Thank You Pete. We are not seeing any signs of IO spike or Disk utilization
being saturated in the storage cells during this time. Below are the
database details.
It's Oracle 19c. And Exadata machine.
Disk_asynch_io - True, filesystemio_options-None.
Platform- linux X86 64bit, CPU - 96, Cores- 48, Sockets-2.
And also just to note , as I am unable to fetch the dash_wait_chain from
during the exact peak period as it is failing with error 'result
concatenation is too long' , so not sure if the above
supplied dash_wait_chain output from immediate before the issue period is
reliable to what extent.
On Fri, Mar 3, 2023 at 10:16 PM Peter Hitchman <pjhoraclel@xxxxxxxxx> wrote:
Hi,
I think some more detail about your set-up will help to get some answers:
Oracle version
Operating system
Number of CPUs
and
The database I/O configuration (disk_asynch_io and filesystemio_options)
and the set-up of the storage being used.
I think that this points to the I/O subsystem not being able to keep up
with the load it is being given.
Regards
Pete
