Hi ,

Database is Oracle 8.1.4 on Sun Solaris 5.8, OLTP environment.

we have a table of 10mb which is frequently used, Its inserted , deleted, selected.

But yesterday we got errors on four similar trace files depicting the same error. Trace file content is displayed below.

================================

*** SESSION ID:(320.34060) 2004-06-16 20:11:13.499
FATAL ERROR IN TWO-TASK SERVER: error = 12592
*** 2004-06-16 20:11:13.515
ksedmp: internal or fatal error
Current SQL statement for this session:
select count(recordhandle) from iway_jmssessionstorejmsstore
----- Call Stack Trace -----
calling call entry argument values in hex
location type point (? means dubious value)
-------------------- -------- -------------------- ----------------------------
ksedmp()+168 CALL ksedst()+0 188 ? 1 ? FFBED670 ?
FFBED114 ? FFBED0F8 ? 0 ?
opitsk()+5344 CALL ksedmp()+0 2 ? 0 ? 0 ? 800 ? 941D5C44 ?
0 ?
opiino()+1792 CALL opitsk()+0 0 ? 1A15E10 ? 1A1888C ? 0 ?
0 ? 3130 ?
opiodr()+5748 PTR_CALL 00000000 FFBEF9F8 ? 80005000 ?
1A17000 ? 1A4E560 ? 0 ?
1A02E80 ?
opidrv()+1308 CALL opiodr()+0 16632B0 ? FFBEF5FC ?
1A18400 ? 0 ? 16983CA ?

Kindly adivce on what could be the error and how to recitfy it.
They have an index in thsi table whihc is just three columns. Is the index necessary , cos per second three to five records gets inserted and deleted. So at a specific point only max of 5 records will be there. But suddenly the waits have increased and at peak times around 1500 records stay in as a stack queue.

Your views are highly appreciated