"Zeugswetter Andreas SB SD" <[EMAIL PROTECTED]> writes:
>> regression=# SELECT update_pg_pwd();
>> ERROR: getTypeOutputInfo: Cache lookup of type 0 failed
> Wouldn't it be nice to make this a feature that allows
> stored procedures (void update_pg_pwd ()) ? Correctly register
> this function to
> Removing the special-case logic in ExecTypeFromTL yields
>
> regression=# SELECT update_pg_pwd();
> ERROR: getTypeOutputInfo: Cache lookup of type 0 failed
Wouldn't it be nice to make this a feature that allows
stored procedures (void update_pg_pwd ()) ? Correctly register
this function to n
> regression=# SELECT update_pg_pwd();
> ERROR: getTypeOutputInfo: Cache lookup of type 0 failed
>
> which is not exactly pretty, but it beats a core dump. "SELECT NULL"
> still works.
Maybe the regression test database should have tests for all the built-in
functions?
Chris
Bruce Momjian <[EMAIL PROTECTED]> writes:
> Feel free to wack it around.
Removing the special-case logic in ExecTypeFromTL yields
regression=# SELECT update_pg_pwd();
ERROR: getTypeOutputInfo: Cache lookup of type 0 failed
which is not exactly pretty, but it beats a core dump. "SELECT NULL"
s
Tom Lane wrote:
> Bruce Momjian <[EMAIL PROTECTED]> writes:
> > If I do this as any user:
> > SELECT update_pg_pwd();
> > it crashes all backends and causes a server-wide restart. Is this
> > acceptable behavior?
>
> There are a number of things we might blame this on, all having to do
> wit
Bruce Momjian <[EMAIL PROTECTED]> writes:
> If I do this as any user:
> SELECT update_pg_pwd();
> it crashes all backends and causes a server-wide restart. Is this
> acceptable behavior?
There are a number of things we might blame this on, all having to do
with the overuse of type OID zero
Does the same thing here.
Sounds like a serious problem to me too.
Greg
On Wed, 2002-03-20 at 20:46, Bruce Momjian wrote:
> If I do this as any user:
>
> SELECT update_pg_pwd();
>
> it crashes all backends and causes a server-wide restart. Is this
> acceptable behavior? I am sure ther
If I do this as any user:
SELECT update_pg_pwd();
it crashes all backends and causes a server-wide restart. Is this
acceptable behavior? I am sure there are other cases too. Isn't it a
problem that we let ordinary users crash the server and cause a restart?
--
