Hi,
we're using Oracle 18c XE for a demo virtual machine together with our software. We just experienced an ORA-600 error which seems to be connected to optimizer statistics.
First couple of lines from the incident file:
[TOC00000]
Jump to table of contents
Dump continued from file: /opt/oracle/diag/rdbms/xe/XE/trace/XE_ora_1542.trc
[TOC00001]
ORA-00600: internal error code, arguments: [19004], [rightcstats out of order], [6589054136243570688], [6589054136243569664], [6589054136243570688], [6589054136243569664], [], [], [], [], [], []
[TOC00001-END]
[TOC00002]
========= Dump for incident 38897 (ORA 600 [19004]) ========
*** 2019-07-26T10:35:40.605925+02:00
dbkedDefDump(): Starting incident default dumps (flags=0x2, level=3, mask=0x0)
[TOC00003]
----- Current SQL Statement for this session (sql_id=7rugrzf1v5thq) -----
SELECT C1.HW_OID32, HWID(C1.HW_OID_LO,C1.HW_VERSION), C2.HW_OID32, HWID(C2.HW_OID_LO,C2.HW_VERSION), L.HW_FRAGMENT_ID FROM HW_LINKS L, HW_OBJECTS O, HW_COMPATIBILITY C1, HW_COMPATIBILITY C2 WHERE L
.HW_TGT_OBJ_OID_LO = :B2 AND L.HW_TGT_OBJ_VERSION IN (:B1 ,0) AND EXISTS ( SELECT 1 FROM TABLE(CAST (:B3 AS HWTABHWID)) T WHERE T.OID_LO = L.HW_SRC_OBJ_OID_LO ) AND EXISTS ( SELECT 1 FROM TABLE(CAS
T (:B4 AS HWTABID32)) LT WHERE LT.ID32 = L.HW_LINKTYPE ) AND O.HW_OID_LO = L.HW_SRC_OBJ_OID_LO AND C1.HW_OID_LO = L.HW_SRC_ANCH_OID_LO AND C1.HW_VERSION = 1 AND C2.HW_OID_LO = O.HW_OID_LO AND C2.HW
_VERSION = O.HW_DEFAULT_WRITE
[TOC00004]
----- PL/SQL Stack -----
----- PL/SQL Call Stack -----
object line object
handle number name
0x63f1baa8 3969 package body HWSYSTEM.SQLLINK_V1.HG_GETLINKS_V1
0x6ce7efb8 2 anonymous block
[TOC00004-END]
[TOC00003-END]
[TOC00005]
----- Call Stack Trace -----
calling call entry argument values in hex
location type point (? means dubious value)
-------------------- -------- -------------------- ----------------------------
ksedst()+119 call kgdsdst() 7FFE306BD468 000000002
7FFE3069EE00 ? 7FFE3069EF18 ?
7FFE306BCC60 ? 000000082 ?
dbkedDefDump()+819 call ksedst() 000000000 000000002 ?
7FFE3069EE00 ? 7FFE3069EF18 ?
7FFE306BCC60 ? 000000082 ?
Did anybody else encounter such errors? I did find a couple bug reports but they should all be fixed in 12.2 and I thought that 18c XE should include those fixes.
Is there an exact database version which was used as the base for 18c XE or is XE kernel still different from the "main stream" database kernel?
Regards
Bernhard
PS: BTW deleting schema_stats and (re)gathering them seemed to fix the problem but I don't know if and when it will show up again