S10 disconnecting sessions

  • This topic is empty.
Viewing 15 posts - 76 through 90 (of 104 total)
  • Author
    Posts
  • #15736
    _David_
    Member
    • Total Post: 3
    • Newbie

    WTOS is at 6.2.0_08

    #15737
    ConfGen
    Keymaster
    • Total Post: 11485
    • Jedi Master
    • ★★★★★★★

    David,

    can you try 6.2.036 in your environment?

    CG

    #15738
    _David_
    Member
    • Total Post: 3
    • Newbie

    Did not fix the issue.

    Issue is definitely (somehow) related to the Sonicwall.

    TZ210 = Screaming users.
    TZ170 = Not a peep.

    #17472
    Nickmancini
    Member
    • Total Post: 9
    • Regular Joe
    • ★★

    I see this topic has gone a bit dormant without a resolution. Wyse tech support has been no help.

    We’re having the same issue – Users get bounced from their Winterm and a “Trap: Protection Fault, EIP=blablabla” error appears on the screen requesting a winterm reboot. (the blablabla can vary)

    This is happening on S10 & V10L (WTNOS) with t any firmware version we throw at it (currently installed firmware is 6.5.23).

    We have determined that this issue can be recreated when users hit specific websites. Recreation is not always 100% accurate, but sooner or later it will bomb.

    Our users use RDP connections to Virtual Windows-7 machines. Not using a Connectiomn Broker at this time, thanks to the poor-man’s connection broker instructions posted here. Static IP Vs. DHCP makes no difference.

    Website that are sure to get the Trap: Protection Fault error:

    Go to http://www.myearthhour.org

    click on “California” then click on “Read more about what is happening in California” at the bottom of the pop-up.

    As soon as the last click is made, the winterm bombs.

    Hope this sheds some light on this issue. we’re not happy with Wyse right now and may be looking for another vendor if this does not get solved – quickly!

    Thanks!

    #17475
    nklykken
    Member
    • Total Post: 22
    • Regular Joe
    • ★★

    @Nickmancini wrote:

    I see this topic has gone a bit dormant without a resolution. Wyse tech support has been no help.

    We’re having the same issue – Users get bounced from their Winterm and a “Trap: Protection Fault, EIP=blablabla” error appears on the screen requesting a winterm reboot. (the blablabla can vary)

    This is happening on S10 & V10L (WTNOS) with t any firmware version we throw at it (currently installed firmware is 6.5.23).

    We have determined that this issue can be recreated when users hit specific websites. Recreation is not always 100% accurate, but sooner or later it will bomb.

    Our users use RDP connections to Virtual Windows-7 machines. Not using a Connectiomn Broker at this time, thanks to the poor-man’s connection broker instructions posted here. Static IP Vs. DHCP makes no difference.

    Website that are sure to get the Trap: Protection Fault error:

    Go to http://www.myearthhour.org

    click on “California” then click on “Read more about what is happening in California” at the bottom of the pop-up.

    As soon as the last click is made, the winterm bombs.

    Hope this sheds some light on this issue. we’re not happy with Wyse right now and may be looking for another vendor if this does not get solved – quickly!

    Thanks!

    I can understand your frustration Nickmancini 😕

    Just to let you know, I’ve tried the website you posted with the accurate instructions on my V10L WNOS 6.3.0_27.01 without getting the problems you describe. Unlike you, I accessed the site through a Citrix connection though..

    One question comes to mind though, do you use an encrypted RDP connection or just plain?

    The issue we had when we got the “Trap protection fault” was changes made with the ICA encryption in a Citrix update. We also had challenges with the Citrix Session Reliability feature in conjunction with our wyse V10L boxes, but that’s most likely not a problem of yours 😛

    One other common thing I’ve seen over the past year is issues related with thin clients and webpages with flash animations. (like the one you linked to) One thing I’ve especially seen is bandwidth use when browsing to flash animated sites (like http://www.vg.no), one user can in worst case scenario use about 3-4 Mbit/s just by keeping this site up on the screen.

    What would also help of course, was if you posted your WNOS.INI file. Just so any misconfiguration can be ruled out.

    Regards

    #17477
    Nickmancini
    Member
    • Total Post: 9
    • Regular Joe
    • ★★

    Thanks for the reply.

    We’re getting this at a few clients who use their systems in entirely different ways.

    The website I outlined in my previous post will eventually bomb the S10 or V10L. Just not all the time. There are more sites too, and as I collect them I will post them.

    We only use RDP, Not ICA.

    Problem manifests itself whether Winterm is connected to a 2008 TS or a Windows-7 VM.

    After the issue, there are times when a reboot of the WT will not reconnect to the host, and a reboot of the host is necessary to kill off the webpage that is causing the issue.

    At both clients, we have broadband connections in excess of 20 mbps with only a handful of users. Plenty to go around.

