This question already has an answer here:

This could sound like a very commercial question, but I really dont know how to justify this, the problem is that sometimes the page due the internet conection (I think) doesnt load, and the browsers returns an error message, monitoring the server, with alerts and all stuffs (incluiding nagios) seems like all services are responding, if I check the memory of the server still 200mb free(enough like to answer a simple web request), there is any other way to contactly check the timeline of a service??

Any idea?

marked as duplicate by EEAA, Ward, mdpc, Jenny D, voretaq7 Apr 10 '13 at 18:51

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.

  • Just to be clear, are you asking for ideas on why a page gets an error sometimes, or asking for ideas on how to prove that your server is up when a client claims that they are unable to connect to it? – DerfK Apr 10 '13 at 0:25
  • I am asking for ideas on how to prove that your server is up during the entire day – user156355 Apr 10 '13 at 15:38
  • Define "up". For telephony even a 1s of downtime could be a problem for a company. Some services could be considered "down" only if they are slower. Imagine 1 second delay on the phone. Now go further and think for GPS/GLONASS satellite systems that needs to have an atomic clock to keep an accurate time. And we are using a time that is drifting from atomic time (TAI) because UTC have to add leap seconds each few years. – Mircea Vutcovici Apr 10 '13 at 15:48
  • And if you think that a leap second does not meter, then read: wired.com/wiredenterprise/2012/07/… – Mircea Vutcovici Apr 10 '13 at 15:52

I think it is time to start monitoring your server from outside your network. There are a few websites that will do the basic monitoring for free.

  • Agree with the above however know that most of the free services simply check the ability to connect to a web page so a connection that returns a lot of "Error: Cannot open X service" will still pass (as will Nagios locally of course). I'd suggest external monitoring to your network which requires to find working strings on the response page (and perhaps requires that others aren't there). – Dave Apr 10 '13 at 0:52
  • there is no something else that could give me a timeline? I mean, external monitories are good, but they ping and try the services every period of time, maybe 10 minutes, and I would like to know if the server was ok every minute during those 10 minutes, any idea? – user156355 Apr 10 '13 at 15:35
  • You can install a remote monitoring service and configure the polling frequency as you like. It should be some providers to offer this kind of service. – Mircea Vutcovici Apr 10 '13 at 15:55

Not the answer you're looking for? Browse other questions tagged or ask your own question.