Re: BDR, near xid wraparound, a lot of files in pg_subtrans directory

[milist ujang <ujang.milist@xxxxxxxxx>]

Hi craig,

I have a bunch of orphaned replication identifier.

How to cleanup those orpaned identifier without affecting currently configured slots?

I mean what is relation the replication identifier with slot_name in replication slots?

On Mon, Sep 18, 2017 at 3:00 PM, milist ujang <ujang.milist@xxxxxxxxx> wrote:

Hi Craig,

So, is it safe to drop those list from this query output?

select riname from pg_replication_identifier where riname not in 

(select external_id from pg_replication_identifier_progress);

I cannot read pg_get_replication_identifier_progress function, is it likely c function?

On Fri, Sep 15, 2017 at 11:14 AM, Craig Ringer <craig@xxxxxxxxxxxxxxx> wrote:

On 15 September 2017 at 11:46, milist ujang <ujang.milist@xxxxxxxxx> wrote:

Hi Craig,

Thanks again for pointing to inactive replication slot.

After inactive replication slot been dropped, the relfrozenxid now moving.

I wonder if  replication identifier will have some issue if left un-chained? since at other side there are inactive replication identifier.

No, that's harmless.

However, if there's still an "other side" at all, you presumably have broken replication. 

--

 Craig Ringer                   http://www.2ndQuadrant.com/
 PostgreSQL Development, 24x7 Support, Training & Services

--

regards

ujang jaenudin | DBA Consultant (Freelancer)
http://ora62.wordpress.com
http://id.linkedin.com/pub/ujang-jaenudin/12/64/bab

--

regards

ujang jaenudin | DBA Consultant (Freelancer)
http://ora62.wordpress.com
http://id.linkedin.com/pub/ujang-jaenudin/12/64/bab