James Mansion wrote:
> Is the intent to replace most uses of (pg)kill with a general
> purpose messaging system between the processes, or
> (just) to address notify per se?
>
> (Presumably with 'fire-and-forget' and also rpc
> semantics? And pub-sub? And some sort of
> write to an fd protected by an atomic flag to
> elide multiple writes when the process hasn't woken
> and acknowledged the ATTN yet?)
>
> If pgkill is not used for signalling, could this reduce the reliance
> on signals
> (except for trying to kill off processes) to the point
> where ot becomes much less scary to link to libraries
> that use signals themselves and/or threaded runtimes?
My intention is to revamp the listen/notify system, pure and simple.
If you have an alternative suggestion them you need to make it now.
We are not intending to use FDs for message passing. They will be stored
in shared memory. See previous discussions for details:
http://groups.google.com/group/pgsql.hackers/browse_frm/thread/e63a5ac43e2508ce/e0892fb3316cc327?hl=en&lnk=gst&q=notify+payload#e0892fb3316cc327
http://groups.google.com/group/pgsql.hackers/browse_frm/thread/6a59675a3e11bedc/87e0ce6dd6cce6a6?hl=en&lnk=gst&q=notification+payload#87e0ce6dd6cce6a6
cheers
andrew
