Moni is a light weight, self contianed (single executable) website <<<<<<< HEAD walker and analyzer. It is basically like a "ping" for an entire website.
Just point at one or more websites that you manage, moni will grab every page of your website, record errors it found some, otherwise the pages response time will also be recorded.
You will be able to tell pretty quickly if there are any problems with your website, including its responsiveness. Moni was written in go with concurrency in mind, moni is light, tight and fast.
Moni has no external dependencies, just provide one or more URLs and determine if your website is behaving like you expect it to.
======= walker and tracker of things connected to networks. moni is small and really fast, because it was written with go, check it out if you are interested! It is also really cool. :)
While moni has no required external dependencies, you can just run the executable, feed it resources to watch and let moni go aboud doing her thing.
Moni has been built as an API first, with clients for the command line and WebUI are clients of the API. REST is currently up and running, with ~hopes~ plans for websocket and graph api lurking just over the horizon.
Moni is also extensible, that is moni can indeed make use of external functionality for storage, caching and persistence, logging facilities. More on this topic later ...
Moni depends heavily on the excellent Go package colly to do the actual fetch and parse of the HTML pages.
It constantly monitors your website(s) for both reliability and performance. Crawl is written go (very small and fast) as a single executable with no required external dependencies (just copy the program and use it.)
74e7161b9026bfee04893cf7115559b79c5c6ce2
How to Use
Crawl is both a daemon and an angel (JK :) and a command line tool. When run as a command line tool crawl can be used to diagnose website troubles in real time.
When run as a daemon or background service crawl monitors the availability and performance of the list of websites.
% ./moni rustyeddy.com
http://rustyeddy.com
https://rustyeddy.com/projects
https://rustyeddy.com/contact
https://rustyeddy.com/projects/crawl
https://rustyeddy.com/
https://rustyeddy.com/interview
https://rustyeddy.com/notes
https://rustyeddy.com/resume
elapsed time 341.752345msOr watch a group of websites with a quick and easy single line output:
http://rustyeddy.com: resp: 659.999µs links: 7
http://oclowvision.com: resp: 986.158µs links: 10
http://orangecountylowvision.com: resp: 888.039µs links: 10
http://sierrahydrographics.com: resp: 1.644021ms links: 16
http://gardenpassages.com: resp: 1.136244ms links: 22
http://skillcraft.com: resp: 1.519748ms links: 14
http://mobilerobot.io: resp: 1.042628ms links: 7This is handy if you manage a lot of websites, or just want to get a quick summary of how your web properties are doing..
Moni walks every page of a website, that it can find, and reports back with the stats gathered above. The output will be produced according to your desires! Text or JSON? You got it!
If desired, Moni will also walk every sub-page of the website, however all external links with NOT be walked otherwise we will quickly be walking every website in the world, and be the next Google.
We don't want to do that, we just want to ensure the good health of our own web properties, and find out immediately when something goes wrong (and it will, this is tech after all).
Moni uses the go package colly to monitor given websites. Colly makes a standard get request to a given URL, gathers the links from that web page, as well as text and other elements, then returns. The return gets a timestamp that we use to record the elapsed time.
Each link of the webpage is now categorized as either internal or external, internal links may also be crawled, where as external links will not be crawled.
Internal and External links will be tested for reachability.
Are links from one page of a given website to another page of the same website. We define two web pages being on the the same website when:
Two page belong to the same website when their respective [canonical] URL's share the same root domain.
For simplicity we defer the definition of root domain to DNS, and take a rather ignorant approach of simply comparing host portion of the URL.
When we crawl a page we gather all of its contents, in particular all links on a given page, as well as content. We then optionally will also crawl any internal links and links to permitted external sites.
The reason for blocking un-allowed external sites is to avoid walking sites like amazon.com and github.com just because the page has a link to it.
Reachability on the other hand is much lighter weight, we simply verify the page is still online and available. If not, it will be flagged as a broken link.
We do not collect any content from the Reachability test and hence we do not have any additional links to crawl.