Comment 4 for bug 387682

Ah, that's interesting. Your problem is a duplicate of bug 264336. However, I was unable so far to reproduce this. I just wonder why starting postgresql a little later suddenly makes _more_ shared memory available; that's surprising because usually the processes that start earlier will consume shared memory, not free it.

Can you please give me the output of "ls -l /etc/rc2.d/S20*"? You can confirm that "sudo mv /etc/rc2.d/S19postgresql-8.3 /etc/rc2.d/S20postgresql-8.3" reliably fixes the problem?