Re: pg_ctl/pg_rewind tests vs. slow AIX buildfarm members - Mailing list pgsql-hackers

From Tom Lane
Subject Re: pg_ctl/pg_rewind tests vs. slow AIX buildfarm members
Date
Msg-id 32333.1441310839@sss.pgh.pa.us
Whole thread Raw
In response to Re: pg_ctl/pg_rewind tests vs. slow AIX buildfarm members  (Andrew Dunstan <andrew@dunslane.net>)
Responses Re: pg_ctl/pg_rewind tests vs. slow AIX buildfarm members  (Alvaro Herrera <alvherre@2ndquadrant.com>)
List pgsql-hackers
Andrew Dunstan <andrew@dunslane.net> writes:
> There is no equivalent of execl, nor a cmd.exe exquivalent of the 
> shell's exec. But surely the equivalent of the fork/execl you're doing 
> here would be a simple CreateProcess(). I don't see why you need a shell 
> in the middle on Windows at all.

The problem is to get the output redirection to work.  I imagine it could
be rewritten without involving the shell, but it'd be less than trivial
(or at least, I'm not going to do it).

If we did get that to work, it could probably be applied to the
service-start case as well, which currently just does a CreateProcess and
pays no attention to what happens to postmaster stderr.
        regards, tom lane



pgsql-hackers by date:

Previous
From: Andrew Dunstan
Date:
Subject: Re: pg_ctl/pg_rewind tests vs. slow AIX buildfarm members
Next
From: Pavel Stehule
Date:
Subject: Re: On-demand running query plans using auto_explain and signals