"Tom Lane" <tgl@sss.pgh.pa.us> writes:
> Gregory Stark <stark@enterprisedb.com> writes:
>> We're seeing a problem where occasionally a process appears to be granted a
>> lock but miss its semaphore signal.
>
> Kernel bug maybe? What's the platform?
It does sound like it given the way my description went. I was worried it may
be some code path not setting waitStatus properly or the compiler caching it
incorrectly somehow.
But now that I check I see it's a pretty old kernel version (Linux 2.6.5) on
the machine. The schedule was completely rewritten by Nick Piggin since then.
I've suggested upgrading it but that may leave us in a bit of a pickle. If we
update it then that throws away all the benchmark history since we can't
really compare new benchmarks against old ones any more.
-- Gregory Stark EnterpriseDB http://www.enterprisedb.com