From: | Magnus Hagander <magnus(at)hagander(dot)net> |
---|---|
To: | Simon Riggs <simon(at)2ndquadrant(dot)com> |
Cc: | pgsql-hackers(at)postgresql(dot)org, Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com> |
Subject: | Re: Summary and Plan for Hot Standby |
Date: | 2009-11-15 10:03:28 |
Message-ID: | 9837222c0911150203i3e7ff384m34979d99535e14fb@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Sunday, November 15, 2009, Simon Riggs <simon(at)2ndquadrant(dot)com> wrote:
> On Sun, 2009-11-15 at 10:00 +0100, Magnus Hagander wrote:
>
>> What does the time depend on?
>
> We need to wait for all current transactions to complete. (i.e. any
> backend that has (or could) take an xid or an AccessExclusiveLock before
> it commits.). Similar-ish to the wait for a CREATE INDEX CONCURRENTLY.
>
> The standby already performs this wait in the case where we overflow the
> snapshot, so we have >64 subtransactions on *any* current transaction on
> the master. The reason for that is (again) performance on master: we
> choose not to WAL log new subtransactions.
>
> There are various ways around this and I'm certain we'll come up with
> something ingenious but my main point is that we don't need to wait for
> this issue to be solved in order for HS to be usable.
Yeah, with that explanation (thanks for clearing it up) I agree - it
will definitely still be hugely useful even with this restriction, so
we realy don't need to delay an initial (or the alpha at least)
commit.
Thus, +1 on the second one as well :)
--
Magnus Hagander
Me: http://www.hagander.net/
Work: http://www.redpill-linpro.com/
From | Date | Subject | |
---|---|---|---|
Next Message | Simon Riggs | 2009-11-15 10:25:04 | Re: Hot standby, overflowed snapshots, testing |
Previous Message | Simon Riggs | 2009-11-15 09:43:52 | Re: Summary and Plan for Hot Standby |