Tom Lane wrote:
> Magnus Hagander <magnus@hagander.net> writes:
>> Tom Lane wrote:
>>> pg_control is certainly not ever deleted or renamed, and in fact I
>>> believe there's an LWLock enforcing that only one PG process at a time
>>> is even touching it. So we need another theory to explain this one :-(
>
>> Right. What we need is a list of which processes have handles open to
>> the file, which can be dumped using Process Explorer (there are other
>> sysinternals tools to do it as well, but PE is probably the easiest)-
>
> Hmm, are you just assuming that the underlying error is
> ERROR_SHARING_VIOLATION? One of the things that's bothered me all along
> is that there are a dozen different Windows error codes that we map to
> EACCES ... perhaps it's time to think about disambiguating that a bit
> better?
I was. Using PE is just one way to prove that was it :-)
But yeah, that's probably a good idea. A quick look at the code says we
should at least ask people who have this problem to give it a run with
logging at DEBUG5 which should then log exactly what the errorcode was.
Or are you seeing more places that need such logging first?
//Magnus