Re: Simple DB presence verifier.

Assuming u are familiar with TCP/IP, the problem is basically to see if
there is a TCP server listening at port 5432.
So the smallest overhead will be to try connecting to the IP of the
server on the said port,
if the connection is successful, disconnect and conclude that the server
is running.
I would not recommend keeping one client connected to to the server and
sending a query every now and then
just to verify server is running. I think the above check is good enough.
I remember u could create a client TCP socket using Java.
--Abbas

Damian C wrote:

> Hello,
> This is a real novice question! We are developing a distributed
> (~1-10 PCs) Java office application that uses Postgres as the central
> data store. We are capable Java developers, but our Postgres skills
> are limited because it is primarily hidden behind Hibernate ORM. Our
> product is aimed at a "technically illiterate small business" market,
> so there will be no fancy equipment or system administrators, and
> "forgetting to turn on the server" will be a common fail case.
>
> Our use case involves the java desktops checking to see if Postgres is
> available at (1) startup, and (2) ongoing "watchdog", perhaps every
> minute or so.
>
> My first attempt has been to write a query such as "SELECT last_value
> FROM est_sequence". If the query fails then we can conclude that the
> database is not present.
>
> What we are looking for is a simple, low overhead, "quick succeed,
> catchable fail", way to check if the database is available.
>
> Can anyone suggest an improved technique to check for database presence??
>
> All suggestions welcome.
> Many thanks,
> -Damian
>
> ---------------------------(end of broadcast)---------------------------
> TIP 9: In versions below 8.0, the planner will ignore your desire to
> choose an index scan if your joining column's datatypes do not
> match
>