Server Fault is a question and answer site for system and network administrators. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I have the following scenario:

a small internal network behind a DSL connection with problems (short outages several times a month) AND a virtual server at a hosting company.

I want to implement monitoring and statistics for my machines (availability, load, ... )

How can I do this in one central place? It's obvious that if my DSL connection is down I can't get the current data for the virtual server, but I would want the solution to "fill the blanks" in the local database afterwards.

From what I have read, Zabbix might be able to do this? Does a Zabbix Proxy buffer the data until the main Zabbix instance is available again?

Any other recommendations for monitoring systems that support this use case? Thanks!

share|improve this question
up vote 2 down vote accepted

Zabbix Proxy can be used for this purposes:

  • Offload Zabbix Server when monitoring thousands of devices
  • Monitor remote locations
  • Monitor locations having unreliable communications
  • Simplify maintenance of distributed monitoring
  • List item

share|improve this answer

The last time I had worked with Zabbix (though it was long time back), I don't think I saw this type of functionality in it. Also, I am not sure whether I had heard anything like this in any other tool.

Though I am sure people will update if there is any tool which had any such type of functionality.

share|improve this answer
I'm not sure that I'd agree that monitoring will take a big chunk of bandwidth. – mdpc Dec 22 '12 at 16:41
I know it's not that great, but monitoring one server over WAN shouldn't put that much stress on the connection... I mean, if I have 60 metrics/updated every 60s thats 1 value per second. – allan Dec 22 '12 at 16:43
Though I meant to write that it depends on how many metrics you want to monitor and how many machines, but I surely need some more coffee to stay awake. Editing my answer. Thanx guys. – GeekRide Dec 22 '12 at 16:47

I'd suggest that in this case you'll have to make all your triggers LESS sensitive, and possibly require more problems in a row if you will before you trigger an alert.

As for other monitoring software, that is a great matter for debate and discussion. I have successfully used nagios/pnp4nagios combination very successfully for a site of about 500 systems. Nice thing it is free and you can usually find packaging for the major Linux distributions as well as a wealth of plug-ins for many many situations contributed by nagios users.

share|improve this answer

Monitoring solutions built on RRDtool, which is most of them it seems, have the ability to switch how they handle 'unknown' values. One setting fills those woth voids, the other reuses the previous value. Changing this behavior varies based on the exact monitoring package.

Backfilling with interpolated values is not supported by RRDtool based systems.

share|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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