Re: restore_command return code behaviour
От | Jacob Champion |
---|---|
Тема | Re: restore_command return code behaviour |
Дата | |
Msg-id | CAOYmi+kL2kATavDhJ7dwJpA0vnJjFB+k7XiRqQZLzeTZdTd__w@mail.gmail.com обсуждение исходный текст |
Ответ на | Re: restore_command return code behaviour ("David G. Johnston" <david.g.johnston@gmail.com>) |
Ответы |
Re: restore_command return code behaviour
|
Список | pgsql-hackers |
On Mon, Jul 28, 2025 at 2:42 PM David G. Johnston <david.g.johnston@gmail.com> wrote: > I don’t understand calling out sigterm as an exception, the same abort-and-shutdown action happens there too. RestoreArchivedFile() has a special case for SIGTERM, though? > And in any case signals are turned into exit status values anyway so naming them specifically seems redundant. > > The “Command not found” error is defined by POSIX as exit status 127 making its specification redundant with > 125 I don't think either is redundant from the point of view of the targeted audience, who may not understand the overlap in the POSIX specification. "My command returned", "my command died", and "my command never ran" are interesting cases to have to consider (and I think it's unfortunate that we can't reliably tell them apart). --Jacob
В списке pgsql-hackers по дате отправления: