GoldenGate common error analysis (1)

1) One of the key points to solve the GoldenGate error is to use the error analysis tool, including the report file, the logdump tool of the ggserr. log discard file, and the GGSCI command line) to determine which component causes the error.

System or network?

Database error or application error?

Error in GoldenGate installation?

Is an error reported by a GoldenGate process?

What is the error in the parameter configuration file of GoldenGate?

SQL statement or stored procedure error?

Then determine the cause of the error and troubleshoot it one by one.

2) When GoldenGate encounters an error, you can use the log and report files to find the cause of the error and troubleshoot it step by step. The error message GoldenGate prompts a corresponding solution.

An error case is described as follows:

Run the following command:

Example 14:

GGSCI> view ggsevt

The error message 4 is displayed.

650) this. width = 650; "src =" http://www.bkjia.com/uploads/allimg/131229/2155034c5-0.jpg "title =" 1.jpg"/>

Figure 4

View report dpeyb shows similar information.

Then observe the error message of the Disaster Recovery replication process:

Example 15:

12:03:37 ERROR OGG-01028 Incompatible record in./dirdat/yb018262, rba 72955479 getting header ).

Log on to the trail file through logdump, as shown in Figure 5.

650) this. width = 650; "src =" http://www.bkjia.com/uploads/allimg/131229/2155035048-1.jpg "title =" 2.jpg"/>

Figure 5

Through analysis and scrutiny, it is confirmed that the shipping process is not recognized because one of the records in the trail file is damaged and cannot automatically roll to the next trail file, the replication process cannot automatically apply to the next trail file. The Pump process uses manual etrolover to manually specify the application to the next trail file through alter, and the fault can be eliminated.

1. aix ggsci cannot run

Error message:

Example 16:

Cannot load ICU resource bundle 'ggmessage', error code 2-No such file or directory

Cannot load ICU resource bundle 'ggmessage', error code 2-No such file or directory

IOT/Abort trap core dumped)

Or GGSCI can be started, but an error is reported when running any command.

Solution: The existing mount point is usually used to install GoldenGate, and the concurrent CIO parameter is used during mount. Create a new file system and mount it again as the GoldenGate installation directory.

Error message:

Example 17:

$./Ggsci

Exec): 050036 Cannot load program GGSCI because of the following errors:

050130 Symbol resolution failed for GGSCI because:

050136 Symbol _ GetCatName _ FiPCc number 158) is not exported from dependent module/usr/lib/libC. a [ansi_64.o].

050136 Symbol _ Getnumpunct _ FPCc number 162) is not exported from dependent module/usr/lib/libC. a [ansi_64.o].

050136 Symbol _ ct _ Q2_3std8_LocinfoFPCci number 183) is not exported from dependent module/usr/lib/libC. a [ansi_64.o].

050192 Examine. loader section symbols with the 'dump-TV 'command.

The reason is that the XLC version is 6.0, and the problem can be solved after the XLC version is upgraded to 10.1 or later.

2. HP-UX GGSCI cannot run

Error message: core dumped

This problem is only found on the HP-UX11.31.

Solution: The environment variable is not set correctly.

3. OGG-01296

Example 18:

ERROR OGG-01296 Oracle GoldenGate Delivery for Oracle, yx_rep3.prm: Error mapping from SGPM. A_PAY_FLOW to SGPM. A_PAY_FLOW.

This error occurs because the table structure is changed at the source end and no target end is notified.

Solution: Execute the corresponding statement on the target end and modify the table structure to be consistent with that on the source end.

4. OGG-01088

Error message:

Example 19:

ERROR OGG-01088 Oracle GoldenGate Delivery for Oracle, pms_rep1.prm: malloc 2097152 bytes failed.

ERROR OGG-01668 Oracle GoldenGate Delivery for Oracle, pms_rep1.prm: process abending.

Solution:

1) "ulimit-a", verify that the operating system has no limit on all resources of the user.

2) split the process into multiple processes.

3) download the latest patch package from support.oracle.com and upgrade GoldenGate.

5. OGG-01224

Example 20:

ERROR OGG-01224 Oracle GoldenGate Manager for Oracle, mgr. prm: No buffer space available

Solution:

Modify mgr. prm to expand the dynamic port range. dynamicportlist 7840-7914.

6. OGG-01031

Example 21:

The ERROR OGG-01031 There is a problem in network communication, a remote file problem, encryption keys for target and source do not match if using ENCRYPT) or an unknown error. reply specified Ed is Expected 4 bytes, but got 0 bytes, in trail. /dirdat/t000026, seqno 26, reading record trailer token at RBA 103637218 ).

11:04:16 ERROR OGG-01668 process abending.

Solution:

It may be that the trail file on the target end has a problem. Roll Back to generate a new send extract xxx rolover or "alter extract xxx rolover ".

7. OGG-01072

Example 22:

ERROR OGG-01072 LOBROW_get_next_chunkLOBROW_row_t *, BOOL, LOBROW_chunk_header_t *, char *, size_t, BOOL, *) Buffer overflow, needed: 132, alloc

Solution:

1) if the version is 11.1.1.0.1 Build 078, upgrade it to the latest patch package.

2) use "ulimit-a" to view resource usage restrictions and adjust the resource to unlimited.

3) Extract: dboptions lobbufsize.

4) replicat: dboptions lobwritesize 1 MB.

8. OGG-01476

Example 23:

ERROR OGG-01476 The previous run abended due to an out of order transaction. issue ALTER etrolover to advance the output trail sequence past the current trail sequence number, then restart. then, use alter extseqno on the subsequent pump EXTRACT, or REPLICAT, process group to start reading from the new trail file created by ALTER etrolover; the downstream process will not automatically switch to the new trail file.

During initialization, because the disaster recovery end is not ready, the production end performs many operations back and forth, resulting in confusion in the extraction of the production end. At this time, the extraction is restarted before RMAN is performed, ignore the chaotic information before calling.

Solution: "alter extract xxx, etrolover ".

9. OGG-00850

Example 24:

ERROR OGG-00850 Oracle GoldenGate Capture for DB2, extxa. prm: Database instance XP1 has both USEREXIT and LOGRETAIN set to off.

ERROR OGG-01668 Oracle GoldenGate Capture for DB2, extxa. prm: process abending.

Solution:

I) for DB2 8.1/8.2, you must set USEREXIT and LOGRETAIN to ON.

2) If it is DB2 5, LOGARCHMETH1 and LOGARCHMETH2 have been used to replace the preceding two parameters. Generally, LOGARCHMETH1 is DISK and LOGARCHMETH2 is TSM. The Archive mode is enabled using these two parameters. In DB2 5, USEREXIT can be set to OFF, but LOGRETAIN still needs to be set to ON.

Oracle video tutorial follow: http://u.youku.com/user_video/id_UMzAzMjkxMjE2.html