Maclean Liu(刘相兵

alter system archive log current;
alter system archive log current;
/

执行5~6次再看看

Maclean Liu(刘相兵

2013-07-18 04:49:17  WARNING OGG-01423  Oracle GoldenGate Capture for Oracle, ext1.prm:  No valid default archive log destination directory found for thread 1.
2013-07-18 04:49:20  INFO    OGG-01515  Oracle GoldenGate Capture for Oracle, ext1.prm:  Positioning to begin time Jul 13, 2013 5:21:30 AM.
2013-07-18 04:49:20  ERROR   OGG-00685  Oracle GoldenGate Capture for Oracle, ext1.prm:  begin time Jul 13, 2013 5:21:30 AM prior to oldest log in log history. Last SQL executed <SELECT 1   FROM dual  WHERE TO_DATE(:begin_time,'YYYY-MM-DD HH24:MI:SS') <    (SELECT MIN(first_time)       FROM v$log_history      WHERE thread# = :ora_thread)>.


肯定是archivelog 配置的有问题 ， 重建 extract或者可以解决问题

Maclean Liu(刘相兵

1) Bug 13840207 & Bug 13628811

The bug occurs when there is a race condition in producer_status that is called from both the consumer and coordinator threads.
Restart of the extract would work fine in this case


The first one is Bug 13628811 which was backported in label OGGCORE_11.2.1.0.1_PLATFORMS_120229.1800
The second is Bug 13840207 which was backported in label OGGCORE_11.2.1.0.2_PLATFORMS_120516.1800
[This section is not visible to customers.]

2) Bug 14105014

The bug occurs due to issues reading the CSN information from output recovery file

renaming the <extract process name>.cps file from dirchk and restarting the extract may help
[This section is not visible to customers.]

If extract generated a core then gdb (dbx for AIX) output would show a error like below

$gdb extract <core file name>
...
Core was generated by `extract'.
Program terminated with signal 10, Bus error.

(gdb) where
#0 0x40000000006c3ce0:0 in parse_supp_token_buf () at
/scratch/aime1/adestore/views/aime1_staoi06/oggcore/OpenSys/src/gglib/ggapp/checkpt.c:2213
#1 0x40000000006ca060:0 in CP_read_checkpoint_csn () at
/scratch/aime1/adestore/views/aime1_staoi06/oggcore/OpenSys/src/gglib/ggapp/checkpt.c:2785
#2 0x40000000003b7850:0 in START_EXTRACT_CHECKPOINT ()
[This section is not visible to customers.]

3) Bug 13894704

Extract could core dump if minimum db level supplemental logging is not enabled

OGGCORE_11.2.1.0.1_PLATFORMS_120403.1800 contains changes from arsepulv_rfi_backport_13894704_11.2.1.0.1[This section is not visible to customers.]
4) Bug 13860160

This bug causes the Extract to core dump when capturing inserts from an IOT table generated by
doing a select from another table:

eg., insert into <iot table> select * from <table>;

5) Bug 14546106

Issue can also happen when the system has insufficient heap size. Here is an example showing the problem

Snippet from the extract report file

Operating System Version:
AIX
Version 6, Release 1
Node: kipe401
Machine: 00F634464C00
soft limit hard limit
Address Space Size : unlimited unlimited
Heap Size : 268435456 unlimited
File Size : unlimited unlimited
CPU Time : unlimited unlimited

Increasing the heap size resolved the issue in this case

customer used LDR_CNTRL=MAXDATA=0xC0000000 to do the increase[This section is not visible to customers.]
6) Bugs 14268459, 14371558 . Core shows bus error like below

Core was generated by `/u02/app/oracle/product/11.2.1.0.1/extract PARAMFILE /u02/app/oracle/product/11'.
Program terminated with signal 7, Bus error.
#0 0x000000314787c613 in ?? ()

7) Bug 14477847

Replicat core dumps and stack shows following

Program received signal SIGSEGV, Segmentation fault.
0x000000364be79ba0 in strlen () from /lib64/libc.so.6
(gdb) where
#0 0x000000364be79ba0 in strlen () from /lib64/libc.so.6
#1 0x000000364be798d6 in strdup () from /lib64/libc.so.6
#2 0x00002aaaaab86725 in CDataItem::CDataItem (this=0x7fffffff8f80, value=@0x7fffffff8f58)
at /scratch/mmar/view_storage/mmar_14066483/oggcore/OpenSys/src/gglib/gglog/CDataItem.cpp:109
#3 0x000000000051e3ec in take_rep_err_action (errtype=218, err=0, err_msg=0x0, extr_ptr=0x1e57270, hdr=0x14b7420,
rec_buf=0x12b4580 "", src_file=0x17f52b0, commit_error=false)
at /scratch/mmar/view_storage/mmar_14066483/oggcore/OpenSys/src/app/er/errors.cpp:376
#4 0x0000000000536a80 in process_extract_loop ()
at /scratch/mmar/view_storage/mmar_14066483/oggcore/OpenSys/src/app/er/processloop.cpp:1791
#5 0x0000000000548722 in main (argc=<value optimized out>, argv=<value optimized out>)
at /scratch/mmar/view_storage/mmar_14066483/oggcore/OpenSys/src/app/er/replicat/replicat.cpp:899
(gdb)

8) Replicat core dumps and stack from the core shows following

libclntsh.so.11.1`OCIStmtExecute+0x30(102b58148, 103a26a40, 102b73068, 3e8, 0, 0)
__1cTDBOCI_exec_stmt_ctx6FpnTOCISESS_context_def_pnODBOCI_stmt_def_pnUBATCHSQdDL_db_params_t__i_+0xbc(1009300e0, 1036f46b0,
102abf990, 0, 10092b000, 103e97fb0)
__1cVDBBATCHSQdDL_exec_batch6FpnOBATCHSQdDL_ctx_t_pnUBATCHSQdDL_db_params_t__i_+0x1b4(1018d7600, 102abf990, 101735440, 102abf990,
103a3c750, 1017353d0)
__1cRexec_batch_thread6Fpv_v_+0x1ac(1036fb150, 230a3c, 102abf990, ffffffff, 87a0000, 10090b550)
__1cRcheck_queue_batch6FpnOBATCHSQdDL_ctx_t_pnQBATCHSQdDL_batch_t_i_i_+0x1e8(1, 1, 0, 100a05, 100a05e78, ffffffff)
__1cTexec_queue_threaded6FpnOBATCHSQdDL_ctx_t_pnQBATCHSQdDL_queue_t__i_+0x294(1018d7600, 103449f10, 1017354b0, 10090b550, 0, 0)
__1cTREP_end_transaction6Fh_i_+0x83c(100911000, 100925540, 100756804, 100926b40, 0, 3e8)
__1cUprocess_extract_loop6F_h_+0x60fc(100800, 0, 0, 101463b68, 10074eb00, 100925540)
main+0xc8c(100756e37, 1009268f0, 1007572f4, 100800, 101752350, 100757205)
_start+0x17c(0, 0, 0, 0, 0, 0)

BATCHSQL can be disabled as a work-around. Fix is in bug 10379816