新装一套19c rac,使用了10月份补丁。其中一个节点的告警日志一直报以下错,cdump把oracle_home目录快撑满了。

DDE: Problem Key 'ORA 7445 [pevm_icd_call_common]' was completely flood controlled (0x6)
Further messages for this problem key will be suppressed for up to 10 minutes
2020-12-31T13:32:37.626035+08:00
Exception [type: SIGSEGV, SI_KERNEL(general_protection)] [ADDR:0x0] [PC:0x1289FAA1, pevm_icd_call_common()+225] [flags: 0x0, count: 1]
2020-12-31T13:33:37.669324+08:00
Exception [type: SIGSEGV, SI_KERNEL(general_protection)] [ADDR:0x0] [PC:0x1289FAA1, pevm_icd_call_common()+225] [flags: 0x0, count: 1]
2020-12-31T13:34:37.731930+08:00
Exception [type: SIGSEGV, SI_KERNEL(general_protection)] [ADDR:0x0] [PC:0x1289FAA1, pevm_icd_call_common()+225] [flags: 0x0, count: 1]
2020-12-31T13:35:37.772123+08:00


$ more zjcdba2_m004_9211.trc
Trace file /u01/app/oracle/diag/rdbms/zjcdba/zjcdba2/trace/zjcdba2_m004_9211.trc
Oracle Database 19c Enterprise Edition Release 19.0.0.0.0 - Production
Version 19.3.0.0.0
Build label:    RDBMS_19.3.0.0.0DBRU_LINUX.X64_190417
ORACLE_HOME:    /u01/app/oracle/product/19.0.0/db_1
System name:    Linux
Node name:      oracle-19c-pro-p002
Release:        3.10.0-957.el7.x86_64
Version:        #1 SMP Thu Nov 8 23:39:32 UTC 2018
Machine:        x86_64
Instance name: zjcdba2
Redo thread mounted by this instance: 2
Oracle process number: 149
Unix process pid: 9211, image: oracle@oracle-19c-pro-p002 (M004)


*** 2020-12-31T13:28:37.475623+08:00 (CDB$ROOT(1))
*** SESSION ID:(4365.41701) 2020-12-31T13:28:37.475667+08:00
*** CLIENT ID:() 2020-12-31T13:28:37.475675+08:00
*** SERVICE NAME:(SYS$BACKGROUND) 2020-12-31T13:28:37.475681+08:00
*** MODULE NAME:(MMON_SLAVE) 2020-12-31T13:28:37.475687+08:00
*** ACTION NAME:(ADR Container Space Management Statistics Flush) 2020-12-31T13:28:37.475693+08:00
*** CLIENT DRIVER:() 2020-12-31T13:28:37.475699+08:00
*** CONTAINER ID:(1) 2020-12-31T13:28:37.475705+08:00

*********** START A RUN OF [ADR Container Space Management Statistics Flush] *************

*** 2020-12-31T13:29:37.495253+08:00 (CDB$ROOT(1))
*** MODULE NAME:(MMON_SLAVE) 2020-12-31T13:29:37.495295+08:00
*** ACTION NAME:(KDILM background EXEcution) 2020-12-31T13:29:37.495304+08:00
*** CONTAINER ID:(1) 2020-12-31T13:29:37.495310+08:00

*********** START A RUN OF [KDILM background EXEcution] *************

*** 2020-12-31T13:30:37.537420+08:00 (CDB$ROOT(1))
*** MODULE NAME:(MMON_SLAVE) 2020-12-31T13:30:37.537466+08:00
*** ACTION NAME:(KDILM background CLeaNup) 2020-12-31T13:30:37.537476+08:00
*** CONTAINER ID:(1) 2020-12-31T13:30:37.537482+08:00

*********** START A RUN OF [KDILM background CLeaNup] *************

*** 2020-12-31T13:31:37.570206+08:00 (CDB$ROOT(1))
*** MODULE NAME:(MMON_SLAVE) 2020-12-31T13:31:37.570253+08:00
*** ACTION NAME:(KDILM background EXEcution) 2020-12-31T13:31:37.570262+08:00
*** CONTAINER ID:(1) 2020-12-31T13:31:37.570268+08:00

*********** START A RUN OF [KDILM background EXEcution] *************
*** MODULE NAME:(MMON_SLAVE) 2020-12-31T13:31:37.571295+08:00
*** ACTION NAME:(Intensive AutoTask Dispatcher) 2020-12-31T13:31:37.571312+08:00
*** CONTAINER ID:(1) 2020-12-31T13:31:37.571319+08:00

*********** START A RUN OF [Intensive AutoTask Dispatcher] *************
*** MODULE NAME:(MMON_SLAVE) 2020-12-31T13:31:37.590475+08:00
*** ACTION NAME:(Automatic Report Flush) 2020-12-31T13:31:37.590503+08:00
*** CONTAINER ID:(1) 2020-12-31T13:31:37.590509+08:00

*********** START A RUN OF [Automatic Report Flush] *************

*** 2020-12-31T13:31:37.601918+08:00 (CDB$ROOT(1))
Exception [type: SIGSEGV, SI_KERNEL(general_protection)] [ADDR:0x0] [PC:0x1289FAA1, pevm_icd_call_c
ommon()+225] [flags: 0x0, count: 1]
Dumping swap information
Memory (Avail / Total) = 197100.00M / 257386.76M
Swap (Avail / Total) = 16384.00M /  16384.00M
ssexhd: crashing the process...
Shadow_Core_Dump = partial
ksdbgcra: writing core file to directory '/u01/app/oracle/diag/rdbms/zjcdba/zjcdba2/cdump'

该问题与mos中这篇文章非常相似:
Automatic Report Flush = ORA-7445[opifcr] or ORA-7445[kkxmsiag] or ORA-7445[evaopn2] or ORA-7445[koxsh2copy] or ORA-7445[plsql_run] Etc. (文档 ID 2309250.1)


自动报告在12c引入,不过存在bug:

   Bug 23342649 - ORA-600[17099] IN M001 PROCESS DURING AUTOMATIC REPORT FLUSH TASK

   BUG 23342649 also occurs in Release 19c.

通过设置隐含参数禁用自动报告功能,恢复正常:
alter system set "_report_capture_cycle_time"=0;