Resume Mode¶
Resume mode picks up a previously interrupted copy without re-creating schema objects.
When to Use¶
- A Full copy was interrupted (network drop, timeout, cancellation)
- Schema was already created but data copy didn't finish
- You don't want to start over from scratch
What Happens¶
- Skips DDL — does not recreate schemas, types, tables, functions, etc.
- Compares row counts — checks each table: source vs destination
- Copies only mismatched tables — tables with fewer rows on destination get truncated and re-copied
- Tables with matching counts are skipped entirely
How It Decides What to Copy¶
For each table in the source:
| Source Rows | Dest Rows | Action |
|---|---|---|
| 1000 | 1000 | ✅ Skip — already copied |
| 1000 | 500 | 🔄 Truncate + re-copy |
| 1000 | 0 | 🔄 Truncate + re-copy (destination table exists but is empty) |
Example Scenario¶
[10:30] Full copy started — 85 tables
[10:45] Network timeout after 60 tables
[10:46] User switches to Resume mode
[10:46] Resume: 60 tables match, 25 need copying
[10:55] Done — all 85 tables verified
Tips¶
Don't modify the destination between runs
Resume mode trusts that the schema is correct. If you manually alter the destination schema between a failed Full copy and a Resume, you may get constraint errors.
Requires the \"All Tables\" selection
Resume is blocked with an explanation while a non-default table selection is active — switch the source panel back to All Tables first.
- Resume is safe to run multiple times — it's idempotent
- If schema creation itself failed, use Full mode instead