Ankündigung

Einklappen
Keine Ankündigung bisher.

- √ - HomeServer very slow

Einklappen
X
 
  • Filter
  • Zeit
  • Anzeigen
Alles löschen
neue Beiträge

  • Gaston
    antwortet
    Zitat von Warichet Beitrag anzeigen
    I think I got it, indeed I made a loop on sheet "Login analyze" by re-using the "OK" internal variable
    Indeed. Happy to hear you got it.

    BTW: Nice paradoxal phrase on your guest login page

    Kind Regards,
    Gaston

    Einen Kommentar schreiben:


  • Warichet
    antwortet
    Zitat von MatthiasS Beitrag anzeigen
    Because this "Dummy" makes it difficult to analyse your project.
    The dummy is sitting on my chair
    I think I got it, indeed I made a loop on sheet "Login analyze" by re-using the "OK" internal variable

    The SETWERT_WORK keeps quite now (me too )

    A big thank you for all who took their time and effort to help me.

    Einen Kommentar schreiben:


  • MatthiasS
    antwortet
    Yes - I wish . Because this "Dummy" makes it difficult to analyse your project.

    Einen Kommentar schreiben:


  • Warichet
    antwortet
    Zitat von MatthiasS Beitrag anzeigen
    I see a large number of error messages according to logic modules 13051, 13052, 13057, 13062 ....
    Yes, indeed, in an attempt to diagnose my problem, I've a bit "massacrated" my project.
    I thought the culprit could be one of my own modules.
    So, I created a very short dummy module, which I renamed and copied in place of the original ones.
    If one of the modules was the culprit, the problem should have gone, it's not the case.
    My conclusion: the problem must be in a logic sheet, not logic module.

    But I can put everything back in place if you wish.

    Einen Kommentar schreiben:


  • MatthiasS
    antwortet
    @Raymond

    I see a large number of error messages according to logic modules 13051, 13052, 13057, 13062 ....

    I am surprised:

    All these logic modules in your project do contain the same lines!?

    Einen Kommentar schreiben:


  • Gaston
    antwortet
    Zitat von Warichet Beitrag anzeigen
    Well, I've already done a raw survey of my logic sheets. Chances are I've overlooked it. Same player shoot again
    Could you email me your project (using Archiving without passwords) ?

    gaston

    Einen Kommentar schreiben:


  • Warichet
    antwortet
    Zitat von Gaston Beitrag anzeigen
    Well, SETWERT_WORK ist the thread working through the SetWertQueue, which in your case is full which indicates a loop somewhere.
    OK, this seems to be clear for me now.

    Zitat von Gaston Beitrag anzeigen
    You should first look for logic sheets where loops exist (same KOs in output and input of sheet).
    Well, I've already done a raw survey of my logic sheets. Chances are I've overlooked it. Same player shoot again

    Thank you very much for the tips.

    Einen Kommentar schreiben:


  • Gaston
    antwortet
    Zitat von Warichet Beitrag anzeigen
    OK, let's forget about a logic problem.
    Any idea about who/what could use the SETWERT_WORK ?
    Well, SETWERT_WORK ist the thread working through the SetWertQueue, which in your case is full which indicates a loop somewhere.

    You should first look for logic sheets where loops exist (same KOs in output and input of sheet). Since the problem occurs almost directly after the boot you may simply disconnect the looped KO in the input and try if the problem is gone.

    Of course, and that might be mor difficult to track down, loops may also cover several sheets: Sheet1:Output->KO1->Sheet2:Input->Sheet2:Output->KO2->Sheet1Input for example.

    Happy digging ,
    Gaston

    Einen Kommentar schreiben:


  • Warichet
    antwortet
    Zitat von Michel Beitrag anzeigen
    If it would be a logic problem, the number would be seen in that section.
    OK, let's forget about a logic problem.
    Any idea about who/what could use the SETWERT_WORK ?

