The response to a conflict can be configured before the conflict occurs. This is done with conflict handlers. The configuration can be specified for each replicated table and for each operation type separately, and for the special case ("data" divergence) when zero or more than one row are affected. The Dbvisit Replicate command to set conflict handlers is SET_CONFLICT_HANDLERS.
...
| Section |
|---|
| Column |
|---|
| | Panel |
|---|
| dbvrep> SET_CONFLICT_HANDLERS FOR TABLE SCOTT.AVI_OBJECTS FOR UPDATE ON NO_DATA TO OVERWRITE |
|
|
| Note |
|---|
Source tablename is used while setting conflict handlers. |
To show what the conflict handler has been set to:
| Section |
|---|
| Column |
|---|
| | Panel |
|---|
| dbvrep> show_conflict_handlers for table scott.avi_objects The table called SCOTT.AVI_OBJECTS on source is handled on apply as follows: UPDATE (error): handler: RETRY logging: LOG UPDATE (no_data): handler: OVERWRITE logging: LOG UPDATE (too_many): handler: OVERWRITE logging: LOG DELETE (error): handler: RETRY logging: LOG DELETE (no_data): handler: RETRY logging: LOG DELETE (too_many): handler: RETRY logging: LOG INSERT (error): handler: RETRY logging: LOG TRANSACTION (error): handler: RETRY logging: LOG |
|
|
| Info |
|---|
Note that previous versions used the DATA handler instead of NO_DATA and TOO_MANY. For compatibility reasons, this is still supported - specifying DATA handler sets both NO_DATA and TOO_MANY handlers. For 2.3.18 and before: Note that the handler is executed AFTER the conflicting statement was run and thus any changes done by the statement are still in effect. This is important to consider especially for TOO_MANY handler. For 2.3.20 and later: For TOO_MANY error, the conflicting statement is rolled back before retry etc. This brings a new conflict resolution: FORCE. While FORCE keeps the effect of the conflicted statement (multiple rows changed), IGNORE/DISCARD rolls the statament back and thus no change is applied. |
Example
The following example shows an example of a duplicate row conflict handler.
Create a simple table with two duplicate rows:
| Section |
|---|
| Column |
|---|
| | Panel |
|---|
| create table scott.x (i number); insert into x values (1); insert into x values (1); commit; |
|
|
On the source database, the user issues the following update:
| Section |
|---|
| Column |
|---|
| | Panel |
|---|
| update x set i=2 where i=1; commit; |
|
|
This update will affect 2 rows in the table.
The Mine process will see this as two row changes, so the Apply process will execute:
| Section |
|---|
| Column |
|---|
| | Panel |
|---|
| (#1) update x set i=2 where i=1;
(#2) update x set i=2 where i=1; |
|
|
The first statement conflicts with "2 rows updated".
There are several ways to handle this conflict:
- FORCE: both rows are changed. Apply now executes #2 - and conflicts again, but this time with "0 rows updated" as both records now i=2. This conflict has to be handled as well with a NO_DATA conflict handler. This second conflict has to be ignored because the second update is no longer needed.
- SQL (rownum=1): only one row is changed. Apply now executes #2 - only 1 row is updated, so no conflict.
Option 2) is the best approach - it handles only duplicates, not other conflicts.
Option 1) would (silently) handle other types of conflicts.
The syntax for Option 2 is:
UPDATE:
| Section |
|---|
| Column |
|---|
| | Panel |
|---|
| SET_CONFLICT_HANDLERS FOR TABLE SCOTT.X FOR UPDATE ON TOO_MANY TO SQL s/$/ and rownum = 1/ |
|
|
DELETE
| Section |
|---|
| Column |
|---|
| | Panel |
|---|
| SET_CONFLICT_HANDLERS FOR TABLE SCOTT.X FOR DELETE ON TOO_MANY TO SQL s/$/ and rownum = 1/ |
|
|
| Info |
|---|
Current conflicts will not be affected by newly defined CONFLICT HANDLERS. To resolve current conflicts there are 2 options: - Manually resolve with RESOLVE CONFLICT, OR
- Restart the APPLY process. This will enable the newly defined conflict handler to manage the current conflict
|