Re: Another possible corruption bug in 9.3.2 or possibly a known MultiXact problem?

On 28 Feb 2014 06:19, "Andres Freund" <andres(at)2ndquadrant(dot)com> wrote:
>
> On 2014-02-27 23:41:08 +0000, Greg Stark wrote:
> > Though I notice something I can't understand here.
> >
> > After activating the new clone subsequent attempts to select rows from
> > the page bump the LSN, presumably due to touching hint bits (since the
> > prune xid hasn't changed). But the checksum hasn't changed even after
> > running CHECKPOINT.
>
> Are you running with full_page_writes=off?

No

> Only delete and update do a PageSetPrunable(), so prune_xid not being
> changed doesn't say much...
>
> > How is it possible for the LSN to get updated without changing the
checksum?
>
> Generally the LSN is computed when writing, not when a buffer is
> modified, so that's not particularly surprising. It'd be interesting to
> see what the records are that end on those LSNs.

The checksum you mean? But that's why I ran checkpoint.

> It'd probably nice to add the capability to dump records that end in a
> particular location to pg_xlogdump...

I have this crazy idea about combining xlogdump and pg_receivexlog to
archive all xlog to a postgres database for querying.....