Re: GiST limits on contrib/cube with dimension > 100? - Mailing list pgsql-hackers

From Tom Lane
Subject Re: GiST limits on contrib/cube with dimension > 100?
Date
Msg-id 2641.1560365369@sss.pgh.pa.us
Whole thread Raw
In response to Re: GiST limits on contrib/cube with dimension > 100?  (Andrey Borodin <x4mmm@yandex-team.ru>)
List pgsql-hackers
Andrey Borodin <x4mmm@yandex-team.ru> writes:
>> 9 июня 2019 г., в 23:05, Siarhei Siniak <siarheisiniak@yahoo.com> написал(а):
>> I've been using cube extension recompiled with
>> #define MAX_DIM 256.
>> But with a version 11.3 I'm getting the following error:
>> failed to add item to index page in <index_name>

> So, you have changed source code (removing dim constraint) and get cryptic error after that. In some way this is
expected...

Yeah.  There is necessarily a limit on the size of index entries,
and it's getting exceeded.

> Nevertheless, we can improve AddTuple messages. But there is not such strict guidelines as with B-tree. Probably,
tuplesshould not be bigger than half of usable page space. 

I do not think "improve AddTuple messages" is going to be enough to fix
this.  Daniel was correct that this is the same problem seen in


https://www.postgresql.org/message-id/flat/AM6PR06MB57318C9882C021879DD4101EA3100%40AM6PR06MB5731.eurprd06.prod.outlook.com

See my reply there.  I think we should continue this discussion on that
thread, since it has seniority.

            regards, tom lane



pgsql-hackers by date:

Previous
From: Bruce Momjian
Date:
Subject: Re: [Proposal] Table-level Transparent Data Encryption (TDE) and KeyManagement Service (KMS)
Next
From: Tom Lane
Date:
Subject: Re: Race conditions with checkpointer and shutdown