My Oracle Support Banner

ORA-27072 Errors When Adding, Dropping ASM Disks or Running Rebalance Operation. Also OS logs show FCPIO_SGL_INVALID (Doc ID 1496269.1)

Last updated on AUGUST 04, 2018

Applies to:

Oracle Server - Enterprise Edition - Version 11.2.0.3 and later
Information in this document applies to any platform.

Symptoms


2 node RAC - 11203 - Linux x86-64 (SLES11) - ASMLIB - Storage: Cisco server using FCoE technolog - multipath (emcpower)

Receiving read errors on all ASM luns when dropping / adding or rebalancing luns within a diskgroup

ASM alert log shows:

Thu Sep 20 15:37:04 2012 --> SQL> alter diskgroup DATA01 drop disk'DATA01_0000'
Thu Sep 20 15:37:09 2012 --> SUCCESS: alter diskgroup DATA01 drop disk'DATA01_0000'
Thu Sep 20 15:37:09 2012 --> ARB0 started with pid=36, OS id=68835
                             /oracle/admin/diag/asm/+asm/+ASM1/trace/+ASM1_arb0_68835.trc:
                             ORA-27072: File I/O error
                             Linux-x86_64 Error: 5: Input/output error
                             Additional information: 4
                             Additional information: 4096
                             Additional information: -1
                             WARNING: Read Failed. group:2 disk:0 AU:2 offset:0 size:1048576
                             NOTE: group 2, file 1, extent 0: successful reread of 256 block(s)
                             NOTE: F1X0 copy 1 relocating from 0:2 to 1:62340 for diskgroup 2 (DATA01)
                             ...
                             ... <errors continue for different AUs:>
                             Additional information: 6144
                             WARNING: Read Failed. group:2 disk:0 AU:3 offset:0 size:1048576
                             ...
                             Additional information: -1
                             Additional information: 1048576
                             WARNING: Write Failed. group:2 disk:1 AU:62390 offset:0 size:1048576
                             ERROR: failed to copy file +DATA01.256, extent 1
                             ERROR: ORA-15080 thrown in ARB0 for group number 2
                             Errors in file /oracle/admin/diag/asm/+asm/+ASM1/trace/+ASM1_arb0_68835.trc:
                             ORA-15080: synchronous I/O operation to a disk failed
                             NOTE: stopping process ARB0
                             ...
                             NOTE: starting rebalance of group 2/0x7217e974 (DATA01) at power 1
                             Starting background process ARB0
Thu Sep 20 15:42:06 2012 --> Received dirty detach msg from inst 2 for dom 5
Thu Sep 20 15:42:06 2012 --> List of instances:  1 2
Thu Sep 20 15:52:42 2012 --> SQL> alter diskgroup DATA01 add disk '/dev/oracleasm/disks/ASM_DATA_01'
Thu Sep 20 15:52:55 2012 --> SUCCESS: alter diskgroup DATA01 add disk '/dev/oracleasm/disks/ASM_DATA_01'
Thu Sep 20 15:52:55 2012 --> ARB0 started with pid=36, OS id=73279
Thu Sep 20 15:57:03 2012 --> Received dirty detach msg from inst 2 for dom 3
Thu Sep 20 15:57:03 2012 --> List of instances:  1 2

 
The following entries also appear at those time in the OS system logs (/var/log/messages*)

Sep 20 15:37:09 msp0lracd018 kernel: [17018.656624] scsi host2: hdr status = FCPIO_SGL_INVALID
...
Sep 20 15:37:09 msp0lracd018 kernel: [17018.657390] scsi host4: hdr status = FCPIO_SGL_INVALID
...
Sep 20 15:37:11 msp0lracd018 [17020.621145] scsi host2: hdr status = FCPIO_SGL_INVALID

 
From the ASM alert log entries, ARB is re-reads the information successfully (i.e., 'successful reread of 256 block(s))
When there is a write error the process is stopped, but the ASM disk group is not dismounted.

.

Cause

To view full details, sign in with your My Oracle Support account.

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


In this Document
Symptoms
Cause
Solution
References

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