[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Beta release of LDM version 5.1.2 available

>To: address@hidden
>From:  Harry Edmon <address@hidden>
>Subject: Re: 20000804: Beta release of LDM version 5.1.2 available 
>Organization: Dept of Atmospheric Sciences, University of Washington
>Keywords: DEC, socklen_t

Harry,

> I am having problems on my new DEC machine with 5.1.2.  I first had to define
> _POSIX_PII_SOCKET to get the definition of socklen_t to be a long instead of 
> an
> int.  

Oops, I didn't know about _POSIX_PII_SOCKET.  Check what the man page
says the type should be for the last parameter to accept(3n) in the
socket API.  I'm assuming it is a socklen_t*, but if not, then maybe
your socket library doesn't correspond to the interfaces you get with
_POSIX_PII_SOCKET defined.  Are there multiple versions of the socket
library?  If so, we need to get the one where the last parameter to
accept is a socklen_t* and socklen_t is defined to correspond to what
the library expects ...

> But if I stop the LDM and start it again without recreating the queue, pqact 
> and
> pqbinstats behave properly.  My guess is that pqact and pqbinstats don't like 
> a
> brand new empty queue.

Hmm, we haven't seen this on Solaris or IRIX.  But I'll see if I can
see why this might happen.

> If you want access to this machine to try this out, let me know.

Thanks for the offer.  I may take you up on it next week, if I can't
figure out what is going on from the traces you provided.

--Russ