Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> writes:
> On Fri, Sep 23, 2016 at 1:21 AM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>> Robert Haas <robertmhaas(at)gmail(dot)com> writes:
>>> So, we could have dsm_postmaster_startup() seed the random number
>>> generator itself, and then let PostmasterRandom() override the seed
>>> later. Like maybe:
>>
>> Works for me, at least as a temporary solution.
> Isn't it better if we use the same technique in dsm_create() as well
> which uses random() for handle?
dsm_create() is executed in backends, not the postmaster, and they
already have their own random seeds (cf BackendRun). Adding more
srandom calls at random places will *not* make things better.
However, it's certainly true that dsm_postmaster_startup() might not
be the only place in postmaster startup that wants to use random().
What seems like the best thing after sleeping on it is to put
"srandom(time(NULL))" somewhere early in PostmasterMain, so that one
such call suffices for all uses during postmaster startup.
regards, tom lane