Re: PANIC in GIN code

Поиск
Список
Период
Сортировка
От Heikki Linnakangas
Тема Re: PANIC in GIN code
Дата
Msg-id 55927423.2000909@iki.fi
обсуждение исходный текст
Ответ на Re: PANIC in GIN code  (Jeff Janes <jeff.janes@gmail.com>)
Список pgsql-hackers
On 06/30/2015 02:18 AM, Jeff Janes wrote:
> On Mon, Jun 29, 2015 at 2:08 PM, Heikki Linnakangas <hlinnaka@iki.fi> wrote:
>
>> I just pushed a fix for this, but unfortunately it didn't make it
>> 9.5alpha1.
>
> Thanks.  I think that that fixed it.  It survived for over an hour this
> time.

Thanks. I grepped through all the other call sites of 
XLogInitBufferForRedo() to look for similar bugs, and sure enough, there 
was one more bug of the same sort, in B-tree page deletion. That one was 
in 9.4 as well. Fixed that too, and added an assertion into 
PageGetSpecialPointer() that would help to catch this sort of bugs 
earlier in the future.

- Heikki




В списке pgsql-hackers по дате отправления:

Предыдущее
От: David Rowley
Дата:
Сообщение: Re: Refactor to split nodeAgg.c?
Следующее
От: Petr Jelinek
Дата:
Сообщение: Re: Dereferenced pointer in tablesample.c