HP-ux oracle 10.2.0.4&…

from itpub 一个帖子，简要整理；
如题 oracle目录下产生大量的.trc .trm文件 （非常多非常多的），现在oralce目录已经满了100%，
-rw-rw---- 1 oracle dba 93 Oct 17 04:09 orcl_j001_9226.trm
-rw-rw---- 1 oracle dba 606 Oct 1 03:56 orcl_j001_9227.trc
-rw-rw---- 1 oracle dba 93 Oct 1 03:56 orcl_j001_9227.trm
-rw-rw---- 1 oracle dba 606 Oct 14 10:31 orcl_j001_9230.trc
-rw-rw---- 1 oracle dba 93 Oct 14 10:31 orcl_j001_9230.trm
-rw-rw---- 1 oracle dba 606 Oct 11 05:41 orcl_j001_9232.trc
-rw-rw---- 1 oracle dba 93 Oct 11 05:41 orcl_j001_9232.trm
-rw-rw---- 1 oracle dba 1212 Oct 12 03:37 orcl_j001_9233.trc
-rw-rw---- 1 oracle dba 143 Oct 12 03:37 orcl_j001_9233.trm
Fri Oct 26 09:58:16 2012
下面是一部分Alert 日志 （ORA-00020引起的吗） 不全太大了不能都传上来，麻烦大家给解答下谢谢啦！！！
ORA-00060: Deadlock detected. More info in file /oracle/app/oracle/diag/rdbms/orcl/orcl/trace/orcl_ora_1123.trc.
Fri Oct 26 09:59:18 2012
ORA-00060: Deadlock detected. More info in file /oracle/app/oracle/diag/rdbms/orcl/orcl/trace/orcl_ora_23081.trc.
Fri Oct 26 10:01:00 2012
ORA-00060: Deadlock detected. More info in file /oracle/app/oracle/diag/rdbms/orcl/orcl/trace/orcl_ora_8851.trc.
Fri Oct 26 10:04:44 2012
Thread 1 advanced to log sequence 6568 (LGWR switch)
Current log# 1 seq# 6568 mem# 0: /oradata/orcl/redo01.log
Fri Oct 26 10:07:22 2012
ORA-00060: Deadlock detected. More info in file /oracle/app/oracle/diag/rdbms/orcl/orcl/trace/orcl_ora_1110.trc.
Fri Oct 26 10:08:10 2012
ORA-00060: Deadlock detected. More info in file /oracle/app/oracle/diag/rdbms/orcl/orcl/trace/orcl_ora_8851.trc.
Fri Oct 26 10:11:00 2012
ORA-00060: Deadlock detected. More info in file /oracle/app/oracle/diag/rdbms/orcl/orcl/trace/orcl_ora_23106.trc.
Fri Oct 26 10:11:06 2012
ORA-00060: Deadlock detected. More info in file /oracle/app/oracle/diag/rdbms/orcl/orcl/trace/orcl_ora_19427.trc.
Fri Oct 26 10:11:06 2012
ORA-00060: Deadlock detected. More info in file /oracle/app/oracle/diag/rdbms/orcl/orcl/trace/orcl_ora_22849.trc.
Fri Oct 26 10:11:11 2012
ORA-00060: Deadlock detected. More info in file /oracle/app/oracle/diag/rdbms/orcl/orcl/trace/orcl_ora_19938.trc.
Fri Oct 26 10:11:12 2012
ORA-00060: Deadlock detected. More info in file /oracle/app/oracle/diag/rdbms/orcl/orcl/trace/orcl_ora_27474.trc.
Fri Oct 26 10:11:15 2012
ORA-00060: Deadlock detected. More info in file /oracle/app/oracle/diag/rdbms/orcl/orcl/trace/orcl_ora_19849.trc.
Fri Oct 26 10:11:45 2012
Thread 1 advanced to log sequence 6569 (LGWR switch)
Current log# 2 seq# 6569 mem# 0: /oradata/orcl/redo02.log
Fri Oct 26 10:16:16 2012
Thread 1 advanced to log sequence 6570 (LGWR switch)
Current log# 3 seq# 6570 mem# 0: /oradata/orcl/redo03.log
Fri Oct 26 10:17:58 2012
ORA-00020: maximum number of processes 150 exceeded
ORA-20 errors will not be written to the alert log for
the next minute. Please look at trace files to see all
the ORA-20 errors.
Fri Oct 26 10:17:58 2012
ORA-00020: maximum number of processes 150 exceeded
ORA-20 errors will not be written to the alert log for
the next minute. Please look at trace files to see all
the ORA-20 errors.
Fri Oct 26 10:17:58 2012
ORA-00020: maximum number of processes 150 exceeded
ORA-20 errors will not be written to the alert log for
the next minute. Please look at trace files to see all
the ORA-20 errors.
Fri Oct 26 10:18:13 2012
Thread 1 advanced to log sequence 6571 (LGWR switch)
Current log# 1 seq# 6571 mem# 0: /oradata/orcl/redo01.log
Fri Oct 26 10:19:58 2012
Fri Oct 26 10:19:58 2012
Fri Oct 26 10:19:58 2012
Fri Oct 26 10:19:58 2012
ORA-00020: maximum number of processes 150 exceeded
Fri Oct 26 10:19:58 2012
Fri Oct 26 10:19:58 2012
ORA-00020: maximum number of processes 150 exceededORA-00020: maximum number of processes 150 exceededORA-00020: maximum number of processes 150 exceeded
Fri Oct 26 10:19:58 2012
ORA-00020: maximum number of processes 150 exceeded
ORA-00020: maximum number of processes 150 exceededFri Oct 26 10:19:58 2012
ORA-00020: maximum number of processes 150 exceeded

