Exadata : ORA 600 [kfk_iodone_invalid_buffer] [IO buffer does not pass HARD checks] When Switching to MAXIMUM AVAILABILITY Data Guard mode (Doc ID 2657346.1)

Last updated on APRIL 17, 2023

Applies to:

Symptoms

Primary RAC database 19c release crashed with below ORA 600 error ::

2020-04-01T15:19:57.044526+02:00
Redo log for group 45, sequence 509 is located on DAX storage                       <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< Incorrectly think this is DAX storage while it is not.
Errors in file /u01/app/oracle/diag/rdbms/testdb/testdb1/trace/testdb1_lgwr_271454.trc  (incident=344355):
ORA-00600: internal error code, arguments: [kfk_iodone_invalid_buffer], [IO buffer does not pass HARD checks], [2], [0], [], [], [], [], [], [], [], []     <<<<<<<<<<
Incident details in: /u01/app/oracle/diag/rdbms/testdb/testdb1/incident/incdir_344355/testdb1_lgwr_271454_i344355.trc
Use ADRCI or Support Workbench to package the incident.
See Note 411.1 at My Oracle Support for error and packaging details.
ERROR: unrecoverable error ORA-600 raised in ASM I/O path; terminating process 271454
Process termination requested for pid 271454 [source = rdbms], [info = 2] [request issued by pid: 271454, uid: 5041]

Changes

Changing Data Guard mode from MAXIMUM PERFORMANCE to MAXIMUM AVAILABILITY.

NOTE:

• Change to MAXIMUM PROTECTION does not hit the ORA 600.
• 12c database within same Exadata rack does not hit the ORA 600.
• ORA 600 is hitting LGWR background process.
• Online or standby redo log files are resided either on Extreme Flash Cell node or High Capacity Cell node.
• Following patch is present or applied : Patch 30165493 - LOG FILE FAST SYNC PARAMETERS NEED TO BE FIXED FOR PMEMLOG

Cause

My Oracle Support provides customers with access to over a million knowledge articles and a vibrant support community of peers and Oracle experts.