On Fri, Apr 01, 2011 at 09:25:30AM +0200, Gabriele Paggi wrote: > Hello Volker, > > thank for your reply! > > > Where is the real error here? This seems completely normal > > to me with such a high debug level. > > The error I'm referring to is: > > [2011/03/31 15:15:02, 2] winbindd/winbindd.c:remove_client(744) > final write to client failed: Broken pipe > [2011/03/31 15:15:02, 2] winbindd/winbindd.c:remove_client(744) > final write to client failed: Broken pipe > > I can of course disregard it but I would still be wondering why it fails > writing to the client :)
Oh, that's in response to a Solaris bug. Instead of just closing the socket we write one byte before closing it. The bug was that the winbind client did not get a 0 read size when we just close the server end of a unix domain socket and thus went into a timeout. Without this bug it can happen that we can't write to that socket anymore, but it's very hard to detect at runtime whether we have this bug. So we live with this error message. Volker -- SerNet GmbH, Bahnhofsallee 1b, 37081 Göttingen phone: +49-551-370000-0, fax: +49-551-370000-9 AG Göttingen, HRB 2816, GF: Dr. Johannes Loxen -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/options/samba
