Hi,
On 2020-02-04 10:15:01 +0530, Kuntal Ghosh wrote:
> I performed the same test in pg11 and reproduced the issue on the
> commit prior to a4ccc1cef5a04 (Generational memory allocator).
>
> ulimit -s 1024
> ulimit -v 300000
>
> wal_level = logical
> max_replication_slots = 4
>
> [...]
> After that, I applied the "Generational memory allocator" patch and
> that solved the issue. From the error message, it is evident that the
> underlying code is trying to allocate a MaxTupleSize memory for each
> tuple. So, I re-introduced the following lines (which are removed by
> a4ccc1cef5a04) on top of the patch:
> --- a/src/backend/replication/logical/reorderbuffer.c
> +++ b/src/backend/replication/logical/reorderbuffer.c
> @@ -417,6 +417,9 @@ ReorderBufferGetTupleBuf(ReorderBuffer *rb, Size tuple_len)
>
> alloc_len = tuple_len + SizeofHeapTupleHeader;
>
> + if (alloc_len < MaxHeapTupleSize)
> + alloc_len = MaxHeapTupleSize;
Maybe I'm being slow here - but what does this actually prove? Before
the generation contexts were introduced we avoided fragmentation (which
would make things unusably slow) using a a brute force method (namely
forcing all tuple allocations to be of the same/maximum size).
Which means that yes, we'll need more memory than necessary. Do you
think you see anything but that here?
It's good that the situation is better now, but I don't think this means
we need to necessarily backpatch something nontrivial?
Greetings,
Andres Freund
