Home » Server Options » RAC & Failsafe » Oracle cold failover cluster (oracle, 12cR2, RHEL6.5)
Oracle cold failover cluster [message #669937] Fri, 25 May 2018 00:42
Prathikesh
Messages: 20
Registered: February 2015
Location: Hyderabad
Junior Member
Hi Leaders,
This is regarding Oracle cold failover cluster setup(Alias Oracle Active and Passive DB setup). As part of this oracle HA setup, we have two nodes say Node1 and Node2. We have the common SAN storage mounted thru network on both the nodes. All the DB related files located under this SAN storage itself.

Now the issue is, there was some network drop for few minutes. So, SAN storage was not available to nodes and cluster. After some time, we found both the nodes got shutdown. Below are the details found in alert log.

Errors in file /home/app/oracle/diag/rdbms/hwprod/hwprod/trace/hwprod_dbw4_24394.trc:
ORA-15080: synchronous I/O operation failed to write block 288 of disk 0 in disk group DG_HWPROD_DATA
ORA-15186: ASMLIB error function = [kfk_asm_ioerror],  error = [0],  mesg = [I/O Error]
Thu May 24 02:45:09 2018
WARNING: failed to write mirror side 1 of virtual extent 0 logical extent 0 of file 259 in group 1 on disk 0 allocation unit 374
WARNING: group 1 file 259 vxn 0 block 288 write I/O failed
KCF: read, write or open error, block=0x120 online=1
        file=4 '+DG_HWPROD_DATA/hwprod/undotbs01.dbf'
        error=15081 txt: ''
Thu May 24 02:45:09 2018
Errors in file /home/app/oracle/diag/rdbms/hwprod/hwprod/trace/hwprod_dbw4_24394.trc:
ORA-15186: ASMLIB error function = [kfk_asm_ioerror],  error = [0],  mesg = [I/O Error]
Thu May 24 02:45:09 2018
Errors in file /home/app/oracle/diag/rdbms/hwprod/hwprod/trace/hwprod_arc0_24486.trc:
ORA-00204: error in reading (block 1, # blocks 1) of control file
ORA-00202: control file: '+DG_HWPROD_DATA/hwprod/control01.ctl'
ORA-15081: failed to submit an I/O operation to a disk
ORA-15186: ASMLIB error function = [kfk_asm_ioerror],  error = [0],  mesg = [I/O Error]
Thu May 24 02:45:10 2018
Errors in file /home/app/oracle/diag/rdbms/hwprod/hwprod/trace/hwprod_arc1_24488.trc:
ORA-00204: error in reading (block 1, # blocks 1) of control file
ORA-00202: control file: '+DG_HWPROD_DATA/hwprod/control01.ctl'
ORA-15081: failed to submit an I/O operation to a disk
ORA-15186: ASMLIB error function = [kfk_asm_ioerror],  error = [0],  mesg = [I/O Error]
Thu May 24 02:45:10 2018
Errors in file /home/app/oracle/diag/rdbms/hwprod/hwprod/trace/hwprod_ckpt_24400.trc:
ORA-00202: control file: '+DG_HWPROD_DATA/hwprod/control01.ctl'
ORA-15081: failed to submit an I/O operation to a disk
ORA-15186: ASMLIB error function = [kfk_asm_ioerror],  error = [0],  mesg = [I/O Error]
Thu May 24 02:45:10 2018
Errors in file /home/app/oracle/diag/rdbms/hwprod/hwprod/trace/hwprod_arc2_24490.trc:
ORA-00202: control file: '+DG_HWPROD_DATA/hwprod/control01.ctl'
ORA-15081: failed to submit an I/O operation to a disk
ORA-15186: ASMLIB error function = [kfk_asm_ioerror],  error = [0],  mesg = [I/O Error]

So, Can you please check and confirm on below? Thank you.
Quote:
Since both nodes are physical blades, and they're attached via iSCSi for certain partitions, is it possible the Oracle Cluster manager, noticing a loss of iSCSi connection due to network drop would reboot the system
Previous Topic: [INS-30512] Automatic Storage Management software is not configured on this cluster
Next Topic: Root.sh failed: ORA-39510: CRS error performing start on instance '+ASM1' on '+ASM'
Goto Forum:
  


Current Time: Thu Mar 28 17:00:33 CDT 2024