Those look like sessions that have been redirected and just waiting to be reaped. Same story, just helps load balancing of non-card sessions. Shorten the reaper time if you'd like them to go away sooner.

On Aug 18, 2011, at 9:51 AM, Steven Gelsie <[email protected]> wrote:

Craig:

Thanks for the explanation. All of our sunray sessions, either regular or kiosk require a java card to connect to the sunray server. I found the results for utsession -p interesting. None of the Unix ID are defined , they are all ???. We also have another issue related to sessions and that is, you are not always connected back to your existing session. It does not happen that often but it happens. I think this happens most often when the network connection drops when we are using the sunray from home. The following is the result from utsession . I changed user names to initials. Token ID Registered Name Unix ID Disp State ------------------------------ -------------------- ---------- ---- -----
Payflex.5007b89500130100       A, M                 ???        50   D
Payflex.5007b89800130100       G, S                 ???        34   D
Payflex.5007b8d300130100       Visual Enterp room   ???        15
Payflex.5007b8dc00130100       S, D.                ???        42
Payflex.500b878f00130100       G,  S                ???        24   D
Payflex.500b879400130100       Z, J                 ???        37
Payflex.500b879500130100       J, G                 ???        70
Payflex.500b879700130100       D, J                 ???        29   D
hotdesk.IEEE802-002128408969   ???                  ???        76   I
Payflex.500b879c00130100       S, S                 ???        36
Payflex.500b87a000130100       B, D                 ???        38   D
hotdesk.IEEE802-0003bad74fdf   ??? ???        74   I
Payflex.500b87a200130100       P, M                 ???        25
Payflex.500b87a300130100       B, J                 ???        27   D
Payflex.500b87a500130100       D, N                 ???        3    D
hotdesk.IEEE802-002128909909   ???                  ???        57   I
Payflex.500b87a600130100       2nd Floor Lab        ???        23
Payflex.500cacb700130100       M, T                 ???        11   D
Payflex.500cacb800130100       R, T J.              ???        51   D
Payflex.500cacb900130100       H, M.                ???        44
Payflex.500cacbb00130100       K, A                 ???        35   D
Payflex.500d1cb500130200       U, D                 ???        58   D
Payflex.500d1cb900130200       S, C                 ???        48
Payflex.500d1ccf00130200       S, A                 ???        75   D
Payflex.500d1cde00130200       H, B                 ???        19
Payflex.500d1cdf00130200       C, G                 ???        16   D
hotdesk.IEEE802-00144fb1cc2b   ??? ???        17   I
Payflex.501265f400130100       H, W                 ???        47   D
hotdesk.IEEE802-00144fe7d57b   ???                  ???        53   I
Payflex.501265f600130100       B, O M.              ???        26   D
Payflex.501265f700130100       V, M                 ???        69
Payflex.501265fa00130100       S, R                 ???        62   D
hotdesk.IEEE802-00144fd34d26   ???                  ???        49   I
Payflex.501265fe00130100       L, D                 ???        20   D
Payflex.501265ff00130100       B, E                 ???        12
Payflex.5012660000130100       S, S                 ???        22   D
Payflex.5012660100130100       W, D                 ???        55   D
Payflex.5012667d00130100       A, G S               ???        14   D
Payflex.5012668300130100       T, B                 ???        80   D
Payflex.5012669f00130100       F, A                 ???        59   D
Payflex.50127d9800130100       P, K                 ???        56   D
Payflex.50127d9900130100       S, E                 ???        2    D
hotdesk.IEEE802-0021286e1988   ???                  ???        61   I
Payflex.50127d9a00130100       Y, T L.              ???        33   D
Payflex.50127d9f00130100       F, P                 ???        64   D
Payflex.50127da000130100       R, M                 ???        73   D
Payflex.50127da100130100       M , C                ???        4    D
Payflex.50127da200130100       L, J                 ???        60   D
hotdesk.IEEE802-0021286e19d9   ???                  ???        32   I
Payflex.50127da300130100       M, J                 ???        68   D
Payflex.50127da900130100       P, M                 ???        67   D
Payflex.50127daa00130100       TPC Kiosk User5      ???        5
Payflex.50127dab00130100       TPC Kiosk User11     ???        28
Payflex.50127dac00130100       TPC Kiosk User17     ???        21
Payflex.50127dad00130100       TPC Kiosk User16     ???        46
Payflex.5013a80a00130100       K, H                 ???        79   D
Payflex.5013a80c00130100       TPC Kiosk User3      ???        10   D
Payflex.5013a81300130100       TPC Kiosk User1      ???        65
Payflex.5013a81400130100       TPC Kiosk User6      ???        41
Payflex.5013a81500130100       TPC Kiosk User4      ???        6
Payflex.5013a81600130100       TPC Kiosk User8      ???        30
Payflex.5013a81800130100       TPC Kiosk User10     ???        8
Payflex.5013a81900130100       TPC Kiosk User9      ???        40   D
Payflex.5013a81b00130100       TPC Kiosk User12     ???        31
Payflex.5013a81d00130100       TPC Kiosk User14     ???        43
Payflex.5013a84e00130100       M, J                 ???        45   D
Payflex.50165e3700130100       M, A                 ???        54   D
Payflex.50165e3e00130100       M, N K.              ???        77   D
Payflex.50165e4100130100       Visual Enterp Room   ???        18
pseudo.0003ba118477            ???                  ???        7
pseudo.0003ba3c0ff4            ???                  ???        9
pseudo.0003bad74bf2            ???                  ???        72
pseudo.00144fd35a84            ???                  ???        63
pseudo.0021284088ab            ???                  ???        13
pseudo.0021286dd890            ???                  ???        52
pseudo.0021286e19d3            ???                  ???        39
pseudo.002128909bd2            ???                  ???        71

