Re: Weirdness using Executor Hooks - Mailing list pgsql-hackers

From Eric Ridge
Subject Re: Weirdness using Executor Hooks
Date
Msg-id CANcm6wacpL6W+dR6_MsN3y7qKdbUrQRLA6b_rDAGCgtt2EXxow@mail.gmail.com
Whole thread Raw
In response to Re: Weirdness using Executor Hooks  (Tom Lane <tgl@sss.pgh.pa.us>)
Responses Re: Weirdness using Executor Hooks  (Eric Ridge <eebbrr@gmail.com>)
List pgsql-hackers
On Thu, Jun 18, 2015 at 6:04 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:

> What we were expecting I guess is that such functions would be declared
> to reference the library underneath $libdir/plugins, not that you'd use
> a symlink.

According to pg_proc.probin, all 32 of my C language functions point
to $libdir/my_extension_lib, which makes sense because as part of the
extension .sql I declare them as LANGUAGE c AS 'MODULE_PATHNAME', and
MODULE_PATHNAME gets substituted by make.

So are you saying that we should instead declare them AS
'$libdir/plugins/my_extension_lib' so that it matches what's happening
in local_preload_libraries?  And is it safe to directly update
pg_proc.probin (and ya know, terminate all existing backends)?

Thanks so much for your time!

eric

ps, I think if we just changed our deploy process to terminate
existing backends this would just disappear, yeah?



pgsql-hackers by date:

Previous
From: Petr Jelinek
Date:
Subject: Re: Inheritance planner CPU and memory usage change since 9.3.2
Next
From: Michael Paquier
Date:
Subject: Re: dblink: add polymorphic functions - review