Create HOL record gets "failed to execute"; high CPU for Oracle
- Article Type: General
- Product: Aleph
- Product Version: 20, 21, 22, 23
The following problems were reported by libraries in this consortium:
" We go through the steps to create a record, in the final step after clicking the Create button, the system keeps looking, then we get the error "Server failed to execute request after 200 seconds. "
" Whenever we try to save a change to a holding record we get 'Remote Service Error (c0203 update-chk 98)' Server failed to execute request after 200 seconds." I have tried making holding record changes on a couple of different computers and used different titles and all produce the same error message. Changes to the bibliographic and item records function normally and do not produce the same error. "
The DBA's reported, "CPU utilization on the Aleph server is very high and climbing. The load is up to 36 and climbing. Oracle is the top process running."
The "top" command showed that the first 30 processes (in terms of cpu usage) were all "oracle": no "rts", no "pc_server_main", no "z39_server", etc. And that some of these oracle processes had very high CPU time (such as 850 seconds).
Based on the experience in KB 16384-38305 (-- which required restart of oracle --), I felt that this situation also required restart of Oracle (and Aleph).
After stopping Aleph, restarting Oracle, and restarting Aleph: I was able to create an HOL record successfully; "top" looked normal; and other system functions were also OK.
In another case (SF Case 00491835), where the customer was getting the 'Remote Service Error (c0203 update-chk 98)', they found that deleting the 856 field from the record solved the problem: there was some defect in the 856 field which was causing the problem.
- Article last edited: 10/8/2013