    126 SETWERT_WORK R 56756.91 154.34
    Zitat von Michel Beitrag anzeigen
    So if the problem still exists, try Gaston´s tip to restart.
    That would be to easy

    Zitat von Michel Beitrag anzeigen
    Another point may be any false configured "Zentraladressen", but this should be seen in the monitor of your ETS.
    As a HS newbee, you can expect the most blatant mistake on my side
    But I don't see anything abnormal on the bus, traffic is very limited and all manual commands seem to work.
    My conclusion so far: the problem is limited to the HS and it's configuration.
    It's difficult for me to troubleshoot this problem having limited knowledge of the HS and almost no diagnostic tools.

    Thank you for taking the time to look into my problem.

    Einen Kommentar schreiben:


  • HartmutB
    antwortet
    @Warichet:

    Does the problem occur, when you disconnect your HS from the KNX bus and restart it (at least the debug page can tell you something about the "Setwert-Queue") ?

    If yes: What about sending your entire project to Dacom for a check ?
    They should have the means to track it down, and if it is a flaw in the HS software, they should be very interested in it. If - on the other hand - it has been "only" a "broken" logic module, just promise them to promote the HS in France ...

    Regards
    Hartmut

    Einen Kommentar schreiben:


  • Warichet
    antwortet
    Zitat von Gaston Beitrag anzeigen
    have you ever tried to restart the HS.
    Yes, several times

    Zitat von Gaston Beitrag anzeigen
    Did the problem still show up after the reboot.
    Yes, almost immediately

    Zitat von Gaston Beitrag anzeigen
    if you did not change anything
    Well, I don't dare to say this. I'm continuously making rather small changes.
    I must admit I haven't analyzed the problem correctly from the beginning. I supposed a punctual network problem.

    Zitat von Gaston Beitrag anzeigen
    The source of the problem is usually a wrong understanding of how a logic module exactly works or of course could be an own module not reacting as expected.
    Well, probably my case

    Zitat von Gaston Beitrag anzeigen
    since it results that the logic drawn does not behave like the drawing shows it should. You may have a given logic sheet, delete two connections and reconnect them using a different order for connecting then you used the previous time and the result of the logic may dramatically change.
    ûûûh
    This might be the case here.
    But I'm unable to trace it back to the origin

    Thank you for the hints.

    Einen Kommentar schreiben:


  • Michel
    antwortet
    Hmm, nothing special to see. If it would be a logic problem, the number would be seen in that section.
    So if the problem still exists, try Gaston´s tip to restart.

    Another point may be any false configured "Zentraladressen", but this should be seen in the monitor of your ETS.

    Einen Kommentar schreiben:


  • Warichet
    antwortet
    Zitat von MatthiasS Beitrag anzeigen
    Debug-page, section Wait-List.
    Here it is, but don't see what you can get from this info.

    Thank you very much for taking care of my misery
    Angehängte Dateien

    Einen Kommentar schreiben:


  • Gaston
    antwortet
    Raymond, have you ever tried to restart the HS. Did the problem still show up after the reboot. if you did not change anything I would expect it to disappear making it difficult to track down via loading new projects.

    Although the problem may be linked to an own logic thsi is not sure. The problem is basically outside of any logic module. The source of the problem is usually a wrong understanding of how a logic module exactly works or of course could be an own module not reacting as expected.

    Another source of this problem is that the implementation of the logic in the HS has not been implemented as logics are usually implemented. Which in my eyes is the only serious drawback for the HS at this time since it results that the logic drawn does not behave like the drawing shows it should. You may have a given logic sheet, delete two connections and reconnect them using a different order for connecting then you used the previous time and the result of the logic may dramatically change.

    Kind regards,
    Gaston

    Einen Kommentar schreiben:


  • MatthiasS
    antwortet
    @Raymond

    Debug-page, section Wait-List.

    Let's have a look!

    Einen Kommentar schreiben:

Lädt...
X