[issue38119] resource tracker destroys shared memory segments when other processes should still have valid access

[Steve Newcomb]

Steve Newcomb <srn at coolheads.com> added the comment:

Sometimes a leak is exactly what's wanted, i.e. a standing block of shared memory that allows sharing processes come and go ad libitum.  I mention this because I haven't seen anyone mention it explicitly.  

While turicas's monkeypatch covers the use case in which a manually-named block of shared memory is intended to remain indefinitely, it would be best if future versions of shared_memory allowed for such a use case, too.  It shouldn't be necessary to monkeypatch in order to have a standing block of shared memory remain ready and waiting even when nobody's using it.

----------
nosy: +steve.newcomb

_______________________________________
Python tracker <report at bugs.python.org>
<https://bugs.python.org/issue38119>
_______________________________________