Report isn't getting run

[mlevin77]

Hi, I'm a first time user of SDJ-Health. I'm working through the manual, trying to monitor my non-battery devices (specifically I need to know if any of my sensors have not been heard from in a while - once a month or two, some lock up and need to be unplugged/replugged into the outlet).

I'm following the example in the Guide of how to send me an email with a report of which ones have failed, but I don't see the device that's shown in the example (SDJ-Health-General Root). What I see is:

which of these should I be triggering off of? And, what do I check for - what does it want for value in this menu (assuming that's even correct):



What I'd love is an email with the list of devices that hadn't been heard from in 24 hours. Thank you!

Current Date/Time: 12/22/2020 2:35:31 PM
HomeSeer Version: HS3 Pro Edition 3.0.0.548
Operating System: Microsoft Windows 7 Professional - Work Station
System Uptime: 74 Days 18 Hours 34 Minutes 52 Seconds
IP Address: 10.0.1.31
Number of Devices: 200
Number of Events: 151
Available Threads: 400
HSTouch Enabled: True
Event Threads: 0
Event Trigger Eval Queue: 0
Event Trigger Priority Eval Queue: 0
Device Exec Queue: 0
HSTouch Event Queue: 0
Email Send Queue: 0
Anti Virus Installed: Microsoft Security Essentials
In Virtual Machine: No MFG: acer
Enabled Plug-Ins
3.1.0.3: Device History
3.0.0.71: EasyTrigger
3.0.0.14: NetCAM
3.0.8.7: SDJ-Health
3.0.6644.26753: UltraLog3
3.0.2.0: Z-Wave

[SteveMSJ]

Hi, I'm a first time user of SDJ-Health. I'm working through the manual, trying to monitor my non-battery devices (specifically I need to know if any of my sensors have not been heard from in a while - once a month or two, some lock up and need to be unplugged/replugged into the outlet).

I'm following the example in the Guide of how to send me an email with a report of which ones have failed, but I don't see the device that's shown in the example (SDJ-Health-General Root). What I see is:

which of these should I be triggering off of? And, what do I check for - what does it want for value in this menu (assuming that's even correct):



What I'd love is an email with the list of devices that hadn't been heard from in 24 hours. Thank you!

Current Date/Time: 12/22/2020 2:35:31 PM
HomeSeer Version: HS3 Pro Edition 3.0.0.548
Operating System: Microsoft Windows 7 Professional - Work Station
System Uptime: 74 Days 18 Hours 34 Minutes 52 Seconds
IP Address: 10.0.1.31
Number of Devices: 200
Number of Events: 151
Available Threads: 400
HSTouch Enabled: True
Event Threads: 0
Event Trigger Eval Queue: 0
Event Trigger Priority Eval Queue: 0
Device Exec Queue: 0
HSTouch Event Queue: 0
Email Send Queue: 0
Anti Virus Installed: Microsoft Security Essentials
In Virtual Machine: No MFG: acer
Enabled Plug-Ins
3.1.0.3: Device History
3.0.0.71: EasyTrigger
3.0.0.14: NetCAM
3.0.8.7: SDJ-Health
3.0.6644.26753: UltraLog3
3.0.2.0: Z-Wave

Hi,

You will want to check at least one of the options, e.g. 'Check last change time'. Then try running a scan again.
If you still don't get anything can you set the LogLevel to 2, on the 'Plug-in Config' page, run a scan and post any SDJ-Health messages you get in the log.

Typically for the Z-Wave interface, I would check all 4 options, except 'hide...' and set polling interval to 2 seconds and poll depth to 3. However, it depends very much on your system and devices.

Let me know how you get on.

Steve

[SteveMSJ]

I think you edited your post whilst I was replying so my reply might make not much sense now. I am in the middle of a zoom meeting so I will have a look later when I have some time.

Steve

[mlevin77]

I think you edited your post whilst I was replying so my reply might make not much sense now. I am in the middle of a zoom meeting so I will have a look later when I have some time. Steve

Yeah sorry, I solved my original problem and decided to edit the post to ask the new issue I was having. Thank you!

[SteveMSJ]

Yeah sorry, I solved my original problem and decided to edit the post to ask the new issue I was having. Thank you!

OK, I was distracted and thought I would save time by replying to your post whilst in the middle of a zoom meeting. Not very sensible as I didn't event notice that the quote I inserted was different to the message I had read and was responding to

The 'General Devices' section of the plug-in is very different to the 'Battery Devices' section. It can be used in many ways with al sorts of different devices and interfaces so can be complicated to figure out.

