richard

87 积分	0 好友	3 主题

发消息

疑似Bug 9949948，请帮忙定位下

1^#

发表于 2012-3-22 15:42:30 | 查看: 10067| 回复: 7

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
20298 oracle 25 0 1689m 145m 137m R 99.8 0.6 1410:49 oracle

这个进程已经跑了很久了，所以对此做了 oradebug ,trace文件太大了，所以只截了一段
SQL>oradebug setospid 20298
SQL>oradebug dump processstate 10

/u01/app/oracle/admin/ecpdisk/bdump/orcl11_m000_20298.trc
Oracle Database 10g Enterprise Edition Release 10.2.0.5.0 - 64bit Production
With the Partitioning, Real Application Clusters, OLAP, Data Mining
and Real Application Testing options
ORACLE_HOME = /u01/app/oracle/product/10.2.0/db_1
System name: Linux
Node name: disk-db1
Release: 2.6.18-194.el5
Version: #1 SMP Tue Mar 16 21:52:39 EDT 2010
Machine: x86_64
Instance name: orcl11
Redo thread mounted by this instance: 1
Oracle process number: 194
Unix process pid: 20298, image: oracle@disk-db1 (m000)
*** ACTION NAME:(Auto-Flush Slave Action) 2012-03-21 16:00:55.474
*** MODULE NAME:(MMON_SLAVE) 2012-03-21 16:00:55.473
*** SERVICE NAME:(SYS$BACKGROUND) 2012-03-21 16:00:55.473
*** SESSION ID:(1063.50673) 2012-03-21 16:00:55.473
WARNING:io_submit failed due to kernel limitations MAXAIO for process=0 pending aio=0
WARNING:asynch I/O kernel limits is set at AIO-MAX-NR=65536 AIO-NR=65504
WARNING:1 Oracle process running out of OS kernelI/O resources aiolimit=0
ksfdgo()+1488<-ksfdaio1()+9848<-kfkUfsIO()+594<-kfkDoIO()+631<-kfkIOPriv()+616<-kfdIOPriv()+95<-kfioSubmitIO()+503<-kfioRequestPriv()+166<-kfioRequest()+689<-ksfd_osmgo()+1286<-ksfdgo()+1488<-ksfdaio1()+9848<-kcflbi()+498<-kcbldio()+1897<-kcblio()+944<-kcblnb()+790
<-kdblnb()+109<-kdblgb()+85<-kdblailb()+1093<-kdblai()+6407<-klclil1r()+910<-qerltRop()+928<-qerflRop()+95<-qerfxFetch()+978<-qerjotFetch()+209<-qerflFetch()+147<-rwsfcd()+88<-qerltFetch()+1008<-insdlexe()+372<-insExecStmtExecIniEngine()+85<-insexe()+867<-opiexe()+9334
<-kpoal8()+2295<-opiodr()+1184<-kpoodrc()+38<-rpiswu2()+409ssd_unwind_bp: unhandled instruction at 0x150d2cf instr=6a
ssd_unwind_bp: unhandled instruction at 0x150ba23 instr=68
<-kpoodr()+554<-upirtrc()+2101<-kpurcsc()+125<-kpuexecv8()+1705<-kpuexec()+2643<-OCIStmtExecute()+41<-kewrose_oci_stmt_exec()+62<-kewrgwxf1_gwrsql_exft_1()+284<-kewrgwxf_gwrsql_exft()+451<-kewrews_execute_wr_sql()+52<-kewa_flusher()+2592<-kewa_flush_cb()+21
<-kewrft_flush_table()+319<-kewrftec_flush_table_ehdlcx()+805<-kewrfat_flush_all_tables()+537<-kewrfos_flush_onesnap()+170<-kewrfsc_flush_snapshot_c()+650<-kewrafs_auto_flush_slave()+827<-kebm_slave_main()+221<-ksvrdp()+1159<-opirip()+748<-opidrv()+583<-sou2o()+114
<-opimai_real()+317<-main()+116<-__libc_start_main()+244<-_start()+41
*** 2012-03-21 19:56:27.678
Received ORADEBUG command 'unlimit' from process Unix process pid: 30254, *** 2012-03-21 19:56:55.123
Received ORADEBUG command 'dump systemstate 10' from process Unix process pid: 30254, image:

复制代码

有人说是Bug 9949948http://blog.chinaunix.net/uid-7628732-id-2980799.html

请帮忙看下哦

未命名.jpg (161.92 KB, 下载次数: 418)

未命名.jpg

分享0

收藏0 回复只看该作者道具举报

Maclean Liu(刘相兵

2^#

发表于 2012-3-22 16:06:11

WARNING:asynch I/O kernel limits is set at AIO-MAX-NR=65536 AIO-NR=65504
WARNING:1 Oracle process running out of OS kernelI/O resources aiolimit=0

AIO-MAX-NR=65536 << 推荐值

# Oracle-Validated setting for fs.aio-max-nr is 3145728
fs.aio-max-nr = 3145728

回复只看该作者道具举报

Maclean Liu(刘相兵

3^#

发表于 2012-3-22 16:07:02

ODM FINDING:

Bug 9949948 - Linux: Process spin under ksfdrwat0 if OS Async IO not configured high enough

Bug 9949948  Linux: Process spin under ksfdrwat0 if OS Async IO not configured high enough
This note gives a brief overview of bug 9949948.
The content was last updated on: 28-OCT-2011
Click here for details of each of the sections below.
Affects:

Product (Component) Oracle Server (Rdbms)
Range of versions believed to be affected Versions >= 10.2.0.4 but BELOW 11.1
Versions confirmed as being affected

      10.2.0.5

Platforms affected

      Linux X86-64bit
      Linux 32bit

   It is believed to be a regression in default behaviour thus:
   Regression introduced in 10.2.0.5

Fixed:

This issue is fixed in

      11.1.0.6 (Base Release)
      10.2.0.5.2 Patch Set Update
      10.2.0.5 Patch 5 on Windows Platforms

Symptoms:

Related To:

Hang (Process Spins)
Waits for "i/o slave wait"

DISK_ASYNCH_IO

Description

This problem is introduced in 10.2.0.5

It only affects platforms where Oracle has to reserve async IO slots,
such as Linux platforms.