Thanks
Steve

On 08/18/11 05:00, [email protected] wrote:
Send SunRay-Users mailing list submissions to
        [email protected]

To subscribe or unsubscribe via the World Wide Web, visit
        http://www.filibeto.org/mailman/listinfo/sunray-users
or, via email, send a message with subject or body 'help' to
        [email protected]

You can reach the person managing the list at
        [email protected]

When replying, please edit your Subject line so it is more specific
than "Re: Contents of SunRay-Users digest..."


Today's Topics:

   1. Re: Session sharing: Sun Ray to Sun Ray (David)
   2. A lot of Xnewt running (Steven Gelsie)
   3. Re: EXT :Re: Third-Party installation package for SRS 5.2 on
      Ubuntu 11.04 available (Phares, Scott (IS))
   4. Re: A lot of Xnewt running (Craig Bender)
   5. Re: Session sharing: Sun Ray to Sun Ray (Niki W. Waibel)
   6. Re: Problems installing SRSS 5.2 on Fedora 15 (J?rg Barfurth)


--- -------------------------------------------------------------------

Message: 1
Date: Wed, 17 Aug 2011 09:09:29 -0400
From: David<[email protected]>
To: SunRay-Users mailing list<[email protected]>
Subject: Re: [SunRay-Users] Session sharing: Sun Ray to Sun Ray
Message-ID:
<CA+Kjnc6B2E=rO9EWjKgYk49it1hHJC6MYc_Df=ebtxuf1+7...@mail.gmail.com >
Content-Type: text/plain; charset=ISO-8859-1

On Wed, Aug 17, 2011 at 3:29 AM, Niki W. Waibel<[email protected]> wrote:
The user who wants to temp share his/her desktop shall execute x11vnc. All users who want to view that desktop run a vnc viewer.

The ray protocol itself can't do this kind of sharing (as of now).

So then the secondary user would have to run a VNC viewer in his/her
SunRay virtual desktop. Correct?

Seems like it might suffer from performance issues due to the two
level remoting.  UI responsiveness is important to my users, and this
is why I'm looking at the Sun Ray solution in the first place.

