Monitoring servers externally?

6 views

Skip to first unread message

Jeff MacDuff

unread,

Mar 8, 2012, 10:22:06 PM3/8/12

to dotnet...@googlegroups.com

I wanted to ask what people are using to monitor their machines (probably pertains more to colo) externally?

We have been using Pingdom up until now and I love the service. We basically keep a extra public firewalled net tap open for each server which can respond to a ping. Pingdom can them monitor the machines and text me the second one of them goes down. The service also gives us global stats on latency from the pingdom DC’s to ours. I can easily see on a map the trace route and historical connection speeds from the US, Europe, etc.

The service works great, and easy to manage. We are doing some adjustments to our infrastructure and I wanted to see if there was something better?

We want something “not in our DC” and something we don’t need to host.

Thanks!

Jeff MacDuff

CTO & Co-Founder, Buddy

Email: je...@buddy.com

Chris Kinsman

unread,

Mar 8, 2012, 10:45:29 PM3/8/12

to dotnet...@googlegroups.com

Really depends on what you are trying to accomplish. Pingdom is great but limited.

Next product I would check out would be prtg by Paessler. Great for monitoring at the protocol level and below. Disk space, CPU, smtp, FTP, bandwidth in, bandwidth out by switch port, etc. provides alerting via email, sms, and provides an iPhone app to look at all the data. It also trands and tracks the data over time.

Final product I would check out would be an external synthetic test suite. Something like Gomez or an analogous product. Gomez is a bit limited for services and/or ajaxy type applications but the idea would be to have a full workflow simulation that logs in, does some work, logs out. This allows you to do a limited functional test and benchmark the performance of each step. You can alert when certain thresholds are exceeded, etc. The tests can be run in a distributed fashion from machines all over the world and give you insight into not only your application working but also how it looks to end users by geography. Very useful but depending on test frequency can be expensive. In the case of Gomez they also provide private agents you can run in customer sites or behind your firewall in your colo to do things like hit individual web servers instead of hitting your publicly exposed load balanced bridge head.