Обсуждение: FATAL logged when starting

We get this FATAL pg_log entry, just after Postgres startup.
(Postgres does start OK)
//9.1.3 on Linux

2013-02-20 10:15:46.637 EST   50eedb22.3602  0
  LOG:  database system is ready to accept connections
2013-02-20 10:15:46.637 EST   50eedb22.3607  0
  LOG:  autovacuum launcher started
2013-02-20 10:15:47.390 EST [unknown] rsaadmin 50eedb23.3609 2/1 0
  FATAL:  no pg_hba.conf entry for host "[local]", user "rsaadmin", databas=
e "postgres", SSL off

Postgres is run under the linux user account "rsaadmin", but we don't have =
a Postgres role by that name.
We use [-w] as an option for pg_ctl start.

As I read postgresql.org/docs/9.1/static/app-pg-ctl.html,

  -w
  . . . When waiting for startup, pg_ctl repeatedly attempts to connect to =
the server . . .


My question:
Is there a way for me to specify a DIFFERENT USER, that pg_ctl will use to =
determine postgres has started ?

If not, how do people work around getting this nasty FATAL log message ?

Thanks,
-dvs-

"Sahagian, David" <david.sahagian@emc.com> writes:
> We get this FATAL pg_log entry, just after Postgres startup.
>   FATAL:  no pg_hba.conf entry for host "[local]", user "rsaadmin", database "postgres", SSL off

> Postgres is run under the linux user account "rsaadmin", but we don't have a Postgres role by that name.

> Is there a way for me to specify a DIFFERENT USER, that pg_ctl will use to determine postgres has started ?

I think you could set PGUSER in pg_ctl's environment.

> If not, how do people work around getting this nasty FATAL log message ?

You could just ignore it ... it's harmless.

            regards, tom lane