Re: BUG #2013: Assertion Failure
| От | Alvaro Herrera |
|---|---|
| Тема | Re: BUG #2013: Assertion Failure |
| Дата | |
| Msg-id | 20051102141537.GI26524@surnet.cl обсуждение исходный текст |
| Ответ на | BUG #2013: Assertion Failure ("Joel Stevenson" <joelstevenson@mac.com>) |
| Список | pgsql-bugs |
Joel Stevenson wrote:
> This occured during a nightly mainentance run which sets the transaction
> isolation level to SERIALIZABLE, moves records from 3 highly active tables
> to 3 storage tables by creating a temp table and selecting candidate record
> ids into it, then using that list via sub-select for the move ala "insert
> into blah_storage (f1, f2, f3) select f1, f2, f3 from blah_active where f1
> in (select f1 from move_records_tmp)"
>
> After the move a VACUUM FULL ANALYZE is issued. It appears from the logs
> written by the maintenace script that the assertion failure occured during
> the vacuum.
>
> I will install RC1 and see if it can be reproduced.
Please be sure to also configure with --enable-debug (you obviously
already have --enable-cassert), and get a stack trace ("bt" command in
gdb) if at all possible. And keep the core dump somewhere safe -- we
will ask you questions about it.
If you already have a core dump, please provide a stack trace.
--
Alvaro Herrera http://www.advogato.org/person/alvherre
"La naturaleza, tan frágil, tan expuesta a la muerte... y tan viva"
В списке pgsql-bugs по дате отправления: