Re: Missing CONCURRENT VACUUM (Was: Release notes for

On K, 2005-08-17 at 23:58 +0300, Hannu Krosing wrote:
> On K, 2005-08-17 at 16:45 -0400, Tom Lane wrote:
> > Hannu Krosing  writes:
> > > On K, 2005-08-17 at 14:48 -0400, Tom Lane wrote:
> > >> While testing this I realized that it does not in fact work as
> > >> advertised.  It will only exclude long-running VACUUMs from other
> > >> VACUUMs' OldestXmin if *all* the transactions in the system are lazy
> > >> VACUUMs.  If there is even one regular transaction in the system,
> > >> that transaction will include the VACUUMs in its MyProc->xmin, and
> > >> thence GetOldestXmin will have to include them in its result.
> > 
> > > Only if these regular transactions are running in SERIALIZABLE isolation
> > > level, else MyProc->xmin is not set inside GetSnapshotData.
> > 
> > Better read the code again.  The first snap in *any* transaction sets
> > MyProc->xmin.
> 
> Can't find the place :(
> 
> Could you point to the file / function that does this.

Nevermind. Found it - GetTransactionSnapshot() always gets a
serializable snapshot first.

-- 
Hannu Krosing