摘要:
本文通过一次Oracle Job任务异常案例诊断，分析其原因及解决过程，从内部揭示Oracle Job任务调度及内部计时机制。
问题及环境
接到研发人员报告，数据库定时任务未正常执行，导致某些操作失败。
开始介入处理该事故.
系统环境:
SunOS DB 5.8 Generic_108528-21 sun4u sparc SUNW,Ultra-4 Oracle9i Enterprise Edition Release 9.2.0.3.0 - Production
解决过程
首先介入检查数据库任务
` $ sqlplus "/ as sysdba"
SQL*Plus: Release 9.2.0.3.0 - Production on Wed Nov 17 20:23:53 2004
Copyright (c) 1982, 2002, Oracle Corporation. All rights reserved.
Connected to:
Oracle9i Enterprise Edition Release 9.2.0.3.0 - Production
With the Partitioning, OLAP and Oracle Data Mining options
JServer Release 9.2.0.3.0 - Production
SQL> select job,last_date,last_sec,next_date,next_sec,broken,failures from
dba_jobs;
JOB LAST_DATE LAST_SEC NEXT_DATE NEXT_SEC B FAILURES
INTERVAL
31 16-NOV-04 01:00:02 17-NOV-04 01:00:00 N 0
trunc(sysdate+1)+1/24
27 16-NOV-04 00:00:04 17-NOV-04 00:00:00 N 0
TRUNC(SYSDATE) + 1
35 16-NOV-04 01:00:02 17-NOV-04 01:00:00 N 0
trunc(sysdate+1)+1/24
29 16-NOV-04 00:00:04 17-NOV-04 00:00:00 N 0
TRUNC(SYSDATE) + 1
30 01-NOV-04 06:00:01 01-DEC-04 06:00:00 N 0
trunc(add_months(sysdate,1),’MM’)+6/24
65 16-NOV-04 04:00:03 17-NOV-04 04:00:00 N 0
trunc(sysdate+1)+4/24
46 16-NOV-04 02:14:27 17-NOV-04 02:14:27 N 0
sysdate+1
66 16-NOV-04 03:00:02 17-NOV-04 18:14:49 N 0
trunc(sysdate+1)+3/24
8 rows selected. `
发现JOB任务是都没有正常执行，最早一个应该在17-NOV-04 01:00:00执行。但是没有执行。
建立测试JOB
` create or replace PROCEDURE pining
IS
BEGIN
NULL;
END;
/
variable jobno number;
variable instno number;
begin
select instance_number into :instno from v$instance;
dbms_job.submit(:jobno, ’pining;’, trunc(sysdate+1/288,’MI’),
’trunc(SYSDATE+1/288,’’MI’’)’, TRUE, :instno);
end;
/ `
发现同样的，不执行。
但是通过dbms_job.run(
1<job>)执行没有任何问题。
2
3进行恢复尝试
4
5怀疑是CJQ0进程失效，首先设置JOB_QUEUE_PROCESSES为0，Oracle会杀掉CJQ0及相应job进程
6
7` SQL> ALTER SYSTEM SET JOB_QUEUE_PROCESSES = 0; `
8等2~3分钟，重新设置
9
10` SQL> ALTER SYSTEM SET JOB_QUEUE_PROCESSES = 5; `
11此时PMON会重起CJQ0进程
12
13` Thu Nov 18 11:59:50 2004
14
15ALTER SYSTEM SET job_queue_processes=0 SCOPE=MEMORY;
16Thu Nov 18 12:01:30 2004
17ALTER SYSTEM SET job_queue_processes=10 SCOPE=MEMORY;
18Thu Nov 18 12:01:30 2004
19Restarting dead background process CJQ0
20CJQ0 started with pid=8
21但是Job仍然不执行，而且在再次修改的时候，CJQ0直接死掉了。
22Thu Nov 18 13:52:05 2004
23ALTER SYSTEM SET job_queue_processes=0 SCOPE=MEMORY;
24Thu Nov 18 14:09:30 2004
25ALTER SYSTEM SET job_queue_processes=10 SCOPE=MEMORY;
26Thu Nov 18 14:10:27 2004
27ALTER SYSTEM SET job_queue_processes=0 SCOPE=MEMORY;
28Thu Nov 18 14:10:42 2004
29ALTER SYSTEM SET job_queue_processes=10 SCOPE=MEMORY;
30Thu Nov 18 14:31:07 2004
31ALTER SYSTEM SET job_queue_processes=0 SCOPE=MEMORY;
32Thu Nov 18 14:40:14 2004
33ALTER SYSTEM SET job_queue_processes=10 SCOPE=MEMORY;
34Thu Nov 18 14:40:28 2004
35ALTER SYSTEM SET job_queue_processes=0 SCOPE=MEMORY;
36Thu Nov 18 14:40:33 2004
37ALTER SYSTEM SET job_queue_processes=1 SCOPE=MEMORY;
38Thu Nov 18 14:40:40 2004
