Fractured Block/Block Corruption with DNFS in 19c

刘泽宇

现象：
Database report frequent block corruption with DNFS configured in 19c.

Database alert logs shows corruption:

＜ORACLE_BASE＞/diag/rdbms/＜db_name＞/＜instance_name＞/trace/alert_＜instance_name＞.log
==============================================================================================
Corrupt block relative dba: 0x327236ed (file 201, block 3290861)
Fractured block found during buffer read
Data in bad block:
type: 6 format: 2 rdba: 0x327236ed
last change scn: 0x0000.074c.bdee778e seq: 0x1 flg: 0x02
spare3: 0x0
consistency value in tail: 0x006f0021
check value in block header: 0x0
block checksum disabled
dbwriter trace file shows:

＜ORACLE_BASE＞/diag/rdbms/＜db_name＞/＜instance_name＞/trace/＜instance_name＞_dbw1_<PID>.trc
=============================================================================================
[76284619665192] kgnfs_flushmsg:9502: CH OUT of ORDER SEND m->order 229595
ch->order 229688 ch 70001006383c2c0
[76284619665518] kgnfs_flushmsg:9502: CH OUT of ORDER SEND m->order 229595
ch->order 229688 ch 70001006383c2c0
[76284619665531] kgnfs_flushmsg:9502: CH OUT of ORDER SEND m->order 229596
ch->order 229688 ch 70001006383c2c0
[76284619665539] kgnfs_flushmsg:9502: CH OUT of ORDER SEND m->order 229597
ch->order 229688 ch 70001006383c2c0
CHANGES
Database were recently upgraded to 19C.

原因：
If a write operation submitted by dnfs to storage does not not complete on time on account of delays encountered by the storage appliance,Subsequent retry of incomplete write operation we don't maintain ordering of said messages properly.End result is often database block corruption.

This was investigated in
BUG 32931941 - DNFS BLOCK CORRUPTION WHEN ATTEMPTING RETRY OF A WRITE IN 19C

处理方法：
Bug 32931941 is fixed in future release. Apply interim patch 32931941, if available for your platform and Oracle version.If no patch exists for your version, please contact Oracle Support for a backport request.