Automations writer error: Reading FME Server response from channel failed with error number 392008

I was using the FMEServerJobSubmitter and got the same error. Seems to be related to permissions (database permissions or other permissions) - so perhaps its the same with you? Does the user that runs the automation have permissions to all databases, and running workspaces?

FMEServerJobSubmitter_2 (ServerFactory): Reason - credentials for user are invalid

FMEServerJobSubmitter_2 (ServerFactory): http://myfmeserver.com - Failed to submit request to run workspace 'Myworkspace.fmw' in repository 'myrepo'

FMEServerJobSubmitter_2 (ServerFactory): Reading FME Server response from channel failed with error number Reading FME Server response from channel failed with error number 392008..

 FMEServerJobSubmitter_2 (ServerFactory): http://myfmeserver.com - Submitting a request to run workspace 'Myworkspace.fmw' in repository 'myrepo'...

FMEServerJobSubmitter_2 (ServerFactory): Reading FME Server response from channel failed with error number Reading FME Server response from channel failed with error number 392008..

FMEServerJobSubmitter_2 (ServerFactory): Reading FME Server response from channel failed with error number Reading FME Server response from channel failed with error number 392008..

I still have this issue. I have even made a workspace that just starts and uses this connection every 6 hours to make sure to "keep it alive". However this does not work - the only thing that works is to open the workspace in FME Desktop and update the access token and republish:

Updating access token for 'myfmeserver_myfmeuser'

Connecting to 'https://myfmeserver.com' (web connection, authenticated login)...

Successfully connected to FME Server 2021.2.3 (build 21812).

Retrieving item parameters for 'Myworkspace.fmw'...

Successfully retrieved item parameters.

It seems that the "Updating access token" does not work on FME Server automatically? And has to be renewed every week? Anyone from Safe enable to help with this issue?

Hi @Sigbjørn Herstad​ ,

It sounds like you've encountered this known issue: https://community.safe.com/s/article/FME-Server-Web-Connection-fails-with-Reason-credentials-for-user-are-invalid

This seems to be the issue, not sure how I couldnt find this by searching, but thanks for the guiding in the right direction @hollyatsafe​. I will try the steps mentioned and report back on this thread if unsuccessfull.

@Sigbjørn Herstad​ We were only made aware of the issue quite recently, so it's possible this article was created after you encountered the problem! I'll look to add the error code into the article as well in case that's what you were searching for.

I think I also found an alternative workaround:

Submit the workspace as normal. When the token has expired (after a week):

1. go into:https://fmeserver/fmeserver/#/tokens/all?type=session&page=1 
2. Find the correct token (you can look at the date and the user). You can try to enable it, but it will fail due to error :

"Token expiration should either be specified as secondsToExpiry or expiration, not both"

"Token expiration should either be specified as secondsToExpiry or expiration, not both"

However if you examine the code that is sent you can examine the request and do the correct request in another software (for instance Postman):

HTTP PUT to: https://fmeserver/fmeapiv4/tokens/TOKENID

Header: Authorization = fmetoken token=FMETOKEN

Body:

{

 "id": "xxxx",

 "name": "Token Service Login - xxxx",

 "description": "Token Service Login",

 "ownerID": "xxxx",

 "owner": "xxxx",

 "type": "session",

 "customPermissions": false,

 "enabled": true,

 "created": "xxxx",

 "updated": "xxxx",

 "expiration": "xxxx",

 "secondsToExpiry": "xxxx"

}

Here you should remove the error from the FMEServer frontend (remove the line secondsToExpiry) and this request will update the token to the new expiration date.

@Sigbjørn Herstad​ The issue I had wasn't failing weekly like you encountered and it did run successfully afterwards with no intervention (just lost the chunk of data the failed job was supposed to process). So I'm not sure if it was the same. Good news for me, it hasn't been occurring lately. Glad you found the solution for your issue.

Hi @Sigbjørn Herstad​ ,

This is a fantastic suggestion, much easier than modifying existing workspaces as my previously found workarounds required. Thank you for sharing. I've added this as an option in the article - we can actually make this call using the HTTP Request action in FME Server Automations and have it running on a schedule, so there is no additional software needed!

Hi @kathyjbryce​ ,

I'm glad your issue hasn't been occurring recently - please do tag me in a comment if it pops up again and I'd be happy to try and help you get to the bottom of what could be causing your error!