Re: Longer startup delay (was Re: Single Index Tuple Chain (SITC) method)

Alvaro Herrera <alvherre(at)commandprompt(dot)com> writes:
> Speaking of which, I think I've noticed a longer delay in server start
> after initdb. I haven't measured nor profiled it, but I think it may be
> because of the heap_inplace_update xlogging that we weren't doing
> previously.

Can't say that I've noticed anything like that. I have gotten a nagging
sense lately that there are strange delays happening at random times
during the regression tests. Like you I'm not quite sure about this,
but I run "make installcheck" often enough to be fairly attuned to it,
and it just seems like it's sometimes hanging up for a couple of seconds
for no apparent reason. OTOH we've stuck some new, slow queries into
the tests recently and maybe I just haven't acclimated to that.

regards, tom lane