In case of multiple modifications before the next checkpoint, only the first before-image is logged in the physical log.
The database server stores the before-images in the physical log only until the next checkpoint. To control the amount of data that the database server logs, you can tune the checkpoint interval configuration parameter CKPTINTVL.
The following dbspace page modifications are not physically logged:
This situation usually occurs when the page was used by some other database server or a table that was dropped.
When fast recovery begins, the database server logs the following message:
Physical recovery started at page chunk:offset.
When the fast recovery completes, the database server logs the following message:
Physical recovery complete: number pages examined, number pages restored.
If the number of pages examined is much larger than the number of pages restored, increase the size of the buffer pool to reduce the number of duplicate before-images. For more information, see the messages appendix in the IBM Informix Administrator's Reference.
The database server pages in the physical log can be any database server page, including simple large objects in tblspaces. Even overhead pages (such as chunk free-list pages) are copied to the physical log before data on the page is modified and flushed to disk.
Overhead pages also include blobspace free-map pages and blobspace bit-map pages. Blobspace blobpages are not logged in the physical log. For further information about blobspace logging, see Logging Blobspaces and Simple Large Objects.
The user-data portion of smart large objects is not physically logged. However, the metadata is physically logged. For information on smart large objects, see Sbspaces.
Home | [ Top of Page | Previous Page | Next Page | Contents | Index ]