Got confirmation from the Dev team stating its happening after moving to
19C. So wanted to understand if any specific changes in 19C is making this
happen and we need to have some different treatment/setup for the memory
component is required?
Regards
Lok
On Fri, Mar 12, 2021 at 3:03 PM Lok P <loknath.73@xxxxxxxxx> wrote:
Hello Listers, We have one Dev database with version- 19.9.0.0.0 and its
on Exadata. We are seeing below errors and during that time we see
select/insert query running slow unexpectedly and also then nobody is able
to login to the database after some time. So we end up restarting the
database to mitigate this. This has happened many times unexpectedly.
Seeing a matching bug as below with similar symptoms which states
combination of hugepages + AMM can cause this. But I verified gv$parameter
, we don't have a memory_target set which means we are not using AMM rather
we have a sga_target set which means we are on ASMM only. So wanted to
check if anybody encountered such an error and how to fix it?
Error:-
Ora-04031 : unable to allocate 12312 bytes of shared memory ("shared
pool","unknown object","KKSSP^1283","kglseshtTable")
Mos Doc:-
ORA-4031 Errors On ASM Instance When Huge Pages Are Enabled On The System
(Doc ID 1625886.1)
