Notes on “Recovery Manager – Performance Troubleshooting”


Some notes taken during the webinar called Recovery Manager – Performance Troubleshooting.

Need timestamps in RMAN logs:

export NLS_DATE_FORMAT="YYYY/MM/DD HH24:MI:SS"

Three (3) main areas of investigation:

  1. RMAN Metadata Query Performance
    • TARGET
    • CATALOG
  2. Disk I/O
  3. Tape I/O

Can use TRACE. Send to separate trace file.

Use 10046 traces on both TARGET and CATALOG. Use ORADEBUG for CATALOG connection.

Known issues in 247611.1

Deep packet inspection by firewalls can cause RMAN to hang.

LARGE POOL is used if native asyncio is disabled. Buffers are allocated in PGA.

Disk backups use Unused Block compression. SBT reads all blocks.

Backup Rate = DB Size/backup time

Calculate input I/O rate as follows:

BACKUP VALIDATE DATABASE;

DIRECT I/O should be used, if supported. (555601.1)

TAPE I/O performance – V$SYSTEM_EVENT.

H/W multiplexing can impact RESTORE performance. RESTORE DATABASE would require multiple scans of the tape. Disk staging area important.

Monitoring RMAN:

  • V$SESSION_LONGOPS
  • V$BACKUP_ASYNC_IO
  • V$BACKUP_SYNC_IO
Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s