Tom Lane writes:
> If that actually is the length of the line, the only answer is to raise
> the memory ulimit setting the postmaster runs under.
Did another test to try to see if the problem is that row or the size of the
row.
Another record of greater size also failed.
Any ideas what this 84MB limit could be from?
I have shared_buffers at 450MB
maintenance_work_mem = 64MB (which I increased to 100MB with the same
result)
OS limit for applications at 1.6GB.
Also when I start postgresql I see several postgresql processes using 400M+
so I don't see how it could be the Os limit.