Problem
After migrating from Studio Enterprise Edition 6.0 to Enterprise Server 2.3 running a transaction appears to hang/loop and eventually a Return Code of 210 appears in the MFDEBUG log.
Resolution
There can be a couple of reasons for the RC210.
Whenever a job starts and ends, mfbsijcl will write a record into the .Sem file. When mfbsijcl is running, it constantly reads this .Sem file checking for the ‘02’ record (Which indicates job ended). The RC210 is essentially saying that the files was read until we reached the timeout but never read/found the job-ended record.
In current versions (3.0 onwards), there is a timestamp in the .Sem file that is used to determine if it is an old file. If it is an old file, it is disregarded and no longer read. This can cause problems in itself where the record is written but the timestamp is incorrect. That is not what is happening here (as this is version 2.3), but it is something to be aware of in later versions.
In this case, there are no timestamps, so that is not an issue. What is a problem is the fact there is an EXTFH file defined with STRIPSPACE=OFF.
Mfbsi relies on the default behaviour of STRIPSPACE=ON. If STRIPSPACE=OFF is not required, simply remove it from the extfh file. Alternatively, and a better solution, is to simply exclude .Sem files from the STRIPSPACE option. So add the following to the EXTFH:
[*.Sem]
STRIPSPACE=ON
This should enable STRIPSPACE for all .Sem files and resolves the issue
#Server
#EnterpriseServer
#Enterprise
