone stumbled accross a similar problem and has a solution ?  

Hi Holly,I am addressing you since your name ison the article(https://community.safe.com/s/article/how-to-create-a-wms-service-using-fme-server?t=1694673698026)and it is not possible to leave a comment I have implemented the WMS via FME Flow and I have noticed a strange behavior that I wanted to report.When requesting data from the WMS service I have noticed that the client (QGIS) creates a request per layer. This is as expected, however on the FME Flow side I get 2 identical requests for each WMS layer. This doubles the response time of the WMS service and uses twice the resources on FME, is this something you can look into and reproduce? Since the WMS request is done via the GetCapabilities workspace is there a way to remove this duplicate request before they trigger the GetMap workspace?

Please excuse the crude drawing. I'm using a SpatialRelator with the red features in the Suppliers port, and the blue feature in the Requestor port. Features #1 and #3 need to pass the spatial predicate due to the relation between them being multiple verts (a line). Feature #2 would not pass, due to only having a single vert in common. None of the default spatial predicates will exclude #2 from the test. So, I've been trying various matrix to find a solution, but none seem to work. What would be the appropriate DE-9IM matrix, given that it is only polygon features being used? I have been referencing this help page: https://docs.safe.com/fme/html/FME-Form-Documentation/FME-Transformers/Transformers/spatialrelations.htm#Dimensio

 Hi,  I'm trying to configure a new FME Flow deployment with a new customer.They would like to keep all storage on the same storageclass: azurefiles-fmeserver.This storage class is created as highlighted in 3. Setup Shared Storage using Azure Files: https://docs.safe.com/fme/html/FME-Flow/AdminGuide/Kubernetes/Kubernetes-Deploying-to-AKS.htmHowever, if storage.postgresql.class is filled in, the deployment does not work.If this parameter is empty, the deployment completes and the storageclass: default is chosen. We have tried adding multiple different parameters in the postgresql.primary parameters, but none that helped us.Is there anyone who has had the same issue?Below is a exert of the Values.YAML file.   storage: useHostDir: false reclai

It seems to me that the Map Configuration window doesn't work (or show) anymore when using the Geometry Picker in 'Full window mode'. I can also imagine that technically the window/menu does work/show, but is placed behind the 'full/expanded window' of the Geometry Picker, causing it not to be accessible/usable. To be clear. What I'm talking about is this menu/window, which does work when using the Geometry Picker in 'Inline mode';However, when I use the Geometry Picker in 'Full window mode', nothing seems to happen/show if I click on the Map Configuration button. E.g.;Are others also experiencing this issue? Is this a known bug? ps. I'm using FME Flow 2023.0.0.

Is there a method to read all workspaces on a server ex. list all workspaces that have specific transformer or list all workspaces that have specific parameter.  Also, where are workspaces on a server? Is there a Folder for workspaces? Are they ascii readable? Thanks in advance

Hi, I am new to using data download or streaming in FME Server, I have only used job submitter up until now. I am wanting a very basic workspace to generate a CSV which is then returned to the user via the web browser, once they run a fme workbench via FME Server directly (Server App).  Does anyone know which method would be best to return just the CSV file, on demand, direct to the users web browser for download? I am looking through the help files for examples on best practice, but its all a bit confusing. I have currently got a workbench published to server, and its using data download as its registered service, but it keeps downloading a zip file with a half dozen sub folders within, then my CSV file with its associated data. I am presuming I have over complicated things. In essence, I am hoping to get the results of this workbench to the user via a CSV which they can download to their PC: 

I have a workspace where the Reader reads an Access database using a lettered drive path (K:\\projects\\databases\\access.mdb) The Writer writes to the same database using the same lettered drive path. When I publish this workspace to FME Flow without uploading/selecting files, the Database Path under Published Parameters is empty. ​But, when I change the database paths in my workspace to a UNC path (\\\\gisserver1\\projects\\databases\\access.mdb), I publish the workspace (without uploading/selecting files) and I see the UNC path to the Access Database populated in Published Parameters - Database Path. ​If my workspace reads from a lettered drive path and the database is Microsoft Access, when I publish the workspace, do I need to upload/select files? This would upload the Access database to FME Server.

We (rarely) get a glut of job submissions to some of our FME Server queues (Running on FME Cloud).  These glut of jobs usually come from someone who has automated a bunch of submissions (we're totally fine with that). We want these jobs to be processed, but at the same time we still want to be able to handle other user requests in a timely manner. I created an idea over here which would help me: https://community.safe.com/s/bridea/a0rDm000001Sf7kIAC/option-to-move-submitted-queued-jobs-to-new-queue It outlines the scenario in more details. Just looking for tips or idea on how to work around this. The service in question is essentially just a clip and snip but depending on what the user wants the request can take anywhere between 10 seconds and an hour or so to process.

I created an FME Server role that provides read-only access to view FME workspaces published there (i.e. - Workspace Viewer). I would also like to add read-only access to view items under Schedules but only see options for Full Access or None for a role's permisssions. is there a way to provide read-only acccess to Schedules? We are currrently running FME Server 2020.1.2.1 (Build 20624).

I have a flow source Movie    Role    ActorM1         R1        A1,A2M1         R2        A1,A3M2         R1        A2 and I need to switch to the vue Movie     R1      R2M1      A1,A2    A1,A3M2        A2how is it possible, I tried AttributePivoter but didn't find a way to switch rows to columns. The Roles list is not known before, so it should take all unique roles as columns.