You can configure it to only send you an email if there are some device failures and that can be made dependant on what the interface is that the devices belong to. In my case, amongst other things, I like to have a report sent every day whether or not there is a failure. So the example event in the guide triggers on the General-Health root device having it's value set or changed, with a condition less than 'Activity Scan in Progress' so it doesn't trigger until the scan is completed. If you wanted it to only trigger if there are some failures then you would add a condition that the value of the General-Health root device is greater than 'Healthy'. Basically the value of the Root device after each scan is the number of devices that failed with 0 being 'Healthy'.

The example uses the General-Health root device but you can use devices created for each interface on your system if you want a bit of finer control on the reporting.

You can choose from the 2 reports, Failure Report or Full Report, depending on what you want to see.

You can set the scans to run automatically at a certain time each day or use events to trigger scans at any time interval or other trigger you want to choose.

If you take Z-Wave devices as an example and set the options as mentioned in my previous reply as well as setting the Period of Inactivity to 24 hours, then when a scan is run, it will:

• Check any Z-Wave devices that haven't reported for 24 hours
• If any haven't it will poll them to see if they respond
• Check if any devices are in an Invalid State
• Report on what it finds

It will optionally do the same for devices through other interfaces on your system, although not all interfaces support polling.

I hope this helps but feel free to ask more questions.

Steve

[SteveMSJ]

An additional point. I also have icons on my HSTouch home screens tracking the SDJ-Health and SDJ-General-Health status graphics so I see that everything is healthy or not. This also reminds me to look at the reports.

For example:



Steve

[mlevin77]

Thanks Steve!! Couple of questions:

> If you wanted it to only trigger if there are some failures then you would add a condition that the value of the General-Health root device is greater than 'Healthy'. Basically the value of the Root device after each scan is the number of devices that failed with 0 being 'Healthy'.

Ok, so I only care about the Z-wave interface, so is this:

the right device - it doesn't say "root" - but "Z-wave SDJ-Health-General Z-wave (General Health)" - is the # of failed devices?

Assuming that's the right device I should be checking,

> Basically the value of the Root device after each scan is the number of devices that failed with 0 being 'Healthy'.

ok so I want to be notified if the # is above 0, but the interface is only letting me enter one number:

so if I enter a number, will it be "anything above this number" or "exactly this number"? The way it's worded, it makes me think if I say "1", it will only trigger if exactly one device is in failure, but not if 2,3, etc. How do I enter "1 or more"?

And lastly: if I only want this email once a day, can I just trigger it like "if time is 4am AND ..."? or is there a way to have it email me if a *new* device goes into failure? I don't want a million emails for a device that's in failure and I already know about it, so do I just have it run once a day or is there a better way?

[SteveMSJ]

There are several different ways of achieving what you are trying to do and it all depends on your particular preferences. If I first explain a bit about the way the pi works it might be a bit clearer. And, in case anybody else reads this, we are only talking about the General Devices section of the plug-in here, the Battery Devices section works very differently. Battery and General Device features are really 2 plug-ins within a single plug-in.

The plug-in identifies all the Interfaces (usually plug-ins) on your system and provides you with a set of options for each so you can choose how the devices owned by each Interface are scanned and reported on. It also has a group for all devices not owned by Interfaces, i.e. virtual devices that you might create. You can choose to ignore any of the Interfaces.

A Parent Root device is created 'General Health Root' which is used to control scans, alerts and reporting on all the interfaces. You can also choose to have child devices created for any of the individual Interfaces. This isn't necessary but the child devices can be used to control scans and alerts for individual interfaces.

A scan of the system can be set to run automatically at a specific time each day. The options set for the various Interfaces will determine what devices are scanned and how they are scanned. You can also trigger scans manually or from events so they can be run whenever you want and as often as you want. If you have selected to have a child device created for an individual interface then you can trigger scans of the devices on that Interface alone if you want.

Whilst scanning, the value (status) of the Parent device, and child devices if they exist, is changed to reflect the progress of the scan. Some of this may be almost instantaneous. These progress status's have values starting at 10000. When the scan is complete the value of the Parent device, and child devices, are set to the number of failed devices with 0 having the status 'Healthy'.

What the scans check is determined by the settings against each interface, they are:

• Check Last Change Time (the period of inactivity to check for can be configured for each Interface)
• Poll if not changed (optionally poll the device if it has been inactive for longer than the set period to see if it responds)
• Check for Invalid State (if HS doesn't receive a successful acknowledgment of a command it sets the Invalid State flag of devices, so this is an indication of communication problems)

Not all interfaces support polling it depends on whether this is implemented in the Interfaces plug-in.

The plug-in treats grouped devices as a single physical device when checking. So for a typical Z-Wave device only the parent or one of the child devices has to have shown activity in the period for the device to pass the test. Also, if the activity check shows a fail then only the parent is polled. That is unless the PollDepth is set greater than 1, in which case, if the parent device doesn't respond child devices are polled in sequence until a success or the PollDepth is reached. The interval between polls can be set to prevent them hogging the system.