If the OS async IO layer is underconfigured and an Oracle process
cannot get sufficient AIO slots then rather than reverting to
using non AIO call the process may go into an infinite spin
under ksfdrwat0.

Rediscovery notes:
   The spin will be preceded by messages in the trace
   file of the form:
      WARNING:io_submit failed due to kernel limitations MAXAIO
                  for process=0 pending aio=0
      WARNING:asynch I/O kernel limits is set at AIO-MAX-NR=65536 AIO-NR=65518
      WARNING:1 Oracle process running out of OS kernelI/O resources aiolimit=0

   Notice specifically that the value for aiolimit is reported as "0"
   for this bug.

   The process then spins in ksfdrwat0 typically with a stack showing
         skgfqio ()
         ksfdgo ()
         ksfdwtio ()
         ksfdwat1 ()
         ksfdrwat0 () <<< Spin point
         ksfdblock ()
         kcflwi ()
         kcflci ()
         kcblci ()
         kcblcio ()
         kcblgt ()
         kcbldrget ()

   It will show repeated waits for "i/o slave wait", which can be
   misleading as that is normally considered an idle wait event.

Workaround
   Raise the OS AIO limits such that the number of concurrent slot
   requirements never exceeds the OS limit.
   ie: Increase AIO-MAX-NR
   OR
   Disable async IO (Set DISK_ASYNCH_IO=FALSE)

回复只看该作者道具举报

Maclean Liu(刘相兵

4^#

发表于 2012-3-22 16:16:15

Advise:

1. 就STACK CALL 看和Bug 9949948差别较大。

更像另一个BUG NOTE:

Bug 8249407: WARNING:IO_SUBMIT FAILED DUE TO KERNEL LIMITATIONS

Hdr: 8249407 10.2.0.4 RDBMS 10.2.0.4 ASM PRODID-5 PORTID-226 8412426
Abstract: WARNING:IO_SUBMIT FAILED DUE TO KERNEL LIMITATIONS

PROBLEM:
--------
Trace files with the following messages are getting generated in the
asm/udump directory for all 3 nodes.

WARNING:io_submit failed due to kernel limitations MAXAIO for process=128
pending aio=118

WARNING:asynch I/O kernel limits is set at AIO-MAX-NR=3145728 AIO-NR=74963

WARNING:1 Oracle process running out of OS kernelI/O resources aiolimit=128

ksfdgo()+1791<-ksfdaio1()+4593<-kfkUfsIO()+587<-kfkDoIO()+631<-kfkIOPriv()+608
<-kfdIOPriv()+95<-kfdParallelIO()+1776<-kfdDiscoverDeep()+145<-kfgDiscoverDeep
()+310<-kfgDiscoverGroup()+616<-kfgTableCb()+2488<-kfgGrpTableCbInternal()+280
<-kfgGrpTableCb()+56<-qerfxFetch()+2944
<-opifch2()+3189<-kpoal8()+2994<-opiodr()+984<-ttcpip()+1012<-opitsk()+1322<-o
piino()+1026<-opiodr()+984<-opidrv()+547<-sou2o()+114<-opimai_real()+163<-main
()+116<-__libc_start_main()+219<-_start()+42
*** ERROR: PGA memory leak detected 79792 > 57768 ********
******************************************************

DIAGNOSTIC ANALYSIS:
--------------------
Issue looked similar to BUG 6908655,Bug 7439681 and Bug 7523755.
CT applied Patch for BUG 7523755.
But still the issue persists.

WORKAROUND:
-----------

RELATED BUGS:
-------------
BUG 6908655,Bug 7439681 and Bug 7523755

REPRODUCIBILITY:
----------------

TEST CASE:
----------

STACK TRACE:
------------

SUPPORTING INFORMATION:
-----------------------
Will upload ASM alert.log file of all the instances.
Trace file with the messages.
RDA

2. 建议你首先尝试修改 fs.aio-max-nr 内核参数，观察是否能绕过该问题。

回复只看该作者道具举报

richard

5^#

发表于 2012-3-23 20:26:39

1、忘记说环境了
RHEL5.5 X64 RAC 10.2.0.5.0

2、尝试修改aio-max
[root@disk-db1 ~]# sysctl -a|grep aio-max
fs.aio-max-nr = 3145728

select * from dba_hist_snapshot where 1=1 order by begin_interval_time desc

无标题.png (9.72 KB, 下载次数: 415)

无标题.png

回复只看该作者道具举报

richard

6^#

发表于 2012-3-23 20:27:09

但是没有见到效果：

$tail -f orcl11_mmon_19970.trc *** 2012-03-23 20:15:00.821
*** KEWRAFC: Flush slave failed, AWR Enqueue Timeout
*** 2012-03-23 20:16:00.874
*** KEWRAFC: Flush slave failed, AWR Enqueue Timeout
*** 2012-03-23 20:17:00.899
*** KEWRAFC: Flush slave failed, AWR Enqueue Timeout
*** 2012-03-23 20:18:00.959
*** KEWRAFC: Flush slave failed, AWR Enqueue Timeout

回复只看该作者道具举报