PATCH: CheckAlive on the maintenance database

Hi,

Steps to reproduce the problem:

1. Connect a server.

    i.e. PostgreSQL 9.0

2. Go to the Database->postgr"es->Schemas->public

    PS: 'postgres' is the maintenance database for the 'PostgreSQL 9.0'.

3. Stop the server in background

    i.e. bin/pg_ctl -D data -m fast stop

4. Click on any of the object under the same database node.

    i.e. Database->postgres->Schemas->public->Functions->abc(text, integer)

You will get an error:

"An error occurred:

FATAL: terminating connection due to administrator command

FATAL: terminating connection due to administrator command"

Error occurred, but nothing happens.

5. Refresh the 'PostgreSQL 9.0' server node, using right click "Refresh".

6. It will collapse all the children nodes and still showed connected.

And, that server node becomes unusable afterwards.

We will have to restart the pgAdmin3 application to use it.

I was able to see this problem on both Ubuntu 10.04 & RHEL 5.5.

Never tried on windows though.

Expected behaviour, it should ask for reconnection after the error message.

Attached are the patches for both master & REL-1_14.

I believe  - these are still not complete patches, because the user will have to click on other node for the message reconnecting the server.