Content-Type: multipart/alternative;
 boundary="------------BMr5CRBA0z939SKJl0c7KbKO"
Message-ID: <a97d7401-e7c9-f771-6a00-037379f0a8bb@gmail.com>
Date: Thu, 19 Sep 2024 12:00:00 +0300
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.4.2
Subject: Re: Large expressions in indexes can't be stored (non-TOASTable)
Content-Language: en-US
To: Nathan Bossart <nathandbossart@gmail.com>, Tom Lane <tgl@sss.pgh.pa.us>
Cc: "Jonathan S. Katz" <jkatz@postgresql.org>,
 PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
References: <b611015f-b423-458c-aa2d-be0e655cc1b4@postgresql.org>
 <ZtipsVMZAGjrO1p_@nathan> <1679756.1725476893@sss.pgh.pa.us>
 <c500a379-7a20-4ab4-bcc9-5ea9aabe017a@postgresql.org>
 <ZuroEfWjxd4Zj6Yq@nathan> <1812111.1726671296@sss.pgh.pa.us>
 <ZusvlZBrPKm913LA@nathan>
From: Alexander Lakhin <exclusion@gmail.com>
In-Reply-To: <ZusvlZBrPKm913LA@nathan>
Archived-At: 
 <https://www.postgresql.org/message-id/a97d7401-e7c9-f771-6a00-037379f0a8bb%40gmail.com>
Precedence: bulk

This is a multi-part message in MIME format.
--------------BMr5CRBA0z939SKJl0c7KbKO
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit

Hello Nathan,

18.09.2024 22:52, Nathan Bossart wrote:
> Committed.  I waffled on whether to add a test for system indexes that used
> pg_index's varlena columns, but I ended up leaving it out.  I've attached
> it here in case anyone thinks we should add it.

I've discovered that Jonathan's initial script:
CREATE TABLE def (id int);
SELECT array_agg(n) b FROM generate_series(1,10_000) n \gset
CREATE OR REPLACE FUNCTION vec_quantizer (a int, b int[]) RETURNS bool
AS $$ SELECT true $$ LANGUAGE SQL IMMUTABLE;
CREATE INDEX ON def (vec_quantizer(id, :'b'));

completed with:
DROP INDEX CONCURRENTLY def_vec_quantizer_idx;

triggers an assertion failure:
TRAP: failed Assert("HaveRegisteredOrActiveSnapshot()"), File: "toast_internals.c", Line: 668, PID: 3723372

with the following stack trace:
ExceptionalCondition at assert.c:52:13
init_toast_snapshot at toast_internals.c:670:2
toast_delete_datum at toast_internals.c:429:60
toast_tuple_cleanup at toast_helper.c:303:30
heap_toast_insert_or_update at heaptoast.c:335:9
heap_update at heapam.c:3752:14
simple_heap_update at heapam.c:4210:11
CatalogTupleUpdate at indexing.c:324:2
index_set_state_flags at index.c:3522:2
index_concurrently_set_dead at index.c:1848:2
index_drop at index.c:2286:3
doDeletion at dependency.c:1362:5
deleteOneObject at dependency.c:1279:12
deleteObjectsInList at dependency.c:229:3
performMultipleDeletions at dependency.c:393:2
RemoveRelations at tablecmds.c:1594:2
ExecDropStmt at utility.c:2008:4
...

This class of assert failures is not new, see e. g., bugs #13809, #18127,
but this concrete instance (with index_set_state_flags()) emerged with
b52c4fc3c and may be worth fixing while on it...

Best regards,
Alexander
--------------BMr5CRBA0z939SKJl0c7KbKO
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: 7bit

<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
  </head>
  <body>
    <div class="moz-cite-prefix">Hello Nathan,<br>
      <br>
      18.09.2024 22:52, Nathan Bossart wrote:<br>
    </div>
    <blockquote type="cite" cite="mid:ZusvlZBrPKm913LA@nathan">
      <pre class="moz-quote-pre" wrap="">Committed.  I waffled on whether to add a test for system indexes that used
pg_index's varlena columns, but I ended up leaving it out.  I've attached
it here in case anyone thinks we should add it.
</pre>
    </blockquote>
    <br>
    I've discovered that Jonathan's initial script:<br>
    CREATE TABLE def (id int);<br>
    SELECT array_agg(n) b FROM generate_series(1,10_000) n \gset<br>
    CREATE OR REPLACE FUNCTION vec_quantizer (a int, b int[]) RETURNS
    bool <br>
    AS $$ SELECT true $$ LANGUAGE SQL IMMUTABLE;<br>
    CREATE INDEX ON def (vec_quantizer(id, :'b'));<br>
    <br>
    completed with:<br>
    DROP INDEX CONCURRENTLY def_vec_quantizer_idx;<br>
    <br>
    triggers an assertion failure:<br>
    TRAP: failed Assert("HaveRegisteredOrActiveSnapshot()"), File:
    "toast_internals.c", Line: 668, PID: 3723372<br>
    <br>
    with the following stack trace:<br>
    ExceptionalCondition at assert.c:52:13<br>
    init_toast_snapshot at toast_internals.c:670:2<br>
    toast_delete_datum at toast_internals.c:429:60<br>
    toast_tuple_cleanup at toast_helper.c:303:30<br>
    heap_toast_insert_or_update at heaptoast.c:335:9<br>
    heap_update at heapam.c:3752:14<br>
    simple_heap_update at heapam.c:4210:11<br>
    CatalogTupleUpdate at indexing.c:324:2<br>
    index_set_state_flags at index.c:3522:2<br>
    index_concurrently_set_dead at index.c:1848:2<br>
    index_drop at index.c:2286:3<br>
    doDeletion at dependency.c:1362:5<br>
    deleteOneObject at dependency.c:1279:12<br>
    deleteObjectsInList at dependency.c:229:3<br>
    performMultipleDeletions at dependency.c:393:2<br>
    RemoveRelations at tablecmds.c:1594:2<br>
    ExecDropStmt at utility.c:2008:4<br>
    ...<br>
    <br>
    This class of assert failures is not new, see e. g., bugs #13809,
    #18127,<br>
    but this concrete instance (with index_set_state_flags()) emerged
    with<br>
    b52c4fc3c and may be worth fixing while on it...<br>
    <br>
    Best regards,<br>
    Alexander<span style="font-family:monospace"><br>
    </span>
  </body>
</html>

--------------BMr5CRBA0z939SKJl0c7KbKO--