Re: Fwd: Regarding tables detach concurrently with run_maintenance_proc()

Hi Respected Team

By default proc() does not detach tables concurrently. How do we implement
tables detach concurrently without blocking running sessions in prod.
why this is very critical to implement for pg_partman.
if this is not available yet on 5.1.0 then when can i expect to get it
if already there then please let me know the implementation of detaching
tables concurrently
Any best way to implement the same please with out concurrently how do we
ensure data integrity and consistency

Regards
Durga Mahesh

Hi Respected Team

with pg_partman By default proc() does not detach tables concurrently. How
do we implement tables detach concurrently without blocking other sessions
Here queries not using date column to detach tables with
run_maintenance_proc() which is not using concurrently based on the
retention policy which leads to scan all available child tables hence need
to trigger this proc with concurrently option to avoid blocking other child
tables beyond rentention policy while running statements on them

Do we have any other alternative rather than using pg_partman()?

your response is valuable

Regards,
Durga Mahesh

Re: Durgamahesh Manne
> with pg_partman By default proc() does not detach tables concurrently. How
> do we implement tables detach concurrently without blocking other sessions
> Here queries not using date column to detach tables with
> run_maintenance_proc() which is not using concurrently based on the
> retention policy which leads to scan all available child tables hence need
> to trigger this proc with concurrently option to avoid blocking other child
> tables beyond rentention policy while running statements on them

You might have more success by filing pg_partman issues at
https://github.com/pgpartman/pg_partman/issues

> Do we have any other alternative rather than using pg_partman()?

Well you can just run the same commands manually that pg_partman would
run.

Christoph

On Fri, Jul 19, 2024 at 7:55 PM Christoph Berg <myon(at)debian(dot)org> wrote:

> Re: Durgamahesh Manne
> > with pg_partman By default proc() does not detach tables concurrently.
> How
> > do we implement tables detach concurrently without blocking other
> sessions
> > Here queries not using date column to detach tables with
> > run_maintenance_proc() which is not using concurrently based on the
> > retention policy which leads to scan all available child tables hence
> need
> > to trigger this proc with concurrently option to avoid blocking other
> child
> > tables beyond rentention policy while running statements on them
>
> You might have more success by filing pg_partman issues at
> https://github.com/pgpartman/pg_partman/issues
>
> > Do we have any other alternative rather than using pg_partman()?
>
> Well you can just run the same commands manually that pg_partman would
> run.
>
> Christoph
>

Hi
You might have more success by filing pg_partman issues at
https://github.com/pgpartman/pg_partman/issues >>> okay
My intention is to have any other extension other than pg_partman to
manage table partitions manually

Regards
Durga Mahesh

Hi ,

You can consider the pg_pathman extension.

*Muhammad Imtiaz*

*PostgreSQL Technical Support Lead *
*/ Pakistan R&D*
*Mobile: +923345072521*
*Email: imtiaz(dot)m(at)bitnine(dot)net <imtiaz(dot)m(at)bitnine(dot)net>*

On Fri, Jul 19, 2024 at 7:55 PM Durgamahesh Manne <maheshpostgres9(at)gmail(dot)com>
wrote:

>
>
> On Fri, Jul 19, 2024 at 7:55 PM Christoph Berg <myon(at)debian(dot)org> wrote:
>
>> Re: Durgamahesh Manne
>> > with pg_partman By default proc() does not detach tables concurrently.
>> How
>> > do we implement tables detach concurrently without blocking other
>> sessions
>> > Here queries not using date column to detach tables with
>> > run_maintenance_proc() which is not using concurrently based on the
>> > retention policy which leads to scan all available child tables hence
>> need
>> > to trigger this proc with concurrently option to avoid blocking other
>> child
>> > tables beyond rentention policy while running statements on them
>>
>> You might have more success by filing pg_partman issues at
>> https://github.com/pgpartman/pg_partman/issues
>>
>> > Do we have any other alternative rather than using pg_partman()?
>>
>> Well you can just run the same commands manually that pg_partman would
>> run.
>>
>> Christoph
>>
>
> Hi
> You might have more success by filing pg_partman issues at
> https://github.com/pgpartman/pg_partman/issues >>> okay
> My intention is to have any other extension other than pg_partman to
> manage table partitions manually
>
> Regards
> Durga Mahesh
>