The Syniti Replicatelog file is an important source of information for details on any replication. Even though a replication process is typically automatic enough to determine choices based on the replication configuration, you sometimes need to examine the log file make adjustments to the replication configuration. For this reason is important to check the DBMoto.log file and understand the difference between:
Information messages | Useful information to follow the replication process as it progresses. |
Warning messages | A warning usually indicates an issue that does not typically generate an error in the replication process. However, you should check the source of the message and assess whether it could affect the results of your replication. |
Error messages | An error message generally indicates an error in the replication. In some cases, Syniti Replicate is able to resolve the issue automatically, but the information is still signaled as an error message because it is a source of possible asynchronism between source and target tables. In some cases, the error is an internal error (this is determined by the presence of a single level error message.) In other cases, the error is generated from the .NET, ODBC or OLE DB provider and the native provider error is shown, together with the Syniti Replicate error message. Sometimes errors appear simply because wrong assumptions are made during the replication setup. For instance, when replicating a table from an IBM Db2 for i source in mirroring mode, if the table is not journalized, an error will indicate the need to create a journal for that table. A duplicate primary key error could occur if inserting a record when the target table already contains that primary key value. |
These are some of the entries you may find in your log file.
Warning A replication conflict has occurred during replication (AddNew) of a record.
A synchronization conflict error. For example, the replicator is trying to insert a new record when the target table already contains that record (identified by the primary key value) but there are some differences with the values in the other columns. Verify why the conflict is occurring and if the current values are valid. Often previous errors in the replication can generate synchronization conflict errors.
Warning A replication conflict has occurred during replication (Delete) of a record.
A synchronization conflict error. For example, the replicator is trying to insert a new record when the target table already contains that record (identified by the primary key value) but there are some differences with the values in the other columns. Verify why the conflict is occurring and if the current values are valid. Often previous errors in the replication can generate synchronization conflict errors.
Replication Complete
You will only see a message in the log file to indicate that a replication has completed for refresh and continuous refresh. For mirroring and synchronization, there is no end to the replication and therefore no such message in the log file.
Date/Time Conversion Errors
These messages are shown when the Replicator reads a null date on the Db2 system(null date set to 01/01/01) because SQL Server doesn’t accept the year. The error is only a notification of the forced conversion and in this cases for every forced conversion an error is sent to the log; but replication is not stopped. However is possible manage date/time conversion errors from the Replication Agent Options dialog, available from the Metadata menu.
(-2147217900) ORA-01293: time or SCN range not fully contained in the listed logfiles ORA-06512: at "SYS.DBMS_LOGMNR", line 42 ORA-06512: at line 1
The range requested from the log is not present in the log files used. Because Syniti Replicate opens the Oracle log based on the last SCN (transaction ID) processed during the last mirroring step (which is the number you see displayed in the "Redo Log" dialog in Management Center), this error might indicate that, at some point, after a certain number of transactions processed, the Oracle log files have reached the maximum size available and are getting over-written from the beginning (in circular way).
If this is the problem, increasing the size of the log files or creating more log files could at least delay the problem if not solve it. Of course all depends on the number of transactions that have to be analyzed, if it is too huge chances are that Syniti Replicate is not able to keep up with this increasing number.
'OracleEllipsePrd' connection transactions have been deleted. Database will be temporarily removed from replicated databases. (-2147217900) ORA-01293: time or SCN range not fully contained in the listed logfiles. ORA-06512: at "SYS.DBMS_LOGMNR", line 42.
Use the Help Center to research issues or contact customer support for help.
Related Topics
Replication Agent Options dialog