Re: Patch: dumping tables data in multiple chunks in pg_dump

Hi Zsolt,

I had finally time to return to this

Please find the latest patch attached

This commit fixes two critical issues :

1. Fixes dependency tracking during parallel restore. The original patch
modified dependencies (adding chunks) after depCount was initialized,
leading to early scheduling of dependent objects (constraints/indexes)
before all data chunks were loaded. This is fixed by moving the
depCount initialization loop to the end of fix_dependencies().

2. Fixes inserts mode. The previous patch missed adding ctid range clauses
to dumpTableData_insert(), causing the entire table to be dumped for
each chunk when using --inserts. This is fixed by adding the segment
ctid filtering logic to dumpTableData_insert() as well.
Also added TAP tests to verify chunked dump/restore in insert mode.

On Mon, Mar 30, 2026 at 11:32 PM Zsolt Parragi
<zsolt(dot)parragi(at)percona(dot)com> wrote:
>
> Hello!
>
> A simple test causes an assertion failure in my testing, dependency
> counting still doesn't seem to work correctly:
>
> pg_restore: >...>/pg_backup_archiver.c:5207: reduce_dependencies:
> Assertion `otherte->depCount > 0' failed.
>
> Without assertions it results in data loss.
>
> 004_pg_dump_parallel also showcases the issue in my testing.
>
> But simple manual testing also confirms it:
>
> 1. create some data
>
> CREATE TABLE tplain (id int UNIQUE);
> INSERT INTO tplain SELECT x FROM generate_series(1,1000) x;
>
> 2. create a dump
>
> dump with --max-table-segment-pages=2
>
> 3. try to restore
>
> restore with --jobs=3