Subject: Re: BUG #17372: Altering statistics during concurrent drop can lead
 to a server crash
To: Tomas Vondra <tomas.vondra@enterprisedb.com>,
 pgsql-bugs@lists.postgresql.org
References: <17372-bf3b6e947e35ae77@postgresql.org>
 <2615f158-b790-4176-11ad-9711e4fd56c7@enterprisedb.com>
From: Alexander Lakhin <exclusion@gmail.com>
Message-ID: <a652653f-b669-cb3a-6116-663e335a5012@gmail.com>
Date: Fri, 21 Jan 2022 09:00:00 +0300
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101
 Thunderbird/78.14.0
MIME-Version: 1.0
In-Reply-To: <2615f158-b790-4176-11ad-9711e4fd56c7@enterprisedb.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Content-Language: en-US
Archived-At: 
 <https://www.postgresql.org/message-id/a652653f-b669-cb3a-6116-663e335a5012%40gmail.com>
Precedence: bulk

Hello Tomas,
20.01.2022 20:57, Tomas Vondra wrote:
> Thanks for the report - I've reproduced it using your script (I had to
> add a short wait before the syscache lookup). The issue is fairly
> simple, we never check we actually found a tuple in the syscache, i.e.
> there's no HeapTupleIsValid() call. We check the OID earlier, but the
> tuple may have disappeared since then.
>
> IMO the usual "cache lookup failed" error that we throw in other places
> in similar cases is good enough. We already fail with other errors
> (tuple concurrently updated/deleted) here.
Yes, the rough, but simple search:
grep -Porz
'.*=\s*SearchSysCache1.*\n(?!(.*|.*\n.*|.*\n.*\n.*|.*\n.*\n.*\n.*)HeapTupleIsValid)'
./
finds only this place.
Your fix works for me.

Thanks!

Best regards,
Alexander