Hi @Jan T

It should be possible to obtain a detailed error from reader by increasing the log level(You can set it to DEBUG) for the connect reader and looking into the log entries when the issue occurs. Depending on the error details you should be able to narrow down the issue and take the actions to mitigate it. 

Thanks,

Kasun

Hi @Jan T,

Please also check if there is a file named SFTP_READER_BANKING_FI_problem_files.xml in \\<IFS_HOME>instance\<InstanceID>\logs folder in Application Server and see if the file names that are not being processed can be found in this file. 

Since you have mentioned the file gets processed once you rename it, this could be the issue. Usually a file is added to problem files when the file is corrupted.

Hope this helps!

The file name contains the date and is such unique, as we only receive one file per day. If it is added to problem files, it is doing so when the original read fails.

Can it be that the SFTP reader reads the file when it is still being written? Then the file is corrupt when it is reading it. It would then mark it as a problem file and works when I rename it    

Hi @Jan T,

Such an issue existed in IFSAPP8, where the reader reads the file while it is still being written and as a result marks the file as problematic, but it was patched. Have a look at LCS patch 120672. Haven’t seen this kind of an issue in IFSAPP9 or IFSAPP10.

I cannot think of any configuration changes that can be done to remedy something like this, off the top of my head. You can try increasing the retry limit and see if it helps. Unless, it would be a good idea to get some help from R&D if the issue persists. 

Hope this helps!

The debug level is set on. But strangely there is nothing in the log at the time when the load fails and the Connect reader queue entry is created.
It is still a mystery. Now it has worked fine 8 days in a row, In total it has failed 15 days in 6 weeks.