Andres Freund <andres@anarazel.de> writes:
> On 2018-09-11 12:26:44 -0400, Tom Lane wrote:
>> Well, there remains the fact that we've seen no field reports that seem
>> to trace to failure-to-acquire-AEL since 9.6 came out. So arguing that
>> this *could* be a probable scenario fails to comport with the available
>> evidence.
> It might even be that we've seen reports of this, but didn't attribute
> the errors correctly.
Yeah, that's certainly possible. One good thing about the change I'm
recommending is that the symptom would be very clear (ie, "out of shared
memory" from the startup process). If we do start getting reports of it,
we'd know where the problem is.
>> My inclination is to fix it as I've suggested and wait to see if there
>> are field complaints before expending a whole lot of effort to create
>> a better solution. In any case, I am not willing to create that better
>> solution myself, and neither is Robert; are you?
> On master I'd be ok with that, but on the backbranches that seems like
> playing with user's setups.
I am not sure which part of "I will not fix this" you didn't understand.
*If* we get clear evidence that it happens for a non-negligible number
of users, I might be willing to reconsider. Right now my reading of
the evidence is that it hasn't, and won't; so I judge that putting in
a complex mechanism to try to cope with the situation would be a net
loss for reliability. Back-patching such a mechanism seems like it'd
be an even worse engineering decision.
regards, tom lane