This worry relies on a zero day bug/memory exploit in one of the most widely used access methods for Postgres. This worry can be applied to every component of the software stack, including the OS.
Hmm, not really. Whether the kernel is managing memory for processes properly is different than asking whether a reused Postgres connection clears all relevant memory.
Shared memory is shared memory. Are the pages zeroed out?
This worry relies on a zero day bug/memory exploit in one of the most widely used access methods for Postgres. This worry can be applied to every component of the software stack, including the OS.
Hmm, not really. Whether the kernel is managing memory for processes properly is different than asking whether a reused Postgres connection clears all relevant memory.
But thanks for info about level of issue.