The Website Monitor lacks clear documentation

david ascher shared this idea 1 week ago
Open Discussion

I've been trying to use the Website Monitor to keep track of the health of one of my websites. Last week, for no apparent reason, I started receiving emails saying that my site was unreachable. Eventually, with the help of a support agent, I was able to track this to my having entered the target word "volunteer" when the site displayed a link with the text "Volunteer". I modified the target word to match the case of the link text and the email notifications stopped. I then modified that target word again, to use a word that was not a link - "grassroots" in this case. The support agent then reported that he was unable to reproduce my original problem where the case of "volunteer" (the target word) didn't match the case of the word on the site. I confirmed that was, for some reason no longer an issue. One support agent told me that the monitor has recently been updated and  the target word now has to match the word that appears on the site. Another reported that this was not part of the updated monitor. I was therefor left not having any idea why I had suddenly started receiving alarm emails for that site from the monitor. Upon further examination, I discovered that the href attribute for the original link from the word "Volunteer" contained the string "volunteer". This discovery only increased my confusion about why I had originally received the alarm emails from the monitor. 

What I would like to see is documentation describing a) whether the target word(s) must match the words on the website and b) whether only "visible" text is checked for a match to the target word(s) or if "invisible" text, contained in attributes, tagnames, directives, etc. are also checked for a match to the target word(s). 

I am not particularly concerned as to what the answers are to those to questions, but it seems obvious to me that somebody configuring the monitor should have the correct information about those two things in order to know if they are choosing a target word that may match something on the site (like a tag name, attribute name or value, etc.) which could be matched even if some hacker had managed to modify the page and in doing so remove the visible target word that was intended to be match.

0 votes

Comments (1)

photo

Hello, David!

The content health monitoring works like this:

- SPanel uses curl to fetch the website's front-end code content (html, css, javascript)

- SPanel checks if the monitored keyword is present in the received output from the curl request

The monitored keyword check is not case sensitive.

If the monitored keyword is not present e.g. the website returns an error, you will get a notification.

Leave a Comment