| From: | Michael Paquier <michael(at)paquier(dot)xyz> |
|---|---|
| To: | Laurenz Albe <laurenz(dot)albe(at)cybertec(dot)at> |
| Cc: | Abhijit Gharami <abhijitgharami(at)gmail(dot)com>, PostgreSQL mailing lists <pgsql-general(at)postgresql(dot)org> |
| Subject: | Re: root page 3 of index "pg_class_oid_index" has level 0, expected 1 |
| Date: | 2019-11-14 07:54:57 |
| Message-ID: | 20191114075457.GF1910@paquier.xyz |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
On Thu, Nov 14, 2019 at 08:26:48AM +0100, Laurenz Albe wrote:
> If you have data corruption, you shouldn't use pg_upgrade to upgrade.
> pg_dumpall / psql is the way to go.
Please refer to this wiki page:
https://wiki.postgresql.org/wiki/Corruption
If you have a cluster in such a state, you have risks of having a
corruption which has spread, and you should not reuse this data folder
as-is. You can think of this REINDEX as a short-term solution to
retrieve as much of your data as you can so as you can dump it, and
then restore it into a safer location. For corruptions with system
indexes, you also have the option of using ignore_system_indexes.
--
Michael
| From | Date | Subject | |
|---|---|---|---|
| Next Message | stan | 2019-11-14 13:53:56 | Problems modifyiong view |
| Previous Message | Laurenz Albe | 2019-11-14 07:26:48 | Re: root page 3 of index "pg_class_oid_index" has level 0, expected 1 |