I've not been able to use SRSS or SunRay yet, so maybe this is a silly
question, but:  Since I'm only planning to use *nix desktop sessions,
can the SRSS simply connect the secondary SunRay to the same X server
instance/display as the primary user?

Thank you


------------------------------

Message: 2
Date: Wed, 17 Aug 2011 10:22:08 -0400
From: Steven Gelsie<[email protected]>
To: [email protected]
Subject: [SunRay-Users] A lot of Xnewt running
Message-ID:<[email protected]>
Content-Type: text/plain; charset="iso-8859-1"; Format="flowed"

  We have a lot (32) of "/opt/SUNWut/lib/Xnewt :13 -nobanner -auth
/var/dt/A:13-WXaWfc -br +bs -terminate" processes running as root that are started by dtlogin. At first I thought was because we are running
in kiosk mode and we have 25 kiosk users but there more then 25 Xnewt
run as root.  Does anybody know why there are so many Xnewt process
running and what determines this number ?  I did kill them but they
restart as expected . We are running Sunrary Server 5.2 on Solaris 10.

Thanks
Steve

--
Steven Gelsie                   Email: [email protected]
Johns Hopkins Univ./APL         Phone: 240-228-4081   DC
11100 Johns Hopkins Rd                 443-778-4081   Baltimore
Laurel, MD  20723-6099          FAX:   240-228-6119
Work Schedule : Mon-Thurs  8:30AM-6PM
                Friday     Work At Home


-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 2490 bytes
Desc: S/MIME Cryptographic Signature
URL:<http://www.filibeto.org/pipermail/sunray-users/attachments/20110817/e52f397b/attachment-0001.bin >

------------------------------

Message: 3
Date: Wed, 17 Aug 2011 14:38:19 +0000
From: "Phares, Scott (IS)"<[email protected]>
To: SunRay-Users mailing list<[email protected]>
Subject: Re: [SunRay-Users] EXT :Re: Third-Party installation package
        for SRS 5.2 on Ubuntu 11.04 available
Message-ID:
<[email protected] >
Content-Type: text/plain; charset="us-ascii"

FYI...I trying this on Ubuntu 11.04 (i386) on a Virtual machine. I see that you have provided 1 piece in i386 format. The others are amd64 format.

I believe that your script will help greatly in installing SRS on Ubuntu and I appreciate the effort you have put forth so far.

Scott

-----Original Message-----
From: [email protected] [mailto:sunray-users- [email protected]] On Behalf Of Jens Langner
Sent: Tuesday, August 16, 2011 5:16 PM
To: SunRay-Users mailing list
Subject: Re: [SunRay-Users] EXT :Re: Third-Party installation package for SRS 5.2 on Ubuntu 11.04 available

Hi Dave,

thanks for the report and sorry that the script didn't work as expected. However, please note that this script/archive is meant to be a constant
work-in-progress. I have prepared it here with the servers we have
running now for quite a while. Thus, I haven't had the time to try my
installation script with a plain ubuntu 11.04 installation yet. So it
might need some additional fine tuning to really get usable for every
ubuntu installation. So if you want to take some time in modifying it
please feel free to submit your changes to the script back to me so that I can constantly update it and post a new version to the sunray wiki. I
will do the same as soon as I found the time to test it again with a
fresh ubuntu installation. In addition, I have currently only tested it
with an amd64 installation, thus the i386 installation might not be
perfect as well.

However, I still hoep that my archive/script is quite valueable for some of the ubuntu people in here because it contains the kernel patches to
allow to use it with kernels 2.6.38+ as well as having the latest
patches to get SRS 5.2.1 in principle running on an ubuntu installation.

So please stay tuned for more additional fine tuning of my script.

best regards,
jens

Dave McGuire schrieb:
On 08/16/2011 05:45 PM, Phares, Scott (IS) wrote:
I am having problems getting this installation to work. It appears
from your message, that this is a real straight forward installation,
however, I getting multiple installation errors when executing the
srs-installation.sh file. Has anyone else seen installation issues
following this thread? Is this working perfectly for anyone else?
I've not yet managed to make it work either. The first problem I hit was the script trying to use apt-get to install packages "sun-java6- jre" and
"libmotif3", which aren't available in the repository.

