Gary Mills wrote:
On Mon, Nov 05, 2007 at 05:02:24PM -0500, Bob Doolittle wrote:
Gary Mills wrote:
On Fri, Nov 02, 2007 at 04:16:14PM -0400, Bob Doolittle wrote:
I didn't see any snoop output. Only an
interpretation that at some point UDP traffic
stopped and you got some ICMP Destination
Unreachables from the *server*. The details of
that ICMP message may be illuminating, as well as
what immediately preceded the problem. Is your X
server dying? It's possible that some process on
the server is dying and hence no app is reading
that port any more.

That's something I certainly should check.  I didn't expect that, but
it's possible that the X server is dying.  There is a black screen on
the Sun Ray immediately after I enter a userid and password on the
dtgreet screen.  Would dtlogin start another one?

It should, if it doesn't have a bug. Also, if it tries repeatedly to start a server for a particular display and fails for some reason it'll "blackball" that display number
and refuse any further requests to start a server for that display.

Run "utwho -a" and see which display number is yours when in this state,
then ps and see if there's an Xsun process for that display.

Are you up to date on dtlogin patches?
Anything unusual in /var/dt/Xerrors immediately after the problem
(unfortunately that file rolls over after a short while)?

Here's the snoop output, in part.  I'm more confused than ever now.
This doesn't look like a NAT problem after all.  This is taken on the
interface to the private network, so it will show traffic after NAT
conversion.

Just before the timeout, there's lots of UDP traffic between my Sun Ray
and the Sun Ray server, over a pseudo-connection.  Here's a sample:

  426 9:12:2.15721 toliman.cc.umanitoba.ca -> 192.168.0.158 UDP D=37969 S=40405 
LEN=776
  427 9:12:2.16197 192.168.0.158 -> toliman.cc.umanitoba.ca UDP D=40405 S=37969 
LEN=48
  428 9:12:2.25569 192.168.0.158 -> toliman.cc.umanitoba.ca UDP D=40405 S=37969 
LEN=64
  429 9:12:2.49271 toliman.cc.umanitoba.ca -> 192.168.0.158 UDP D=37969 S=40405 
LEN=48

After the timeout, my Sun Ray sends UDP, but nothing is listening
on the server:

  431 9:12:3.35094 toliman.cc.umanitoba.ca -> 192.168.0.158 UDP D=37969 S=40405 
LEN=24
  432 9:12:3.35144 192.168.0.158 -> toliman.cc.umanitoba.ca UDP D=40405 S=37969 
LEN=48
  433 9:12:3.37728 toliman.cc.umanitoba.ca -> 192.168.0.158 ICMP Destination 
unreachable (UDP port 40405 unreachable)

After that, all UDP traffic stops.  There's still occasional TCP traffic
but only in one direction, from my Sun Ray to the server.  This stays
the same until the Sun Ray terminal reboots.

The tcp/7009 traffic should always be bidirectional because SRSS is
constantly exchanging heartbeats with the client.  If no heartbeats
are sent to the client in some period (a minute or two) the client
will reset.  That seems to be what you're seeing.  So - why should
utauthd become unresponsive at the same time the X server is
dying?  I'd check your system patch levels and make sure you're
up-to-date, then run some hardware diagnostics.

-Bob

_______________________________________________
SunRay-Users mailing list
[email protected]
http://www.filibeto.org/mailman/listinfo/sunray-users

Reply via email to