X

Alejandro Vargas' Blog

  • December 10, 2006

ASM Resilience Tests

Alejandro Vargas
Technical Leader, ACS Global Delivery, Infrastructure & BigData

These scenarios were run several times in a Single Instance 10g R2 database on Linux RH3.
The result was satisfactory in all cases.

All these scenarios were run together with a script that simulates heavy dml.

1- ASM instance failure while add disk in progress
 

1) run load scripts :  /scripts/ins_tab.csh &
2) add disk to datadg
3) check that rebalance is in progress
4) kill asm instance
5) check that both databases are down
6) start ASM, check views
7) Start DB
8) check asmcmd: lsdg, cd datadg, ls �l testasm
9) check that rebalance continue

2- ASM instance failure while drop disk in progress

1) wait until rebalance after add disk complete. run load scripts : /scripts/ins_tab.csh &
2) drop disk on datadg  (ALTER DISKGROUP DATADG DROP DISK DATADG_0001)
3) check that rebalance is in progress
4) kill asm instance
5) check that both databases are down
6) start ASM, check views
7) Start DB
8) check asmcmd: lsdg, cd datadg, ls �l testasm
9) check that rebalance continue

3- ASM instance failure while add and drop disks are in progress simultaneously.


1) run load scripts :  /scripts/ins_tab.csh &
2) add disk on datadg  ALTER DISKGROUP DATADG ADD DISK DATADG_0001, check that rebalance start.
3) drop disk on datadg  ALTER DISKGROUP DATADG DROP DISK DATADG_0002
4) check that rebalance is in progress. Rebalance for add seems to stop. New rebalance process start from 0.
5) kill asm instance
6) check that both databases are down
7) start ASM, check views :select * from v$asm_disk, done twice.
8) Start DB
9) check asmcmd: lsdg, cd datadg, ls �l testasm
10) check that rebalance continue

4. ASM server failure while a disk removed from one disk group is added to second disk group, and rebalance is in progress.

1) run load scripts :  /scripts/ins_tab.csh &
2) add disk on archdg  ALTER DISKGROUP DATADG ADD DISK ARCHDG_0002, check that rebalance start.
3) drop disk on archdg  ALTER DISKGROUP DATADG DROP DISK ARCHDG_0001, check that rebalance start and finish.
4) add the disk removed from archdg on backdg  ALTER DISKGROUP DATADG ADD DISK BACKDG_0002, check that rebalance start.
5) reboot the server, wait until the server is up again.
6) start ASM, check views :select * from v$asm_disk, check twice.
7) Start DB
8) check asmcmd: lsdg, cd datadg, ls �l testasm
9) check that rebalance continue

Be the first to comment

Comments ( 0 )
Please enter your name.Please provide a valid email address.Please enter a comment.CAPTCHA challenge response provided was incorrect. Please try again.