-Dave

--
Dr. Jens Langner                                     Ph: +49-351-2602757
Helmholtz-Zentrum Dresden-Rossendorf
Institute of Radiopharmacy - PET Center                [email protected]
Germany                                              http://www.hzdr.de/


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


------------------------------

Message: 4
Date: Wed, 17 Aug 2011 08:50:35 -0700
From: Craig Bender<[email protected]>
To: [email protected]
Subject: Re: [SunRay-Users] A lot of Xnewt running
Message-ID:<[email protected]>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

Hi Steve,
So you are running kiosk? What's your current policy? Any chance you
are running regular sessions for non-card and kiosk for card?

The output of utsession -p would help me tell you exactly what is
happening here. I'd expect quite a few of these to be have a Unix ID of
"???" and in a state of "DI".

Around the third patch of 4.1, we implemented a feature to help load
balancing in non-card kiosk sessions.  What used to happen that is a
cold restart was done, all those session would go to first server up.
This resulted in really poor load balancing.

There are a many factors that go into Sun Ray load balancing algorithm,
one of which is sessions, which of course require an Xserver.  The
enhancement to assist non-card kiosk scenarios was to keep greeter
sessions around for a certain amount of time, even if the actual user
session was redirected somewhere else.  This helps load balance the
DTU's across the host group much better than was possible before in the
even that NSCM was not being used (NSCM load balances just like smart
card sessions, however, NSCM doesn't really fit with Kiosk mode).

The amount of time these greeter sessions will hang around before
getting reaped is 900 seconds (15 minutes)

You can customize this by copying /etc/opt/SUNWut/ reaper.conf.template
to /etc/opt/SUNWut/reaper.conf and changing the line that read
REAPER_TIMEOUT=900 to what makes you comfortable.

On 8/17/11 7:22 AM, Steven Gelsie wrote:
We have a lot (32) of "/opt/SUNWut/lib/Xnewt :13 -nobanner -auth
/var/dt/A:13-WXaWfc -br +bs -terminate" processes running as root that are started by dtlogin. At first I thought was because we are running in kiosk mode and we have 25 kiosk users but there more then 25 Xnewt run as root. Does anybody know why there are so many Xnewt process running and what determines this number ? I did kill them but they restart as
expected . We are running Sunrary Server 5.2 on Solaris 10.

Thanks
Steve



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

------------------------------

Message: 5
Date: Thu, 18 Aug 2011 01:48:10 +0200
From: "Niki W. Waibel"<[email protected]>
To: "SunRay-Users mailing list"<[email protected]>
Subject: Re: [SunRay-Users] Session sharing: Sun Ray to Sun Ray
Message-ID:<[email protected]>
Content-Type: text/plain; charset="utf-8"

----- Urspr?ngliche Nachricht -----
Von: David
Gesendet: 17.08.11 15:09 Uhr
An: SunRay-Users mailing list
Betreff: Re: [SunRay-Users] Session sharing: Sun Ray to Sun Ray

On Wed, Aug 17, 2011 at 3:29 AM, Niki W. Waibel<[email protected]> wrote:
The user who wants to temp share his/her desktop shall execute x11vnc. All users who want to view that desktop run a vnc viewer.

The ray protocol itself can't do this kind of sharing (as of now).

So then the secondary user would have to run a VNC viewer in his/her
SunRay virtual desktop. Correct?
yes. very correct. there could be even more viewers (ie think of a training session).

Seems like it might suffer from performance issues due to the two
level remoting.
that's not the case. all vnc traffic is often within the same server (there are many X servers on a sunray server). if you run a sunray cluster (sunray server group), the vnc traffic goes between those sunray servers only.

obviously the sunray clients update their displays also, which is traffic from the sunray server to the sunray client. but that's the regular case anyway.

