I should have been good enough to post the answer to this one too, but,
only got it going a little over a week ago.
It appears that adding in support for microsoft's VC flag being zero is
the culprit. Not in prior releases, this flag, when sent with a value of
zero means to kill existing connections, cause the client doesn't think
there should be any. Well, what the client doesn't know is what kills
us. Other connections can be done to the same server through other
names, and so, VC=0 happens quite often.
The fix is, unfortunately, in a patch. I'm sure the next release code
base will probably contain it too. I am not quite sure to even call it
a fix, it isn't really a bug, but rather a way to shut off this check. On
an updated system, you have a new flag that can be set so that the
check of the flag being zero is ignored. Presto, fixed.
Note, this has struck most 'compliant' servers, each having to do
something similar.
Side note: only happens on 445 connections. On 139 connections,
it doesn't occur (since the encapsulation doesn't contain it). MS
references the same difference between 445 and 139 with the same
VC=0 flag when people are doing proxying .
Tom
Terrol Guyah wrote:
Hello ..
Try this link
http://support.microsoft.com/kb/2444558
Hope it helps!
_______________________________________________
cifs-discuss mailing list
cifs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/cifs-discuss
