Friday, January 26, 2024

Exadata Scrubbing

 

1. Появление сбойных секторов на диске автоматически запускает дополнительный scrubbing:

8.2.4 Adaptive Scrubbing Schedule
In release 12.1.2.3.0, if a bad sector is found on a hard disk in a current scrubbing job, Oracle Exadata System Software will schedule a follow-up scrubbing job for that disk in one week. When no bad sectors are found in a scrubbing job for that disk, the schedule will fall back to the scrubbing schedule specified by the hardDiskScrubInterval attribute.If the user has changed the hardDiskScrubInterval to less than or equal to weekly, Oracle Exadata System Software will use the user-configured frequency instead of the weekly follow-up schedule even if bad sectors are found.

Exadata adaptively and automatically increase the frequency of scrubbing on that disk until all corruptions are repaired.

2. Scrubbing - не применяется к Flash-селлам

3. Работу по восстановлению избыточности выполняет ASM:  
If scrubbing detects a sector is corrupted, the storage server requests ASM to repair the sector from one of the mirrors on another storage server. This is reason why multiple ASM-mirrors are essential.
Получается, что если ASM остановлен, то Scrubbing не может исправить данные на дисках

4. Scrubbing is an automated process on Exadata that kicks in when the disks are idle ( less than 25% busy )

5. How do you see if scrubbing is in action?

CellCLI> list metriccurrent where name = 'CD_IO_BY_R_SCRUB_SEC' and metricObjectName like 'CD.*'

CD_IO_BY_R_SCRUB_SEC    CD_00_exadbm01celadm01  115 MB/sec  <<< scrubbing sectors at a rate of around 115MB/s.
CD_IO_BY_R_SCRUB_SEC    CD_01_exadbm01celadm01  118 MB/sec
CD_IO_BY_R_SCRUB_SEC    CD_02_exadbm01celadm01  117 MB/sec

     ...
The cell above represents an idle cell. If it were under load, the values on the right would drop to 0 MB/sec.


6. Hard disk drives in the High Capacity storage servers connect to a disk controller, which includes a 2G cache. AWR is reporting the number of IOPS serviced by the disk controller cache, not the physical IOPS serviced by the disk itself.

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.

Does DEALLOCATE UNUSED or SHRINK SPACE will free space occupied by LOB segment?

Lets check how it works. My env is DB 19.20@Linux-x64 1) I created the table with 4 LOB columns of 4 different LOB types: BASICFILE BLOB, BA...