Re: lifetime of the old CTID

Am 06.07.22 um 07:54 schrieb Andreas Kretschmer:
>
>
> Am 06.07.22 um 07:44 schrieb Christophe Pettus:
>>
>>> On Jul 5, 2022, at 22:35, Matthias Apitz <guru(at)unixarea(dot)de> wrote:
>>> Internally, in the DB layer, the read_where() builds the row list
>>> matching
>>> the WHERE clause as a SCROLLED CURSOR of
>>>
>>>     SELECT ctid, * FROM d01buch WHERE ...
>>>
>>> and each fetch() delivers the next row from this cursor. The functions
>>> start_transaction() and end_transaction() do what their names
>>> suggest and
>>> rewrite_actual_row() does a new SELECT based on the ctid of the
>>> actual row
>>>
>>>     SELECT * FROM d01buch WHERE ctid = ... FOR UPDATE
>>>     ...
>>>     UPDATE ...
>> On first glance, it appears that you are using the ctid as a primary
>> key for a row, and that's highly not-recommended.  The ctid is never
>> intended to be stable in the database, as you have discovered.  There
>> are really no particular guarantees about ctid values being retained.
>>
>> I'd suggest having a proper primary key column on the table, and
>> using that instead.
>
>
> 100% ACK.
>
> Andreas
>
>

it reminds me somehow on how people used he OID in old times - and now
we removed the OID completely.

Andreas

--
Andreas Kretschmer
Technical Account Manager (TAM)
www.enterprisedb.com