BUG #17863: Unable to restore dump 12.12 -> 15.2 - Mailing list pgsql-bugs

From PG Bug reporting form
Subject BUG #17863: Unable to restore dump 12.12 -> 15.2
Date
Msg-id 17863-3243d0500e135b20@postgresql.org
Whole thread Raw
Responses Re: BUG #17863: Unable to restore dump 12.12 -> 15.2  (Daniel Gustafsson <daniel@yesql.se>)
Re: BUG #17863: Unable to restore dump 12.12 -> 15.2  (Andres Freund <andres@anarazel.de>)
List pgsql-bugs
The following bug has been logged on the website:

Bug reference:      17863
Logged by:          Andrey Lizenko
Email address:      lizenko79@gmail.com
PostgreSQL version: 15.2
Operating system:   Ubuntu 20.04.5 LTS
Description:

Please close, if already reported or fixed. But its too many release notes,
to check everything and, on the other hand, I suppose, that dumps might be
compatible. 

12.12 
 pg_dump -Fc -Z 9 -v -f ./<db_name>_20230322.dmp <db_name>

15.2
pg_restore -U <user_name> -h 127.0.0.1 -v -d ./<db_name>_20230322.dmp

In short:
pg_restore: processing data for table "<db_name>.consensus_estimate"
pg_restore: error: could not uncompress data: (null)

No other errors in logs.
Here is the table:
<db_name>=# \d+ <db_name>.consensus_estimate
                                                               Table
"<db_name>.consensus_estimate"
       Column        |       Type       | Collation | Nullable |
           Default                         | Storage  | Stats target |
Description

---------------------+------------------+-----------+----------+--------------------------------------------------------+----------+--------------+-------------
 event_time          | bigint           |           |          |
                                           | plain    |              |
<field_name>          | text             |           | not null |
                                            | extended |              |
 source              | text             |           |          |
                                           | extended |              |
 disabled            | boolean          |           |          |
                                           | plain    |              |
 ext                 | jsonb            |           |          |
                                           | extended |              |
 sid                 | text             |           |          |
                                           | extended |              |
 period              | text             |           |          |
                                           | extended |              |
 currency_code       | text             |           |          |
                                           | extended |              |
 ymd                 | integer          |           |          |
                                           | plain    |              |
 reference_period    | text             |           |          |
                                           | extended |              |
 adjusted_eps_mean   | double precision |           |          |
                                           | plain    |              |
 adjusted_eps_median | double precision |           |          |
                                           | plain    |              |
 reported_eps_mean   | double precision |           |          |
                                           | plain    |              |
 reported_eps_median | double precision |           |          |
                                           | plain    |              |
 id                  | integer          |           | not null |
nextval('<db_name>.consensus_estimate_id_seq'::regclass) | plain    |
      |
 ext_hash            | uuid             |           |          |
                                           | plain    |              |
Indexes:
    "idx_consensus_estimate_unique" UNIQUE, btree (<field_name>, ymd,
ext_hash)
Access method: heap


pgsql-bugs by date:

Previous
From: David Rowley
Date:
Subject: Re: BUG #17862: Overall query cost ignores window function
Next
From: PG Bug reporting form
Date:
Subject: BUG #17864: pg_notify is publishing message twice