Re: XLogInsert scaling, revisited
| От | Jeff Janes |
|---|---|
| Тема | Re: XLogInsert scaling, revisited |
| Дата | |
| Msg-id | CAMkU=1wEz3bhbY9KuRoYZh-7tXajXr95w36tgBCPFD4KfmLYjg@mail.gmail.com обсуждение исходный текст |
| Ответ на | Re: XLogInsert scaling, revisited (Heikki Linnakangas <hlinnakangas@vmware.com>) |
| Список | pgsql-hackers |
On Sat, Jun 22, 2013 at 4:32 AM, Heikki Linnakangas <hlinnakangas@vmware.com> wrote:
Hmm, it looks like the xlog-switch is trying to wait for itself to finish. The concurrent TRUNCATE is just being blocked behind the xlog-switch, which is stuck on itself.
I wasn't able to reproduce exactly that, but I got a PANIC by running pgbench and concurrently doing "select pg_switch_xlog()" many times in psql.
Attached is a new version that fixes at least the problem I saw. Not sure if it fixes what you saw, but it's worth a try. How easily can you reproduce that?
With v23, it got stuck both times I tried it, once after 4 hours and once after 6 hours.
With v24, it has been running for 30 hours so far with no problems. So there is a pretty good chance that it is fixed.
This one? http://www.postgresql.org/message-id/CAMkU=1xoA6Fdyoj_4fMLqpicZR1V9GP7cLnXJdHU+iGgqb6WUw@mail.gmail.comThis is using the same testing harness as in the last round of this patch.
Yes. I have cleaned it up some and added use of checksum, I don't know if any of those things are needed to invoke the problem.
Cheers,
Jeff
В списке pgsql-hackers по дате отправления: