Skip to main content

Hi community,

We are trying to retrieve the output of a FME Server hosted 'child' workspace using HTTPCaller (GET) transformer on the main 'parent' workspace.

This seems to work perfectly when running the 'parent' workspace on local machine with HTTPCaller getting the output of the 'child' workspace which is hosted on FME Server.

The problem:

When we upload this 'parent' workspace to FME Server, the HTTPCaller seems to fail. Log as below:

 

HTTPCaller (HTTPFactory): HTTP/FTP transfer error: 'Timeout was reached'

HTTPCaller (HTTPFactory): Please ensure that your network connection is properly set up

HTTPCaller (HTTPFactory): No proxy settings have been entered. If you require a proxy to access external URLs, please ensure the appropriate information has been entered

 

Is there anything we need to change so that the HTTPCaller works when published on Server too? We are using FME Cloud, so proxies should not be a problem and currently have 2 engines, so 2 workspaces should be able to run at the same time. I have also tried to set 'timeout' to 0, but the workspace will run indefinitely.

 

Main HTTPCaller parameters are as below:

URL being used: https://@@@@@.fmecloud.com/fmedatastreaming/Services/Geometry%20Validation.fmw

Any ideas as to how fix this problem would be really appreciated.

Many thanks,

Ana

Is there a reason you are using the httpCaller instead of the FMEServerJobSubmitter?


I am not that familiar with FMESererJobSubmitter but I believe you might need to write the output in a folder on Server and then retrieve those results. I didn't took this approach as I thought it was going to affect the overall time of this workspace to run as I am trying to optimise it as much as possible. If you have experience using it, would you say it is worth trying it?


I am not that familiar with FMESererJobSubmitter but I believe you might need to write the output in a folder on Server and then retrieve those results. I didn't took this approach as I thought it was going to affect the overall time of this workspace to run as I am trying to optimise it as much as possible. If you have experience using it, would you say it is worth trying it?

I use it when chaining multiple workspaces together via a triggered controller workspace.

 

 

I don't use the dataStreamingService all that much, I couldn't tell you what the difference in run time is compared to the jobsubmitter.

I use it when chaining multiple workspaces together via a triggered controller workspace.

 

 

I don't use the dataStreamingService all that much, I couldn't tell you what the difference in run time is compared to the jobsubmitter.

Many thanks, I might give it a go if I am unable to make HTTPCaller works.


Many thanks, I might give it a go if I am unable to make HTTPCaller works.

Do you have different queues for the two engines? Are the parent and child workspace in the same queue?


Do you have different queues for the two engines? Are the parent and child workspace in the same queue?

Yes, the parent workspace will run in one engine whereas the other engine will have the queue ready for the child workspace to be run.