Re: Progress report of CREATE INDEX for nested partitioned tables - Mailing list pgsql-hackers

From Ilya Gladyshev
Subject Re: Progress report of CREATE INDEX for nested partitioned tables
Date
Msg-id CBFBF527-1564-4E09-96D5-A389A2276AEF@gmail.com
Whole thread Raw
In response to Re: Progress report of CREATE INDEX for nested partitioned tables  (Tomas Vondra <tomas.vondra@enterprisedb.com>)
Responses Re: Progress report of CREATE INDEX for nested partitioned tables
List pgsql-hackers
> 17 янв. 2023 г., в 23:44, Tomas Vondra <tomas.vondra@enterprisedb.com> написал(а):
> Do we actually need the new parts_done field? I mean, we already do
> track the value - at PROGRESS_CREATEIDX_PARTITIONS_DONE index in the
> st_progress_param array. Can't we simply read it from there? Then we
> would not have ABI issues with the new field added to IndexStmt.

I think it’s a good approach and it could be useful outside of scope of this patch too. So I have attached a patch,
thatintroduces pgstat_progress_incr_param function for this purpose. There’s one thing I am not sure about, IIUC, we
canassume that the only process that can write into MyBEEntry of the current backend is the current backend itself,
thereforelooping to get consistent reads from this array is not required. Please correct me, if I am wrong here. 


Attachment

pgsql-hackers by date:

Previous
From: Peter Eisentraut
Date:
Subject: Re: Underscores in numeric literals
Next
From: Scott Mead
Date:
Subject: Re: [BUG] Autovacuum not dynamically decreasing cost_limit and cost_delay