Fri Oct 26 10:19:58 2012
ORA-20 errors will not be written to the alert log forORA-20 errors will not be written to the alert log forORA-20 errors will not be written to the alert log for
ORA-20 errors will not be written to the alert log for the next minute. Please look at trace files to see all

ORA-00020: maximum number of processes 150 exceeded
the next minute. Please look at trace files to see all the next minute. Please look at trace files to see all the next minute. Please look at trace files to see all
ORA-20 errors will not be written to the alert log for
the next minute. Please look at trace files to see all
the ORA-20 errors.
the ORA-20 errors.
the ORA-20 errors.
the ORA-20 errors.
the ORA-20 errors.
ORA-20 errors will not be written to the alert log for
the next minute. Please look at trace files to see all
ORA-20 errors will not be written to the alert log for
the ORA-20 errors.
the next minute. Please look at trace files to see all
the ORA-20 errors.
Fri Oct 26 10:19:58 2012
ORA-00020: maximum number of processes 150 exceeded
ORA-20 errors will not be written to the alert log for
the next minute. Please look at trace files to see all
the ORA-20 errors.
ORA-00020: maximum number of processes 150 exceeded
ORA-20 errors will not be written to the alert log for
the next minute. Please look at trace files to see all
the ORA-20 errors.
ORA-20 errors will not be written to the alert log for
the next minute. Please look at trace files to see all
the ORA-20 errors.
Fri Oct 26 10:19:58 2012
Process m000 submission failed with error = 20
Fri Oct 26 10:19:58 2012
Thread 1 cannot allocate new log, sequence 6572
Checkpoint not complete
。。。。。
省略

vi /oracle/app/oracle/diag/rdbms/orcl/orcl/trace/orcl_ora_5765.trc
"/oracle/app/oracle/diag/rdbms/orcl/orcl/trace/orcl_ora_5765.trc" 23222 lines, 1063331 characters
Trace file /oracle/app/oracle/diag/rdbms/orcl/orcl/trace/orcl_ora_5765.trc
Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 - 64bit Production
With the Partitioning, OLAP, Data Mining and Real Application Testing options
ORACLE_HOME = /oracle/app/oracle/product/11.2.0/dbhome_1
System name: HP-UX
Node name: hpuxdb
Release: B.11.31
Version: U
Machine: ia64
Instance name: orcl
Redo thread mounted by this instance: 1
Oracle process number: 0
Unix process pid: 5765, image: [url=mailto racle@hpuxdb]oracle@hpuxdb[/url]