39ALTER SYSTEM SET job_queue_processes=10 SCOPE=MEMORY;
40Thu Nov 18 15:00:42 2004
41ALTER SYSTEM SET job_queue_processes=0 SCOPE=MEMORY;
42
43Thu Nov 18 15:01:36 2004
44ALTER SYSTEM SET job_queue_processes=15 SCOPE=MEMORY; `
45尝试重起数据库，这个必须在晚上进行：
46
47` PMON started with pid=2
48DBW0 started with pid=3
49LGWR started with pid=4
50CKPT started with pid=5
51SMON started with pid=6
52RECO started with pid=7
53CJQ0 started with pid=8
54QMN0 started with pid=9
55.... `
56CJQ0正常启动，但是Job仍然不执行。
57
58没办法了...
59
60继续研究...居然发现Oralce有这样一个bug
61
62` 1. Clear description of the problem encountered:
63slgcsf() / slgcs() on Solaris will stop incrementing after
64497 days 2 hrs 28 mins (approx) machine uptime.
65
662. Pertinent configuration information
67No special configuration other than long machine uptime. .
683. Indication of the frequency and predictability of the problem
69100% but only after 497 days.
704. Sequence of events leading to the problem
71If the gethrtime() OS call returns a value > 42949672950000000
72nanoseconds then slgcs() stays at 0xffffffff. This can
73cause some problems in parts of the code which rely on
74slgcs() to keep moving.
75eg: In kkjssrh() does "now = slgcs(&se)" and compares that
76to a previous timestamp. After 497 days uptime slgcs()
77keeps returning 0xffffffff so "now - kkjlsrt" will
78always return 0. .
795. Technical impact on the customer. Include persistent after effects.
80In this case DBMS JOBS stopped running after 497 days uptime.
81Other symptoms could occur in various places in the code. `
82好么，原来是计时器溢出了，一检查我的主机:
83
84` bash-2.03$ uptime
8510:00pm up 500 day(s), 14:57, 1 user, load average: 1.31, 1.09, 1.08
86bash-2.03$ date
87Fri Nov 19 22:00:14 CST 2004 `
88刚好到事发时是497天多一点.ft.
89
90安排重起主机系统..
91
92这个问题够郁闷的，NND，谁曾想Oracle这都成...
93
94Oracle最后声称:
95
96fix made it into 9.2.0.6 patchset
97
98在Solaris上的9206尚未发布...晕.
99
100好了，就当是个经历吧，如果有问题非常不可思议的话，那么大胆怀疑Oracle吧，是Bug，可能就是Bug。
101
102重起以后问题解决，状态如下:
103
104` $ sqlplus "/ as sysdba"
105SQL*Plus: Release 9.2.0.3.0 - Production on Fri Nov 26 09:21:21 2004
106Copyright (c) 1982, 2002, Oracle Corporation. All rights reserved.
107Connected to:
108Oracle9i Enterprise Edition Release 9.2.0.3.0 - Production
109With the Partitioning, OLAP and Oracle Data Mining options
110JServer Release 9.2.0.3.0 - Production
111SQL> select job,last_date,last_sec,next_date,next_sec from user_jobs;
112
113JOB LAST_DATE LAST_SEC NEXT_DATE NEXT_SEC
114---------- --------- ---------------- --------- ----------------
11570 26-NOV-04 09:21:04 26-NOV-04 09:26:00
116SQL> /
117JOB LAST_DATE LAST_SEC NEXT_DATE NEXT_SEC
118---------- --------- ---------------- --------- ----------------
11970 26-NOV-04 09:26:01 26-NOV-04 09:31:00
120SQL>
121SQL> select * from v$timer;
122HSECS
123----------
1243388153
125SQL> select * from v$timer;
126HSECS
127----------
1283388319
129SQL>
130FAQ `</job>