This error is caused by a nice Oracle documentation typo: the enumerated constant StatelessWithFastRestPackageState is misspelled in table 37-19 (in both 10.2 and 11.1 DBMS_EPG chapters of PL/SQL Packages and Types Reference), but is correctly reported elsewhere (too bad I read the wrong one first...).
begin dbms_epg.set_dad_attribute('APEX','session-state-management','StatelessWithFastRestPackageState'); commit; end; / Error starting at line 3 in command: begin dbms_epg.set_dad_attribute('APEX','session-state-management','StatelessWithFastRestPackageState'); commit; end; Error report: ORA-31038: Invalid enumeration value: "StatelessWithFastRestPackageState" ORA-06512: at "SYS.DBMS_EPG", line 89 ORA-06512: at "SYS.DBMS_EPG", line 222 ORA-06512: at line 2
As you can easily guess, an "e" must be inserted: StatelessWithFastResetPackageState
By the way, according to the HTTP server documentation (referring to Apache as a matter of fact) the default value for this attribute should be
StatelessWithResetPackageState, however the best value for Oracle XE (as well as Oracle 9 and above) should be StatelessWithFastResetPackageState.
And now for the real question is: how am I going to measure any benefit from this change?
The short answer is: I don't know, but I hope to see lower elapsed_time values in apex_workspace_activity_log.
See message translations for ORA-31038 and search additional resources.