Announcement

Collapse
No announcement yet.

Alerting after HS4 reboot

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    Alerting after HS4 reboot

    For the most part SDJ-Health has been working fine, after I learned how to "tune" the poll settings for the different types of battery devices I have, which all report battery life on different intervals.

    But one issue that plagues me is that it seems like every time HS4 is rebooted or down for a brief time, the SDJ-Health plugin will start to alert on devices missing a poll even though HS4 and the plugin have only been down for like 5 minutes or less. It's almost like the PI loses its place in terms of the current state of all monitored devices and takes a while for it to sync up again.

    Is there some kind of config setting that I need to set? What else could I possibly be doing wrong?

    #2
    Originally posted by TC1 View Post
    For the most part SDJ-Health has been working fine, after I learned how to "tune" the poll settings for the different types of battery devices I have, which all report battery life on different intervals.

    But one issue that plagues me is that it seems like every time HS4 is rebooted or down for a brief time, the SDJ-Health plugin will start to alert on devices missing a poll even though HS4 and the plugin have only been down for like 5 minutes or less. It's almost like the PI loses its place in terms of the current state of all monitored devices and takes a while for it to sync up again.

    Is there some kind of config setting that I need to set? What else could I possibly be doing wrong?
    Hi

    That isn't something I have experienced or had reported with polled devices, so I will need a little more information to see if I can pin down the issue.

    Most of my battery devices are non-listening, so the plug-in monitors them by wake-up not by polling. I do have 2 z-wave listing devices though, an Everspring siren and a Spirit valve, which the plug-in in monitors by polling. My HS4 is mostly up 24/7 but there are occasional reboots for Windows updates or when I update HS4 itself. On rare occasions a device might wake-up during that reboot and HS4 will obviously miss that wake-up. Shortly after I will get a notification from the plug-in of a missed wake-up, but as I have also had a notification of an HS4 reboot I just ignore it and the next time the device wakes-up it notifies everything is ok. I haven't noticed this with devices monitored by polling. If desired the odd missed wake-up can be ignored by configuring the settings to ignore 1 or more missed wake-ups before alerting.

    What devices are you monitoring?
    Can you explain a bit more about what you mean by 'tune the poll settings'?
    Can you post a screenshot of your Battery Device Monitoring Settings page?
    If you have overridden the settings for individual devices can you post a screen shot of the SDJ-Health tab Settings for a typical device that you have the issue with?

    Steve
    ​​​​​​

    Comment


      #3
      Originally posted by TC1 View Post
      For the most part SDJ-Health has been working fine, after I learned how to "tune" the poll settings for the different types of battery devices I have, which all report battery life on different intervals.

      But one issue that plagues me is that it seems like every time HS4 is rebooted or down for a brief time, the SDJ-Health plugin will start to alert on devices missing a poll even though HS4 and the plugin have only been down for like 5 minutes or less. It's almost like the PI loses its place in terms of the current state of all monitored devices and takes a while for it to sync up again.

      Is there some kind of config setting that I need to set? What else could I possibly be doing wrong?
      Originally posted by SteveMSJ View Post

      That isn't something I have experienced or had reported with polled devices, so I will need a little more information to see if I can pin down the issue.

      Most of my battery devices are non-listening, so the plug-in monitors them by wake-up not by polling. I do have 2 z-wave listing devices though, an Everspring siren and a Spirit valve, which the plug-in in monitors by polling. My HS4 is mostly up 24/7 but there are occasional reboots for Windows updates or when I update HS4 itself. On rare occasions a device might wake-up during that reboot and HS4 will obviously miss that wake-up. Shortly after I will get a notification from the plug-in of a missed wake-up, but as I have also had a notification of an HS4 reboot I just ignore it and the next time the device wakes-up it notifies everything is ok. I haven't noticed this with devices monitored by polling. If desired the odd missed wake-up can be ignored by configuring the settings to ignore 1 or more missed wake-ups before alerting.

      What devices are you monitoring?
      Can you explain a bit more about what you mean by 'tune the poll settings'?
      Can you post a screenshot of your Battery Device Monitoring Settings page?
      If you have overridden the settings for individual devices can you post a screen shot of the SDJ-Health tab Settings for a typical device that you have the issue with?

      Steve
      ​​​​​​
      TC1 I haven't heard back from you on this so I don't know whether it resolved or not.

      One thing I thought of was that if your Z-Wave interface takes a while to initialize after HS4 is started, the first poll of a battery device may fail. However, if you have more than one battery device being monitored by polling it would only effect one of them as the polls are already spread out over the polling interval. So, it probably is not the cause of what you describe.

      Anyway, to allow a bit more time for Z-Wave interfaces to initialize I have added a 60 second delay after the PI starts before any battery polling commences. This delay can also be adjusted, if necessary, by adding a parameter with a delay value in seconds to the INI file. This is in version 3.1.0.2 which I have added to the Beta section of the updater and the optional delay parameter is noted in 4.2.1(3) of the updated guide.

      This change only effects Battery devices that are monitored by polling and in general won't be noticed.

      Steve

      Comment


        #4
        Hi Steve,

        I haven't responded back because I wanted to more scientific in my next response, ie, run some controlled experiments and collect some data rather than convey my anecdotal experiences.

        Also note that the majority of my battery devices are not Z-wave, but mostly Zigbee. I do a cyclic monitor on them, they either report activity (temp, motion, humidity, etc) and/or battery levels within the period I have setup.

        Unfortunately, other issues have my attention right now, but running some experiments regarding this is on my to-do list.

        Thanks for checking in

        Comment


          #5
          Originally posted by TC1 View Post
          Hi Steve,

          I haven't responded back because I wanted to more scientific in my next response, ie, run some controlled experiments and collect some data rather than convey my anecdotal experiences.

          Also note that the majority of my battery devices are not Z-wave, but mostly Zigbee. I do a cyclic monitor on them, they either report activity (temp, motion, humidity, etc) and/or battery levels within the period I have setup.

          Unfortunately, other issues have my attention right now, but running some experiments regarding this is on my to-do list.

          Thanks for checking in
          No problem, whenever you get round to it

          Comment

          Working...
          X