It is the unknown unknowns that are the scariest. Those days when a server unexpectedly runs out of disk space. It usually happens to me because I have a server connected to my Dropbox account. With the folder that the vBrownBag crew uses to make TechTalks at conferences. Last week at OpenStack summit we made 20GB of video, so my Dropbox folder grew by 20GB in three days. When this has happened in the past my server stopped syncing with Dropbox to avoid completely filling the drive. As a result, my Mac and my VDI desktop didn’t the same files, which gets in the way of my business workflows.
Happily, the week before OpenStack Summit I was at VMworld EMEA and took part in the TFDx festivities. As usual, refer to my TFD disclaimer post. One of the presentations was by Paessler, who make PRTG. I had seen their presentation at the VMworld US edition of TFDx, so while I half-listened to the presentation I also deployed PRTG into my lab. It was a simple install onto a Windows server and management through a web interface. The install and configuration were done in about a quarter hour. PRTG immediately had some things to say about my lab, most of the warnings relate to SSL configuration on the VMware Horizon View services. I poked around at the warnings and then promptly forgot that I had installed PRTG at all.
During the week of OpenStack Summit, we were flat out making videos, over 80 videos in three days makes for busy days. On day three I got a notification email from PRTG. My server was running low on disk space. I knew the culprit was the Dropbox folder full of video, both the OpenStack video and the video from VMworld EMEA too. The low space warning reminded me to move the VMworld EMEA videos to a backup location and avoid the cessation of Dropbox files.
For a more complete look at PRTG, I suggest you look at David Varnum’s post or Lino Tera’s one.
Visibility into what is going on in your data center is critical to avoid unpleasant surprises. My lab is a pretty small datacenter and it can still be painful when surprises occur. Eliminating as many as possible of the unknown unknowns from a larger data center will be very valuable. PRTG started life as a router and network monitor, but I only use it to watch my servers, ESXi and VMs. In a larger organization, a unified view health of multiple types of devices is very valuable.
© 2016, Alastair. All rights reserved.
RSS - Posts