Keep getting false alarm

Help and Support
,
1

Hi,
I have setup a very simple monitor. The only rule in the monitor is:

image
image638×217 9.76 KB

My understanding being that the rule is triggered if the part of the text changed contains the word Datograph. However I get many alerts and when I look in the page that caused the alert there is no track whatsoever of the word Datograph.
Am I doing something wrong?
Thanks

2

@aguida based on the condition, the monitored text must have contained the term “Datograph” when it triggered a notification.

can you check the notification status to see why the condition matched? see Optimizing Alerts with Conditions – Distill

click the bell icon in change history to see the notification status and condition evaluation’s result.

3

Thank you so much, and apologies for the noob questions, I am really new to this.
I will check next time I get an alarm. I had three this night but I cannot find them anymore.

I have a dedicated server, running the desktop version running the monitor every 30 seconds.
Unless I am there to look for the notification soon after received the history will be overloaded by hundreds of new monitor runs.

How do I find back to the one who generated the notification?

If I click on the mail I received “Change History”

image
image280×114 7.91 KB

I get a running line that never stops running (probably because the history is too long?)

How do I find the run who caused the notification when it is 100 or more back in time?

If that is not possible, is it possible to add an event that once the notification is sent it stops the monitor? So I will get a chance to analyze the notification and restart it afterwards.

Thank you, still love the tool. Great job!

4

you can setup an email alert so that older changes get sent to your mailbox. the emails can then be used to see that the text was present in the content.

this is a good ask. we don’t have this feature yet but we plan to support similar functionality in the future.

5

Apologies for getting back to this. Your last suggestion was very helpful (but I still hope you will implement the stop action as mentioned above):
However I have another similar issue now. When I look at the watchlist I often get “Unread” monitors.

image
image263×217 56.5 KB

Not sure if I am misunderstanding, but I guess this means that there has been an alarm, I haven’t looked at yet. But I did not receive any mail or message (both are enabled). What is then the meaning of “Unread”? And of course I cannot find anything scrolling through all the diff screens, also because checking every 30 seconds there are hundreds of them.
.
Thanks

6

Which brings me to a new suggestion (and again apologies if this is stupid, as I am rather new to this tool):

Would it be possible to add a setting so that the tool will cache and store only the checks where there has been triggered an alarm? Right now I have to go through hundreds of them that have not triggered any alarm before (if I am lucky) finding the one who did:
image
This for me would be even more valuable that my previous suggestion to introduce a stop action

7

a monitor is marked as unread when there was a change that met the conditions

actions are triggered only if the condition is satisfied. if conditions are not set, all changes trigger actions.

note that the monitor will remain unread even if there are new updates that didn’t satisfy the condition.

it is possible to discard a check’s result using a macro. macro has a function named panic. this function will error out if it gets a true value when called. following post shows an example usage of the panic function in a related use case:

let me know if you have any other questions.