UI responsiveness is important to my users, and this
is why I'm looking at the Sun Ray solution in the first place.
sunray technology IS the most reposive solution you can have. server->client traffic is going via UDP. all other solutions (i know) use TCP, which has its clear disadvantages in remote desktop environments.

I've not been able to use SRSS or SunRay yet,
you should be using it :-)

so maybe this is a silly
question, but: Since I'm only planning to use *nix desktop sessions,
can the SRSS simply connect the secondary SunRay to the same X server
instance/display as the primary user?
for each sunray session an X server (Xnewt) is started on a sunray server. this X server transmits it's framebuffer to a single sunray client only. the ALP (Appliance Link Protocol) is not (yet?) designed to transmit to 2 clients.

the clients tell the server that they have received (or maybe lost) traffic. the server would get confused (current implementation, as of my knowledge), if more then 1 client would send such replies.
... and what to do with more then 1 mouse pointer?
... and what to do with usb sticks in more then 1 sunray?
... and what to do with audio output/input with more then 1 sunray?
i am sure this could be solved, but currently you stick with x11vnc. performance is fine, as long as you don't do extensive 3D or video.

hope this helps,
niki


------------------------------

Message: 6
Date: Thu, 18 Aug 2011 10:42:05 +0200
From: J?rg Barfurth<[email protected]>
To: SunRay-Users mailing list<[email protected]>
Subject: Re: [SunRay-Users] Problems installing SRSS 5.2 on Fedora 15
Message-ID:<[email protected]>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

Rodney Sparapani schrieb:
Kaya Saman wrote:
Hi Kaya:

Fedora will definitely not work (we tried).  However, you should
check out Oracle Linux 6.1. It's not as bleeding edge as Fedora, but
it is more modern than RHEL 5.7.

Don't. See below.

I am guessing that SRSS will need to be updated and modified slightly in order to cope with the newer GDM in order to get things going..... :-(

First GDM will need to be modified to support multiple 'seats' again.
There are patches available in the GDM project, but they have not been
accepted into the main version by the (Red Hat) maintainer yet.

AFAIK no Linux distro currently ships GDM (>= 2.22) with these patches.
Solaris 11 Express has them.

SRSS has some modifications to work with the patched GDM, but needs
additional work to get this fully ready on any platform.

This modification however, may or may not happen depending on what
Oracle decide for their own Linux variant; but, if generic
compatibility wants be achieved this seems to be the needed path to take!

I can't comment on Oracle's plans for SRSS on OEL, but you may have
noticed that Oracle VDI just added OEL (5.6) as a supported platform in
its latest release.

I obviously can't say whether and when other Linux distributions will
pick up multi-seat patches for GDM (and ConsoleKit, but Red Hat
apparently is moving away from ConsoleKit again), so you may have to
build your own patched version of GDM as one prerequisite for installing
SRSS for some time to come.


I definitely need to check out Oracle Linux anyway so I may build a VM of both Solaris 11 and OL6.1 as learning both is going be incredibly
helpful.

We haven't gotten to the stage of installing on Oracle Linux 6.1
yet. But, Oracle support has assured us that it does work.

They should not have. No released version of SRSS is supported on OEL
6.x - and afaik they won't work at all. SRSS 5.2 is supported on OEL 5.6
only.

- J?rg

--
J?rg Barfurth                     http://blogs.oracle.com/joergb

Disclaimer: I am employed by Oracle. The statements and opinions
expressed here are my own and do not necessarily represent those
of Oracle Corporation.


------------------------------

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


End of SunRay-Users Digest, Vol 91, Issue 17
********************************************


--
Steven Gelsie                   Email: [email protected]
Johns Hopkins Univ./APL         Phone: 240-228-4081   DC
11100 Johns Hopkins Rd                 443-778-4081   Baltimore
Laurel, MD  20723-6099          FAX:   240-228-6119
Work Schedule : Mon-Thurs  8:30AM-6PM
       Friday     Work At Home

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

Reply via email to