Hi Holly!
In my case it actually helped taking away the xml file before it even gets in to the (generic) featurereader. I used the zipextractor to unzip and then a tester to route the file away. Which worked well in this script since I didn't needed the xml file. However it might for some other people be a need for it later, but maybe the particular file can be routed in the same way and then processed directly in to an xml-featurereader :)
Great, I'm glad to have helped somewhat. If you are able to discover what component in your workflow is causing the "A fatal error has occurred" message to be displayed after the Terminator, then please do let me know. It would be good to update the issue to include known causes other than the <rejected> port one.
Hi Holly!
Thanks for the feedback, it makes more logic sense now:) Im not sure if its the problem with features going into the rejected port with this case, the terminator . We will rather try to access the full log and search through it, so we can write out the correct termination message. It seems as a weakness if we cant rely on how the log looks like in the end (I guess other unpredicted errors can come that we might not know about), and that will confuse the end-user. Ill come back here if we need some more information, but im very grateful for this insight!
/Anne
Thank you for providing those log files. In the instance where the terminator message is not being displayed as the statusMessage this is not because of the failure to read XML but it is because that error is also reported at the bottom of the log underneath the Terminator message.
This is the key problem on FMEServer because the Result Data: statusMessage only returns the last error that is reported in the log.
It appears this is a known issue (FMEENGINE-48688) in certain workflows, previously this has occurred if the Terminator is connected to the <rejected> port of a transformer somewhere along the translation. If this is not the cause in your workflow then it would be useful for us to to review your .fmw to identify what else causes this additional error message to be printed. Please let me know if you are happy to provide this.
I tried again with another file that also contains xml-file, and here it actually showed the correct termination message in the result data window, but looking at the log file it also complained about the xml-file, so no Im not sure anymore what influences the statusmessage. job_25344.txt
I seem to have interpreted it wrog. I now actually believe it is because it does not read the xml-file. And that the message which returns to statusmessage is caused by that.
This is not the log from the screenshot above, but the same is happening here. What I can read is that it had some problems with reading source data, but was able to solve it. Do you think this might influence what I get in the statusmessage? Because otherwise I understand the logfile that it terminated because of the terminator_5. And in the result data it only said: A fatal error has occurred. Check the logfile above for details.
| Unable to recognize the source dataset as an ADAC file, 'C:\Apps\FMEServerSystemShare\resources\temp\RAA\TF11_V9_UTM33N_shp\Point_lamningar_UTM33N.shp.xml' |
| A fatal error has occurred. Check the logfile above for details |
| A fatal error has occurred. Check the logfile above for details |
As david_r mentioned you cannot manipulate the end message displayed here. However if you are using a Terminator to end the translation the Termination Message should be displayed in the result data as you are expecting. I have tested in b17725 and can confirm I get the desired result.
Based on the error message returned in your case this indicates to me that your translation is failing before any features hit the Terminator - as the error message shown is quite standard in logging.
Please could you provide the full log file for this job so we can check where in the workspace the job is failing?
The user seem som suggest the prefix (transformer name I guess) removed (then I assume that is because the prefix is alrady there?)
Unfortunately, the post you link to is just an idea for improvement, not a description on how it is/was.
The good news is that you can vote for the idea, the more votes, the more likely it is to get implemented :-)
As I understand it was possible before: https://knowledge.safe.com/idea/22972/terminator-improvements.html
"When a job terminates on Server, it passes the Termination Message on to the statusMessageattribute from the job result. Just like in workbench, it always prefixes this message with "<Transformer Name>: [......] "
As far as I know you cannot influence what is shown in the server UI. You will have to look at the job log for more details.