View Issue Details
ID | Project | Category | View Status | Date Submitted | Last Update |
---|---|---|---|---|---|
0005546 | SymmetricDS | New Feature | public | 2022-10-21 14:20 | 2022-12-16 15:20 |
Reporter | emiller | Assigned To | |||
Priority | normal | ||||
Status | closed | Resolution | no change required | ||
Product Version | 3.14.0 | ||||
Summary | 0005546: Add conversion tool for upgrading from H2 1.x to H2 2.x | ||||
Description | When upgrading to 3.15, any existing H2 databases will have to be upgraded to the latest 2.x version. It would be convenient if there was a conversion tool available to fix the existing H2 database URL(s) and migrate existing data to the new H2 database(s). | ||||
Tags | No tags attached. | ||||
|
This conversion tool will be necessary for any load only, extract only, or log mining nodes. |
|
Can we set MODE=LEGACY on the db.url for a smoother upgrade? |
|
The necessity of the conversion tool has become not necessary now since the new H2 version will continue to run in page mode if the database already exists |
|
Here is the list of functionalities supported by MODE=LEGACY: Empty IN predicate is allowed. TOP clause in SELECT is allowed. OFFSET/LIMIT clauses are allowed. MINUS can be used instead of EXCEPT. IDENTITY can be used as a data type. MS SQL Server-style IDENTITY clause is supported. Legacy SERIAL and BIGSERIAL data types are supported. AUTO_INCREMENT clause can be used instead of GENERATED { ALWAYS | BY DEFAULT } AS IDENTITY. If a value for identity column was specified in an INSERT command the base value of sequence generator of this column is updated if current value of generator was smaller (larger for generators with negative increment) than the inserted value. Identity columns have implicit DEFAULT ON NULL clause. It means a NULL value may be specified for this column in INSERT command and it will be treated as DEFAULT. Oracle-style CURRVAL and NEXTVAL can be used on sequences. TOP clause can be used in DELETE and UPDATE. Non-standard Oracle-style WHERE clause can be used in standard MERGE command. Attempt to reference a non-unique set of columns from a referential constraint will create an UNIQUE constraint on them automatically. Unsafe comparison operators between numeric and boolean values are allowed. IDENTITY() and SCOPE_IDENTITY() are supported, but both are implemented like SCOPE_IDENTITY() SYSDATE, SYSTIMESTAMP, and TODAY are supported. I don't see anything on there that would be required, but we could add the string to the URL or add the SET MODE LEGACY to the db.init.sql property in the engine file. |
|
Closing, we will set db.init.sql=Set SET MODE LEGACY in engine file instead |
Date Modified | Username | Field | Change |
---|---|---|---|
2022-10-21 14:20 | emiller | New Issue | |
2022-10-21 14:21 | emiller | Relationship added | related to 0005545 |
2022-11-09 20:35 | cquamme | Assigned To | => cquamme |
2022-11-09 20:35 | cquamme | Status | new => assigned |
2022-11-14 15:14 | cquamme | Assigned To | cquamme => |
2022-11-16 19:41 | cquamme | Status | assigned => acknowledged |
2022-12-13 13:23 | pmarzullo | Note Added: 0002242 | |
2022-12-13 13:25 | elong | Note Added: 0002243 | |
2022-12-13 13:50 | pmarzullo | Note Added: 0002244 | |
2022-12-13 13:53 | pmarzullo | Note Added: 0002245 | |
2022-12-16 15:19 | elong | Relationship added | related to 0005637 |
2022-12-16 15:20 | elong | Status | acknowledged => closed |
2022-12-16 15:20 | elong | Resolution | open => no change required |
2022-12-16 15:20 | elong | Note Added: 0002247 |