ogg源端投递进程告警OGG-01496、OGG-01668

denglj

问题描述：ogg源端投递进程告警OGG-01496、OGG-01668，如下所示：

源    端：oracle10.2.0.4 64位 + oel 5.11 64位

目 标 端：oracle19.16 64位 + centos 7.9 64位

1、异常重现

GGSCI(leo-10g-ogg) 4> info all

Program     Status     Group       Lag           Time Since Chkpt

MANAGER    RUNNING                                          

EXTRACT     ABENDED    DP1         00:00:00      00:15:46   

EXTRACT     RUNNING    EXTA        00:00:00      00:00:06   

2022-12-1115:43:03  INFO    OGG-01226 Socket buffer size set to 27985 (flush size 27985).

SourceContext :

  SourceModule            : [er.recovery]

  SourceID                :[/scratch/sganti/view_storage/sganti_core_lin64/oggcore/OpenSys/src/app/er/recovery.c]

  SourceFunction          : [RECOVERY_initialize]

  SourceLine              : [1712]

  ThreadBacktrace         : [7] elements

                          :[/u01/app/ogg/extract(CMessageContext::AddThreadContext()+0x26) [0x661466]]

                          :[/u01/app/ogg/extract(CMessageFactory::CreateMessage(CSourceContext*, unsignedint, ...)+0x7b2) [0x657f32]]

                          :[/u01/app/ogg/extract(_MSG_ERR_RECOVERY_OPEN_FAILED(CSourceContext*, charconst*, int, CMessageFactory::MessageDisposition)+0x9b) [0x64235b]]

                          :[/u01/app/ogg/extract(RECOVERY_initialize()+0x350) [0x50f7d0]]

                          :[/u01/app/ogg/extract(main+0x7dc) [0x5252bc]]

                          :[/lib64/libc.so.6(__libc_start_main+0xf4) [0x3c4701d9f4]]

                          :[/u01/app/ogg/extract(__gxx_personality_v0+0x1f2) [0x4f23ba]]

2022-12-1115:43:03  ERROR   OGG-01496 Failed to open target trail file ./dirdat/ra000002, at RBA 20660.

2022-12-1115:43:03  ERROR   OGG-01668  PROCESS ABENDING.

2、异常原因

目标端trail文件丢失，查看$OGG_HOME/dirdat下确实没有文件存在.

3、解决方案

--源端执行

GGSCI(leo-10g-ogg) 6> dblogin userid ogg@ORCL,password ogg

Successfullylogged into database.

GGSCI(leo-10g-ogg) 7> stop dp1

EXTRACTDP1 is already stopped.

GGSCI (leo-10g-ogg)8> alter extract dp1 etrollover

2022-12-1115:57:55  INFO    OGG-01520 Rollover performed.  For eachaffected output trail of Version 10 or higher format, after starting the sourceextract, issue ALTER EXTSEQNO for that trail's reader (either pump EXTRACT orREPLICAT) to move the reader's scan to the new trail file;  it will not happen automatically.

EXTRACTaltered.

说明：ETROLLOVER作用为前滚重新生成一个新的队列文件，注意只能在对应的EXTRACT停止后执行该命令.

--目标端执行

GGSCI(leo-19c-ogg) 8> dblogin userid ogg@SIMDB,password ogg

Successfullylogged into database.

GGSCI(leo-19c-ogg as ogg@simdb) 9> stop rep1

SendingSTOP request to REPLICAT REP1 ...

Requestprocessed.

--目标端修改复制进程trial文件

GGSCI(leo-19c-ogg as ogg@simdb) 10> alter replicat rep1 extseqno 000003, extrba 0

2022-12-1116:01:55  INFO    OGG-06594 Replicat REP1 has been altered. Even the start up position might beupdated, duplicate suppression remains active in next startup. To overrideduplicate suppression, start REP1 with NOFILTERDUPTRANSACTIONS option.

REPLICAT altered.

说明：“000003”根据报错信息“000002”加1所得.

GGSCI(leo-19c-ogg as ogg@simdb) 11> info all

Program     Status     Group       Lag at Chkpt  Time Since Chkpt

MANAGER   RUNNING                                          

REPLICAT    STOPPED    REP1        00:00:00      00:00:13  

--源端启动投递进程

GGSCI(leo-10g-ogg) 10> start dp1

SendingSTART request to MANAGER ...

EXTRACTDP1 starting

GGSCI(leo-10g-ogg) 11> info all

Program     Status     Group       Lag           Time Since Chkpt

MANAGER    RUNNING                                          

EXTRACT     RUNNING    DP1         00:00:00      00:04:30   

EXTRACT     RUNNING    EXTA        00:00:00      00:00:05

--目标端启动应用进程

GGSCI(leo-19c-ogg as ogg@simdb) 12> start rep1

SendingSTART request to MANAGER ...

REPLICATREP1 starting

GGSCI(leo-19c-ogg as ogg@simdb) 13> info all

Program     Status     Group       Lag at Chkpt  Time Since Chkpt

MANAGER   RUNNING                                          

REPLICAT    RUNNING    REP1        00:20:45      00:00:00   

[oracle@leo-19c-oggdirdat]$ pwd

/u01/app/ogg/dirdat

[oracle@leo-19c-oggdirdat]$ ll

total 4

-rw-r-----1 oracle oinstall 1195 Dec 11 16:02 ra000003

说明：目标端trail文件重新生成.