    LAN infrastructure is first rate: Cat5e cabling; Dell 5000 series gigabit switches. Dell T710 server with quad NICs, 48 GB RAM, 15k rpm disks, RAID-6 running vSphere (ESX 4), etc., etc. Horsepower is not an issue.

    WNOS.INI is here:

    DelCertificate=all
    ClearLicense=Yes
    FixLicense=Clean
    AltCacheDisable=yes
    EnableCacheIni=no

    signon=NTLM
    DisableGuest=yes
    lastusername=yes

    DomainList=corp

    DisableDomain=no

    AutoLoad=1

    language=us

    Autopower=yes

    FormURL=biglogo.bmp

    Priviledge=none
    LockDown=no
    HideSysInfo=yes
    HidePPP=yes
    HidePN=yes
    HideConnectionManager=yes
    EnableNetworkTest=no
    EnableTrace=no
    ShowDisplaySettings=no
    EnableKeyboardMouseSettings=no
    KeepDHCPRequestIP=yes
    SuppressTaskBar=yes

    adminmode=no

    SessionConfig=ALL
    unmapprinters=no
    unmapserials=no
    smartcards=yes
    mapdisks=yes
    disablesound=no
    unmapusb=no
    DisksReadOnly=no
    WyseVDA=yes

    Device=audio
    volume=high

    TimeZone=’GMT – 08:00′
    ManualOverride=yes
    Daylight=yes
    Start=030207
    End=110107
    TimeZoneName=PST
    DayLightName=PDT

    TimeServer=192.168.1.10
    TimeFormat=”12-Hour Format”
    DateFormat=”mm/dd/yyyy”

    VNCPrompt=no
    ViewOnly=no
    ActiveVisible=no

    AutoSignoff=yes
    Shutdown=no

    ScreenSaver=30
    LockTerminal=0
    Type=2
    Image=biglogo.bmp

    NoticeFile=Notice.txt
    Resizable=yes
    Timeout=360

    Community=public

    EnableGKey=yes

    KeySequence=yes
    ctrl+alt+del=no
    ctrl+alt+up=no
    ctrl+alt+down=no
    ctrl+alt+left=no
    ctrl+alt+right=no

    Shutdowninfo=no

    DisableButton=no

    Inactive=0

    TCXLicense=xxxxx-xxxxx-xxxxx-xxxxx-xxxxx

    Include=$MAC.ini

    $MAC.INI is here:

    #Mac.ini file
    Terminalname=Terminal01
    Resolution=1440×900 Refresh=60

    Printer=LPT1
    Name=”Jeanine’s Printer”
    PrinterID=”HP Deskjet F4400 series”
    Class=PCL5
    Enabled=yes
    EnableLPD=no

    USER.INI is here:

    connect=rdp
    description=”Connecting…”
    host=jeanine-vm
    icon=smalllogo.bmp
    UserName=$UN
    Password=$PW
    domainname=$DN
    Colors=true
    Fullscreen=yes
    autoconnect=1
    LocalCopy=No
    KeepAlive=10
    Lowband=no
    Mapdisks=yes
    NoReducer=no
    Rdp_No_Animation=no
    Rdp_No_Dragging=no
    Rdp_No_Fontsmoothing=no
    Rdp_No_Theme=no
    Rdp_No_Wallpaper=no
    Reconnect=no

    #17483
    Nickmancini
    Member
    • Total Post: 9
    • Regular Joe
    • ★★

    Quick update: Wyse is recommending FW 6.5.0_27 which supposedly fixes many trap Protection Fault errors.

    Keeping my fingers crossed and will advise of the results.

    #17580
    Nickmancini
    Member
    • Total Post: 9
    • Regular Joe
    • ★★

    Here’s the latest:

    Now running 6.5.29 on both S10 & V10L.

    We continue getting TPF errors on both units, but it’s isolated to Win7 VMs and Windows Terminal Server 2008. We can recreate the TPF about 50% of the time.

    When connecting via RDP to XP or TS 2003, we cannot recreate the TPF.

    This appears to be an issue isolated to Winterms connected via RDP to Win7 VMs.

    Tests have only been run using Win7 x64 VMs.

    Additional tests will be performed using Win7 x86 in a few days and I’ll post the results.

    In the meantime, I have a pretty unhappy customer 👿 .

    #17743
    MHzBurglar
    Member
    • Total Post: 10
    • Regular Joe
    • ★★

    My environment has several S30 terminals which have been converted to S10s running WTOS 6.3.0_12 and I have been getting random disconnects from ICA sessions and the “bad file number” error as well. After a user receives this error, the Winterm will not launch any new ICA sessions, and will only spit out the same “bad file number” error if a user attempts to do so.

    Rebooting the Winterm allows the user to connect to Citrix again, but their previous session will usually be lost as we have several servers and the load balancer will usually launch a new session for the user on a different server.

