SEO

Magnus Hagander wrote:
> Updated patch attached.

A few more comments:

- why did you add the client address to PgStat_MsgHdr, rather than
PgStat_MsgBestart? It would be nice to avoid sending the client address
for each and every stats message, as it shouldn't change over the life
of the session.

- I think the backend_client_addr and backend_client_port view columns
should be named "client_addr" and "client_port", respectively.

- Is there a reason you need to expose and then call the private
network_in() function, rather than using inet_in(), which is already public?

- Is the backend startup time sensitive information? Considering this
information is available via ps(1), perhaps it would be better to allow
any user to see any backend's startup time, rather than providing a
false sense of security.

- We return (NULL, -1) for the client address/port if connected via a
unix domain socket, and (NULL, NULL) for the address/port if the
selecting user isn't a superuser / the owner of that client connection.
It seems a little ugly to return NULL for the address in both cases,
since the same value is used for two completely different meanings. Not
sure of a better convention, though -- just idly complaining

You needn't respin the patch, as I can make the necessary changes myself
-- I'm just looking for comments on the above before applying.

-Neil

---------------------------(end of broadcast)---------------------------
TIP 6: Have you searched our list archives?

http://archives.postgresql.org

Neil Conway <neilc@samurai.com> writes:
> - Is the backend startup time sensitive information? Considering this
> information is available via ps(1), perhaps it would be better to allow
> any user to see any backend's startup time, rather than providing a
> false sense of security.

Remote database users don't have access to ps(1). I think that hiding
the startup time is reasonable.

> - We return (NULL, -1) for the client address/port if connected via a
> unix domain socket, and (NULL, NULL) for the address/port if the
> selecting user isn't a superuser / the owner of that client connection.
> It seems a little ugly to return NULL for the address in both cases,
> since the same value is used for two completely different meanings. Not
> sure of a better convention, though -- just idly complaining

Perhaps 0.0.0.0/0 or some such for the address in the unix domain case?
Not sure that this is an improvement over NULL though.

Other comments seem on-target to me.

regards, tom lane

---------------------------(end of broadcast)---------------------------
TIP 4: Don't 'kill -9' the postmaster