Re: hung backends stuck in spinlock heavy endless loop

Поиск
Список
Период
Сортировка
От Peter Geoghegan
Тема Re: hung backends stuck in spinlock heavy endless loop
Дата
Msg-id CAM3SWZTfE0vHzYkd9qXwcfq5xTRPwdLDM6s8ShLyfcbH=4=0Sw@mail.gmail.com
обсуждение исходный текст
Ответ на hung backends stuck in spinlock heavy endless loop  (Merlin Moncure <mmoncure@gmail.com>)
Ответы Re: hung backends stuck in spinlock heavy endless loop  (Andres Freund <andres@2ndquadrant.com>)
Список pgsql-hackers
On Tue, Jan 13, 2015 at 2:29 PM, Merlin Moncure <mmoncure@gmail.com> wrote:
> On my workstation today (running vanilla 9.4.0) I was testing some new
> code that does aggressive parallel loading to a couple of tables.

Could you give more details, please? For example, I'd like to see
representative data, or at least the table definitions involved. Do
you have any idea what index is involved with the _bt_compare() calls
above? I'm curious as to what its definition is.

You might also consider using my B-Tree verification tool on a
preserved data directory:
http://www.postgresql.org/message-id/CAM3SWZRK2YxKegj94_2wKs_vbEMJeHc1yE23ECsS01wRAKzkjg@mail.gmail.com

I'm inclined to think that this is a livelock, and so the problem
isn't evident from the structure of the B-Tree, but it can't hurt to
check. The tool is bleeding edge, so don't use it in production.

Thanks
-- 
Peter Geoghegan



В списке pgsql-hackers по дате отправления:

Предыдущее
От: Merlin Moncure
Дата:
Сообщение: Re: hung backends stuck in spinlock heavy endless loop
Следующее
От: Andres Freund
Дата:
Сообщение: Re: hung backends stuck in spinlock heavy endless loop