Job Search is not working in E-Recruitment

Question

Hi All,
When I used to search the Job with the link "Job Search"  as an internal candidate that has been published already with the released status, am getting an error message as "An Internal error occurred". As with the forum, i checked with the job RCF_PERIODICAL_SERVICES has been scheduled or not. First it was not maintained, then i maintained it. TREX is already configured and checked with the transaction SKPR07 and checked with the transaction SRMO for RFC destinations and also for the Index category but still i couldn't able to search the Job in Job Search link.
But i can able to find the Job that has been published using "Apply Directly". Please give me the solution to fix the issue.
Thanks in Advance,
Magesh S.

Answer 1

Hello Laura,
usually going through the note does the trick. If not, it is difficult to say from the distance. I had some issues within an EhP4 system that some other configuration error resulted in a SLG1 message for the search but so far I have not found a pattern which could cause the error other than the settings mentioned in the note.
So you will probably have to dig into the system to get to the reason. Depending on your project position you might need some developer support here. First of all you have to find out when / where the error is written into SLG1 first general question - is it within the periodical services or is a frontend user doing something in the application which writes the logs. Usually you can see in the log entry which user caused it. If you have a determined batch user it is easy to find out otherwise you might need to check log time or temporarly assign a user not in the frontend.
If it is in the periodical services I would deactivate them and after some time (depending how often the error appeared you need to wait till it is highly probable that a new service run writes a log) i would run each service at once. Therefore you get into periodical service customizing get the classes for the services and run the "run"-Method in SE80 / SE24 class test. After each class check if there is a SLG1 entry to get the right service. If you have the service I'd wait again, put a breakpoint into cl_hrrcf_exception_handler method write_exception_log (central method writing SLG1 entries) run the services again via the class test and check where the error is caused. From here it is common developer work to get the reason.
If the SLG1 entry is caused by the frondend interaction I'd directly start with debugging in the mentioned method.
Sorry to be not able to offer an instance solution.
Best Regards
Roman