An important point to appreciate is that, if Polling is requested, it is still only carried out on devices that have exceeded the inactivity period and only during a scan, so is not going to cause mass polling of your network.

You can optionally exclude individual devices from the scanning process if you require.

At the end of a scan 2 html reports are produced. One is a full report on all the devices in your system grouped against the selected Interfaces. The other report is a shortened version just for devices that Failed. These reports can be viewed from the HS web interface and/or sent be email as an attachment.

Moving on to your specific questions:

> If you wanted it to only trigger if there are some failures then you would add a condition that the value of the General-Health root device is greater than 'Healthy'. Basically the value of the Root device after each scan is the number of devices that failed with 0 being 'Healthy'.

Ok, so I only care about the Z-wave interface, so is this:

the right device - it doesn't say "root" - but "Z-wave SDJ-Health-General Z-wave (General Health)" - is the # of failed devices?

I can see that you have chosen a child device be created for the Z-Wave Interface. You can trigger on the Child Device or the Parent/Root Device. The difference is that the Child Device will trigger when the Z-Wave Interface has been scanned where as the Parent Device won't trigger until all the select interfaces have been scanned. If you only have the Z-Wave interface selected then there isn't much difference. However, I would generally recommend using the Parent.

ok so I want to be notified if the # is above 0, but the interface is only letting me enter one number:

so if I enter a number, will it be "anything above this number" or "exactly this number"? The way it's worded, it makes me think if I say "1", it will only trigger if exactly one device is in failure, but not if 2,3, etc. How do I enter "1 or more"?

You can set a range but the easiest way is to use:
IF
The Device has its value set and is greater than 'Healthy'
(with a condition)
AND IF
The Device has its value set and it is less than 'Activity Scan in Progress'

The example event in the guide is similar but the trigger is 'just had its value set or changed' but with the same condition as above. The difference being the example will always email a report after a scan, whether or not there are any failures.

And lastly: if I only want this email once a day, can I just trigger it like "if time is 4am AND ..."? or is there a way to have it email me if a *new* device goes into failure? I don't want a million emails for a device that's in failure and I already know about it, so do I just have it run once a day or is there a better way?

The triggers shown above will only send an email once after a scan because they value of the device is only changed during, and at the end of, a scan. They won't generate multiple emails. You could use a time of day as a trigger, with suitable conditions, but you would want to ensure it was shortly after a scan. It is better to trigger the scan at a time of day (automatically from the pi) and then let that trigger the email as per the examples.

I hope this helps,
Steve

[mlevin77]

Ok great. I'm using the Z-wave (child) because I don't care about the others. So, since the "Global Health Checking" is set for 2am (as the default, in the SDJ-Health config), I've got it set for 2:10am as follows:

Weirdly, it's complaining that the report doesn't exist, but if I trigger it manually, I get an email with a report attached, so I think this part is working fine. Hopefully I got the tests correct based on what you said above? Does this look right?

[SteveMSJ]

Ok great. I'm using the Z-wave (child) because I don't care about the others. So, since the "Global Health Checking" is set for 2am (as the default, in the SDJ-Health config), I've got it set for 2:10am as follows:

Weirdly, it's complaining that the report doesn't exist, but if I trigger it manually, I get an email with a report attached, so I think this part is working fine. Hopefully I got the tests correct based on what you said above? Does this look right?

Whilst you can use a time trigger I would recommend triggering on the device value, as my examples, because it ensures that the scan has finished. It also has the advantage of picking up any additional scans you may choose to run.

The attachment path in your screenshot looks convoluted so there is something wrong with your event.

Steve

[mlevin77]

Ok so I've gotten rid of the time condition, since it should only trigger once a day anyhow, and just kept the device values:

is that what you meant? Also, please note that the Attach File path looks normal, but as soon as I click the yellow triangle to hide the details, it starts to look weird:

(convoluted, and in red). I don't understand why, since I don't change the path, merely click on the yellow/white triangle (and also it works, to attach the correct file to the email it sends).

[SteveMSJ]

Ok so I've gotten rid of the time condition, since it should only trigger once a day anyhow, and just kept the device values:

is that what you meant? Also, please note that the Attach File path looks normal, but as soon as I click the yellow triangle to hide the details, it starts to look weird:

(convoluted, and in red). I don't understand why, since I don't change the path, merely click on the yellow/white triangle (and also it works, to attach the correct file to the email it sends).

Yes to the Trigger.

On my system the path shows as just '\html\SDJ-Health\SDJ-HealthGeneralReport-1.html' without the leading HS directory. This might just be an HS peculiarity if it sends the attachment when you test run the event.

Steve