Exadata: Random disk offline issues along with ORA-07445:[_int_malloc] and ORA-07445[_int_free] errors. (Doc ID 2062786.1)

Last updated on OCTOBER 09, 2015

Applies to:

Oracle Database - Enterprise Edition - Version 11.2.0.3 to 11.2.0.4 [Release 11.2]
Information in this document applies to any platform.

Symptoms


Customer may see random Read or Write failures in the RDBMS instances which use shared server configuration.  As a result ASM may take the disk offline temporaily. This happens without any HW issues or IO errors reported on the storage side.

It is possible that multiple disks go offline with in a short span of time, and happen to be the partner disks in the diskgroup. This can result in the dismount of the diskgroups and cause the DB instances to crash, causing an outage situation.

[[ DB Instance alert log file ]]

WARNING: Read Failed. group:2 disk:72 AU:0 offset:0 size:4096
path:o/192.168.20.29;192.168.20.30/DATA_XDB5_CD_00_xdb5cel07
incarnation:0xe969bf97 synchronous result:'I/O error'
subsys:OSS iop:0x7ffd9ee99140 bufp:0x7ffd9eccf400 osderr:0x18 osderr1:0x0

ERROR: cannot read disk header of disk DATA_XDB5_CD_00_XDB5CEL07 (72:3916021655) 
NOTE: process _l395_ocdbqs1 (5459) initiating offline of disk 72.3916021655 (DATA_XDB5_CD_00_XDB5CEL07) with mask 0x7e[0x7] in group 2

DB alert log file would show intermittent ORA-07445:[_int_malloc()+632] and ORA-07445[_int_free()+1633] errors reported for shared server processes, due to <bug 16817656> 

 

Cause

Sign In with your My Oracle Support account

Don't have a My Oracle Support account? Click to get started

My Oracle Support provides customers with access to over a
Million Knowledge Articles and hundreds of Community platforms