FME Flow

I'm testing FME Server 2019.1, We created a test workbench that does nothing really and created a server app for users to test. We get an error. "There was an error submitting the job. " Anyone know what i should look at and/or the help documents relating to this configuration?

Hi Partners, A new article has been published which rounds up all the resources related to upgrading an FME Server deployment (Express Installation) to a Fault Tolerant deployment. How to Upgrade from an Express Install to a Fault Tolerant Install

Hello, I noticed that notifications (emails) from the Community site have stopped since some point within past few days. Can you receive notifications according to the settings on your preference?

Hello,  If there is a FmeServerJobSubmitter transformer with options:Wait for Job to Complete = true Will it keep http connection untill job is done? 

Upgraded to 2018.1. Server has 7 engines and 64 GB or RAM. Some jobs will get error "An error occurred while trying to obtain an FME session" but the second time it runs will run OK. Will open a ticket with Safe later today but if anyone has experienced this issue let me know the fix.

Hello, i have question about testing workbench that is uploaded on FME server.Is there simple method how to work with scripts that are uploaded to FME server ?Testingis scripts in such enviroment is quite uncomfortable - a have to maintanin two wersion of script - one with local path, one with server paths to data input.Is there simple way to work in such enviroment?

Is there any way to restart FME engines on a set schedule ? Like from a rest call or something like that ? Just exploing ways to clear all database connections an engine might have open, context is firewall timeouts.

Hello,We are in a process of upgrading fme 2016 to 2019. After some tests that passed well we need to move all our components to use fme 2019.  Normally what we would do is to create backup of fme 2016 and restore it on 2019. And this is what we did initially, Then for 2 weeks both 2016 and 2019 was running thats why we have some new things on 2016 and some on 2019.  What we need to migrate from 2016 to 2019 are only schedules that were created after given date.  For some reason I can't create normal backup on fme 2016 right now. I wrote it to safe support and they know that in this build there is a bug so we would need to apply fot fix.  My idea to do it in a different way is to:  1. Use fme 2016 desktop to export FME Server Configuration to ffs file (choosing repository and scheduling) 2. Run fme 2016 desktop to write from ffs file to fme server configuration of fme 2019 only scheduling with insert mode.  Will it work and I will not do any damages to new fme 2019 database configu

Hello All,I am trying to import an external module named openpyxl in FME but unfortunately isn't able to setup the environment. Can anyone please help me with identifying what all is required to setup the module in FME?Every time i try to run it inside the python shutdown script it throws the error as module doesn't exist.

Hello, I'm running FME 2019 server. In workspace I have ServerJobSubmitter with parameters like that (I updated to newest version of this transformer):  My expectation was that test.fmw will be executed on the same engine but it was executed on another engine (which is installed on the same engine server).  Why is it like that? Is it possible to have that job executed on the same engine like it was in previous versions?  

Hello I have just upgraded to FME Desktop 2019.1 and am getting a few errors when running a published workspace in FME Cloud.How is FME Cloud Upgraded? and is it necessary to keep the two in sync?

I upgraded FME Server to 2019.1 and my streaming services are failing. It appears that 'FME_SERVER_WEB_URL' in the request did not pick up the change when I enabled HTTPS during the install. It is still showing the original HTTP url with Port 8080 (http://SERVERNAME:8080). Is there another file besides fmeServerConfig.txt that needs to be modified? I have a distributed architecture deployment: All components are on a single machine except the Web Services component, that is on a separate machine. I upgraded from 2019.0.1 to 2019.1 and now my streaming services are not providing an output. Thanks in advance.-Mark

I'm testing out the new automations in FME Server 2019 and I am needing to chain workspaces together while passing parameters (file outputs) between them. I am not finding a way to do so without using the FME Server Job Submitter transformer that I have used in the past. Is there a way to pass parameters between chained workspaces in the automations rather than using the job submitter transformer?  I've gone through the tutorials here: https://knowledge.safe.com/articles/87453/getting-started-with-automations.html , but haven't been able to figure that out.  

Hi,if I click on "Topic monitoring" in the fme server portal I get the error:"There was an error connecting to monitoring service"The port 7078 is open and if I request http://fmedev-test.de:7078/, I get the answer "not a WebSocket handshake request: missing upgrade". So the firewall rules should be right and a telnet command to fme server via 7078 also works.I've also set WEB_SOCKET_SERVER_NAME inUtilities\\tomcat\\webapps\\fmeserver\\WEB-INF\\conf\\propertiesFile.propertiesThe logs from websocket.log in publishers and subscribers are:Opening log file using character encoding UTF8 Tue-13-Feb-2018 02:45:58.721 PM INFORM main 801001 : Publisher "websocket" is starting up ... Tue-13-Feb-2018 02:45:58.721 PM INFORM main 801002 : Publisher "websocket" is updating configurations ...Opening log file using character encoding UTF8 Tue-13-Feb-2018 02:45:52.909 PM INFORM main 803001 : Subscriber "websocket" is starting up .

