Daniele Varrazzo

You can attach gcc to the stuck process, or kill it generating a core.

You can also try to make the process responsive again (killable at python level) switchin...

With AWS or with regular Postgres?

You may be bumping in some quirkiness of AWS pooling or throttling that the libpq is not prepared to handle. Getting a C stac...

Ticket #112 you reference is about queries, not about connections.

The driver is not referencing a deallocated object: it destroyed one and allocated a new one ...

For what I can see here it's the server not completing the connection. I tend to blame the AWS servers, not the client side.

If you want to perform operation "s...

A way to make the query interruptable is to work in green mode with a wait callback that can tolerate being broken by ctrl-c:

from select import select
from ps...

A way to make the query interruptable is to work in green mode with a wait callback that can tolerate being broken by ctrl-c:

from select import select
from ps...

Please install python-dev package (and libpq-dev too).

You are observing PostgreSQL's behaviours, not psycopg.

piro=> select '[1,10]'::int4range;
 int4range 
-----------
 [1,11)
(1 row)

Check the docs about the c...

Hello,

yes, I know the issue. No, it's not fixed in the master and I don't know how to fix it; actually I don't think it's generally fixable in psycopg: you may...

Fixed: the segfault was caused by a missing check in a path normally not used, triggered executing a COPY statement via execute().