On Dec 28, 2010 3:58 PM, "Tom Lane" <t...@sss.pgh.pa.us> wrote: > > Magnus Hagander <mag...@hagander.net> writes: > > Attached patch implements a function called pg_primary_conninfo() that > > returns, well, the primary_conninfo used on the standby when in > > streaming replication mode (otherwise NULL). > > > Objections? > > What's the use case? And aren't there security reasons to NOT expose > that? It might contain a password for instance.
Good point - should be made superuser only. > > > + if (recptr.xlogid == 0 && recptr.xrecoff == 0 && conninfo[0] != '\0') > > + PG_RETURN_NULL(); > > This test seems a bit incoherent. I used that to test that streaming repl is enabled at all. Is there a better way? /Magnus