*** 2012-10-26 10:15:37.741
Ioctl ASYNC_CONFIG error, errno = 1
Ioctl ASYNC_CONFIG error, errno = 1
Trace file /oracle/app/oracle/diag/rdbms/orcl/orcl/trace/orcl_ora_5765.trc
Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 - 64bit Production
With the Partitioning, OLAP, Data Mining and Real Application Testing options
ORACLE_HOME = /oracle/app/oracle/product/11.2.0/dbhome_1
System name: HP-UX
Node name: hpuxdb
Release: B.11.31
Version: U
Machine: ia64
Instance name: orcl
Redo thread mounted by this instance: 1
Oracle process number: 0
Unix process pid: 5765, image: [url=mailto racle@hpuxdb]oracle@hpuxdb[/url]

*** 2012-10-26 18:16:24.289
Ioctl ASYNC_CONFIG error, errno = 1
Ioctl ASYNC_CONFIG error, errno = 1
*** 2012-10-26 18:16:26.428
*** SESSION ID 33.15) 2012-10-26 18:16:26.428
"/oracle/app/oracle/diag/rdbms/orcl/orcl/trace/orcl_ora_5765.trc" 23222 lines, 1063331 characters

Ioctl ASYNC_CONFIG error, errno = 1

查了一下metalink资料
Applies to:

oracle Server - Enterprise Edition - Version: 9.2.0.1 to 11.2.0.2 - Release: 9.2 to 11.2
HP-UX PA-RISC (64-bit)
HP-UX Itanium
Goal

How to disable async_io on oracle in HP-UX to avoid this message in trace file:
Ioctl ASYNC_CONFIG error, errno = 1
Solution

If /dev/async is present and correctly configured, oracle (for HP only) will use it for memory locking which is needed for async i/o for SGA memory pages via ioctl.

oracle (for HP only) opens /dev/async successfully only if the /dev/async HP-UX device driver is properly configured for read and write.

This is irrespective of whether the DISK_ASYNCH_IO parameter is set to TRUE.

Problem is oracle (for HP only) still uses async I/O regardless of the value of init settings
(disk_asynch_io=FALSE and filesystemio_options=none).
The parameter filesystemio_options=none doesn't work.

From the 9.2 until 10.1 version, to inactivate ASYNCH_IO with oracle, the workaround is to:

chown bin:bin /dev/async
chmod 660 /dev/async
Note: This may impact other applications that wish to use asynch I/O.
It is recommended to shutdown oracle database before inactivating asynch I/O.

Grant dba group the MLOCK privilege to avoid the Ioctl ASYNC_CONFIG trace file errors:

(1) # /usr/sbin/setprivgrp dba MLOCK
(2) # vi /etc/privgroup
-- This should contain dba MLOCK RTSCHED RTPRIO
(3) # cat /etc/privgroup
dba MLOCK RTSCHED RTPRIO
Since the oracle 10.2 version, the problem is resolved as these settings stop the tracing:
disk_asynch_io=FALSE
filesystemio_options=none

查看当前系统参数配置：
SQL> show parameter disk_asynch_io ;

NAME                                                                TYPE                                                        VALUE
------------------------------------ -------------------------------- ------------------------------
disk_asynch_io                                            boolean                                                  TRUE
SQL> show parameter filesystemio_options;

NAME                                                                TYPE                                                        VALUE
------------------------------------ -------------------------------- ------------------------------
filesystemio_options                                string                                                    asynch


L------------------------------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------------------------------