Re: Reindex doesn’t not working replica nodes

Hi Mohammed

If we are talking physical replication then the wall includes changes to
index’s data. If we are talking logical replication the wall does not
include index changes.

Med venlig hilsen

Peter Gram
Sæbyholmsvej 18
2500 Valby

Mobile: (+45) 5374 7107
Email: peter(dot)m(dot)gram(at)gmail(dot)com

On Fri, 16 Aug 2024 at 13.20, khan Affan <bawag773(at)gmail(dot)com> wrote:

> Hi Mohammed,
>
> PostgreSQL replication is based on replicating WAL (Write-Ahead Log)
> records, which capture changes to data files such as inserts, updates,
> deletes, and the creation of new tables. However,
>
> WAL does not include the physical structure of indexes.
>
> If you're encountering issues like zero rows being fetched from the
> replica, it could be due to index inconsistencies or replication lag. To
> ensure your read replica has updated indexes, you can promote the replica
> to be the primary node, perform the REINDEX operation, and then revert the
> roles, making the original primary node the primary again.
>
> Regards
> Muhammad Affan
>
> On Wed, Aug 14, 2024 at 5:58 PM Mohammed Afsar <vmdapsar(at)gmail(dot)com> wrote:
>
>> Dear experts,
>>
>>
>> We have initiated reindex on source db but it not replicated to replica
>> db and we have having select query fetching with zero rows but data exiting
>> on the table.
>> Postgres 11.10 streaming replication
>>
>> Regards,
>> Mohammed Afsar
>>
>