Re: Patch for fail-back without fresh backup

On Tue, Oct 8, 2013 at 3:16 PM, Andres Freund <andres(at)2ndquadrant(dot)com>wrote:

>
>
> It is my impression that there still are several people having pretty
> fundamental doubts about this approach in general. From what I remember
> neither Heikki, Simon, Tom nor me were really convinced about this
> approach.
>
>
IIRC you and Tom were particularly skeptical about the approach. But do you
see a technical flaw or a show stopper with the approach ? Heikki has
written pg_rewind which is really very cool. But it fails to handle the
hint bit updates which are not WAL logged unless of course checksums are
turned on. We can have a GUC controlled option to turn WAL logging on for
hint bit updates and then use pg_rewind for the purpose. But I did not see
any agreement on that either. Performance implication of WAL logging every
hint bit update could be huge.

Simon has raised usability concerns that Sawada-san and Samrat have tried
to address by following his suggestions. I am not fully convinced though we
have got that right. But then there is hardly any feedback on that aspect
lately.

In general, from the discussion it seems that the patch is trying to solve
a real problem. Even though Tom and you feel that rsync is probably good
enough and more trustworthy than any other approach, my feeling is that
many including Fujii-san still disagree with that argument based on real
user feedback. So where do we go from here ? I think it will really help
Sawada-san and Samrat if we can provide them some solid feedback and
approach to take.

Lately, I was thinking if we could do something else to track file system
updates without relying on WAL inspection and then use pg_rewind to solve
this problem. Some sort of prelaod library mechanism is one such
possibility. But haven't really thought through this entirely.

Thanks,
Pavan

--
Pavan Deolasee
http://www.linkedin.com/in/pavandeolasee