Pre-daemonise steps in child process (was: PEP 3143 and customizations)

[Ben Finney]

On 21-May-2014, Helmut Grohne wrote:
> Yeah, if there was no need to do setup in the child, the whole
> problem would go away. Just setup first and then daemonize.

That's the procedure envisaged when I wrote the code, yes.

> Unfortunately some things must be set up in the child, but with the
> current implementation the child can no longer signal any failures
> back to the user. All it can do is fail silently. Bad.

Can you point to, and/or write, a minimal example program which I can
run to see this case?

Ideally it should be suitable for a unit test:

* Fail in an obvious way that is due to the lack of behaviour you're
  expecting.

* Succeed when the issue being described is addressed.

As it stands, I'm having difficulty understanding exactly why this
feature is needed. A minimal example may help me to see the
distinction.

-- 
 \       “Pray, v. To ask that the laws of the universe be annulled in |
  `\     behalf of a single petitioner confessedly unworthy.” —Ambrose |
_o__)                           Bierce, _The Devil's Dictionary_, 1906 |
Ben Finney <ben at benfinney.id.au>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.alioth.debian.org/pipermail/python-daemon-devel/attachments/20140717/50bb407a/attachment.sig>