    The “bad file number” disconnects seem to mostly occur at specific times of the day when network traffic is heavy due to the “shift change” of our production staff (which in turn means mass log-offs and log-ons during a relatively short timeframe.)

    I have tried disabling session reliability, but I have found that while it appears to get rid of the “bad file number” disconnects, random disconnects still occur and sometimes more frequently. The only difference is that with session reliability disabled, the disconnects become completely random; they can occur at any time of day on any WTOS Winterm with no distinctive pattern and they do not give any error message at all upon disconnect.

    My only theory for this is that the WTOS ICA client may be less tolerant of network “hiccups” than the Windows CE and Windows XP ICA clients, and by disabling session reliability, the ICA session drops much more easily.

    It seems to be a catch-22, really. No matter what I do, I get complaints from several users about random disconnects. I’m going to see if I can get WTOS 6.5 from Wyse and hope that whatever is causing this problem is fixed in the latest release.

    I’ll update this thread with my findings once I get the new version.

    #17836
    mcsa_ccna
    Member
    • Total Post: 1
    • Newbie

    Hello, We have the same issue here at hsbc. im a network admin, We have wyse s10 terminals across several sites running RDP sessions to windows server 2003 sp2 which itself is running windows terminal services, license server, and the dhcp component etc. and are too having random disconnecting issues. the local sites have a 1gb infrastructure.

    we know event veiwer doesnt show anything, the network monitor shows no irregularities, all firmware updates – still dont work,,,, if u check terminal server, session manager -rdp connection properties u will see that the wyse box thats having the issue is constantly forming and dropping rdp (tcp/ip)sessions, so u might find several rdp sessions being created. After many nice long conversations with wyse, they still cant fix it, or reproduce the effect.

    updated several wyse to rdp cleint 6.0 – still random disconnects
    hacked the registry settings for terminal server, tcp/ip stack ,Rdp application layer – waiting for test results (maybe 1 week observation)

    2 out of the 10 sites are due to be upgraded to windows server 2008
    so will test the wyse on these sites

    * we believe that the issue could be an unstable tcp/ip connection, cauing issues from transport layer-through to session and presentation layer protocols*

    I think in the end, will just replace with fujitsu or hp thin clients

    #17840
    ConfGen
    Keymaster
    • Total Post: 11485
    • Jedi Master
    • ★★★★★★★

    Can you cross check with a PC? Is this working constantly?
    Which firmware have ou tested on your S10s?

    CG

    #17846
    Nickmancini
    Member
    • Total Post: 9
    • Regular Joe
    • ★★

    We have gone as far as setting up an exact replica of our client’s infrastructure and allowed Wyse to access it 24/7.

    They claim to only have suffered 1 TPF during their tests on our systems.

    We can get it to bomb like clockwork, with more and more websites added to the list of sites that cause the TPF.

    At this point, we have no choice but to move on with non-wyse thin clients. This issue has 3 of our most important clients furious.

    FYI – we’re using 6.5.0_30 on S10 & V10L.

    (anybody want to buy a bunch of S10s and V10Ls, cheap?)

    #17852
    ConfGen
    Keymaster
    • Total Post: 11485
    • Jedi Master
    • ★★★★★★★

    Check out 6.5.0_30.10. They have fixed a lot of stuff in there.

    CG

    #17890
    awelker
    Member
    • Total Post: 17
    • Regular Joe
    • ★★

    I’m starting to see this issue on several v10l’s running 6.5.0_30
    Its only reported in the training lab, and it happens to be the only area using the TCX suite of software.

    #18106
    damien65
    Member
    • Total Post: 3
    • Newbie

    Running V10Ls here.

    Have roughly 600 of these deployed.

    Upgraded to Firmware 6.50_30 when it came out, roughly a month ago.

    Am receiving reports that these devices are disconnecting.
    No Trace errors reported or seen by me.

    Upon inspecting them myself, I found the devices are locally responsive, but as far is the network is concerned, have fallen off.

    Remote pings fail 49 times out of 50. Pings and name resolution from the device both fail. The jacks are fine as plugging in a different device yields a workable network connection.

    Devices are DHCP, set to auto. Test laptop and TC both negotiate at 100Full.

    There is no pattern to the drops that I can identify.

    Rebooting the terminal, it comes up cleanly and correctly.
    Gets IP, connects to FTP, configures, launches sessions, without issue.

    I am configuring the trace option on some today to collect ica trace logs.

    I will also soon test Static IPs.

    These devices have been in the field for over a year or so, I have to assume this is something with the new firmware.

    Anyone else tried pinging the client when a drop is reported?
    Perhaps the sessions are not the issue at all, and the local device network stack is having problems.

Viewing 15 posts - 76 through 90 (of 104 total)
  • You must be logged in to reply to this topic.