I need to create FME Smallworld Edition license server. Do I need to install smallworld core on that system?

I'm using directory watch with FME server 2019 to trigger a workspace when a file is updated. the Directory watch is set to watch the directory every days. For now, when there is an update, It run the workspace at noon, and i wish to wait till midnight before the workspace start. Is there a way to add a time option for when the directory watch start ? or a waiting time before the workspace start ?Thank's

My company has an MITM proxy setup, which is resulting in many cloud based service requests failing. Obviously it only applies to any requests via HTTPS, so I am wondering how other people have gotten around these issues? I have set the correct proxy settings within the 'Network Proxy' options and have even setup custom proxies for each of the cloud based services, but still no success.For instance, attempting to use the HTTPCaller to access the HERE Geocoder results in the following error (the blacked out proxy is correct):An example of another issue I am encountering is when I attempt to create a CartoDB connection, I am receiving the following error (the blacked out domains are correct):I have tried everything other than asking my IT department to white list the cloud based services. IT have confirmed that they can see the requests reaching the proxy, but have claimed that the authentication needs to be set to NTLM, but this only results in similar errors. Even when I use th

I have a workspace on FME Server that sometimes will receive strings with accent as a parameter. I am calling the workspace with the direct url from a javascript application. I serialize the parameters with browser's encodeURIComponent function to be valid in the direct url. So a string parameter like NomProjet = é become NomProjet=%C3%A9 I found out that parameters, when they are containing accents, are automatically encoded into hexadeciamal by FME Server. So the parameter é become E9 in the workspace. I have to use a TextDecoder to decode them to Latin-1. However, when the parameter does not contain accents, it's not encoded to hex and the TextDecoder produce an error. I would like to prevent FME Server to encode those parameters into hexadecimal. Utf-8 would be perfect. Is there a way to do that? 

I'm working with FME Server 2019 and was wondering if/how it is possible to use the automations functionality as followed:So, i start a Data Download Service workflow with REST. This workflow performs its arts and then sends a mail if valid with the results if available or no mail when no new results are the (but the workflow still did the job). And failed if there is a bug/rejected stuff. Is there a way to check/test the {downloadUrl} that is used in the Subscription before sending the link?Sending format C is easy because that is just one topic and subscription connected to the failed results. But when the job is successful and the URL in {downloadUrl} is valid it should sends the mail in format A and if the URL in {downloadUrl} is _undefined_ it sends the mail in format B. I was wondering if there is a way to keep the Data Download service as-is but test the content of the {downloadUrl} for further actions regarding the formatting of mail mail.

Update: Newb issue with FME Server. Opened up port 587 on the FME Cloud security settings. So as with most things I solved my own problem minutes after I gave up and asked for help.I am building an application using FME Server (cloud) and part of it is an email exchange. Using the extensive tutorials provided by Safe (thank you for that) I am able to receive and process the email from an Office 365 accout with imap settings outlook.office365.com and publish to a READ_EMAIL notifier. The issue I am now having is setting up the SMTP settings. Following the great tutorials by Safe I can successfully send via my gmail account. However, Office365 seems to be another animal. I've logged into the account and used the specified settings:Server name: smtp.office365.comPort: 587Encryption method: TLSI've tried every permutation of this (SSL, TLS, nothing, etc.) The only place I can see the difference is there is no option to verify the certificate (yes or no.) Microsoft specifies not to

I'm using fme desktop 2018 (18586) on ubuntu 18.04 and that was working perfectly. After an update on the system, fme workbench isn't initializing anymore but the other applications are working fine(data inspector, license manager, and quicktranslator)

I have a workspace on FME Server that sometimes will receive strings with accent as a parameter.I am calling the workspace with the direct url from a javascript application. I serialize the parameters to be valid in the direct url.            serialize = function(obj) {              var str = [];              for (var p in obj) {                if (obj.hasOwnProperty(p)) {                  str.push(encodeURIComponent(p) + "=" + encodeURIComponent(obj[p]));                }              }              return str.join("&");            }So the string parameter NomProjet = Test ML é become NomProjet=Test%20ML%20%C3%A9 However, at the end of the workspace, I write the value of the parameter in an Excel file, but the value is blank.If the parameter does not have accent the output is correct.What is the problem here? What can be done?

Hello,There is something wrong with my fmeserver, now it can't run any datastreaming services, F12 my chrome, it appears "provisional headers are shown",and response nothing,checked the server side logs, nothing written,means server side didn't receive the post,but successfully run jobsubmitter, datadownload and others. It happened many times, I have fix this problem by restart the "FMEServer_ApplicationServer.exe" task in my serverIs there any better way to fix it?Regards, yujian@rylanatsafe @rahulsharma @daleatsafe

Hello,Is there anyway to generate token id from postman?Thanks.

Hello,We have fme 2016.We have two engine servers, on each one 4 engines. What is the algorithm for choosing engine to each new job will be assigned? Let's say that we are not using any routing tags.