CEO's Blog

As of the new Apple operating system Leopard, Macintosh users are able to monitor their resources and their hardware based on SNMP. John Welch has posted an article explaining the implementation and functionality of this newly included SNMP system at The Unofficial Apple Weblog. You can find the article under http://www.tuaw.com/2007/12/05/john-welchs-guide-to-snmp-on-leopard. SNMP is one of the ways to monitor your systems with both our flagship software products, PRTG and IPCheck.

Some customers have had issues with installations of SNMP Helper Pro when the Data Execution Prevention (DEP) feature of Windows 2003 server was enabled. Also in some situations the packet sniffer feature of PRTG has shown problems with DEP enabled.

Read more in our knowledgebase

PRTG allows administrators to discern actual bandwidth usage in their network based on multiple parameters, such as IP addresses, port numbers, protocols, etc., using either packet sniffing or NetFlow collector sensors. A new article in our knowledgebase explains how this feature can be used to find out who (or what PC/Server) is causing traffic peaks and excessive bandwidth usage in a network.

Read the knowledgebase article

When creating “Custom SNMP Sensors” in PRTG Traffic Grapher you need to choose between a “Delta counter (difference per period)” and a “Gauge counter (absolute value)”. Sometimes you may want to change this setting later.

Our latest knowledgebase article explains how to do it.

Apparently a traffic accident is to blame for almost 3 hours of downtime for our corporate websites in the night from Monday to Tuesday. A truck driver lost control over his vehicle due to low blood sugar and drove into a power transformer on the side of the Dallas-area data center of Rackspace Inc. in Grapevine, Texas, which hosts our server park.

(more…)

Today we have published a new beta version of PRTG V6.2. Our developers have recently worked on PRTG’s SNMP engine and optimized it heavily.

It can now query more SNMP values in less time and with less network load. Especially for installations with some hundred sensors or more (and for installations with small intervals, e.g. below 10 seconds) this new version requires less CPU load and causes less network traffic while showing better performance.

Download this beta (the beta expires Dec 12th 2007).

There are SNMP devices that sometimes change the OID for a specific counter. This is common for systems running Windows 2003. If this happens PRTG can not continue to monitor this counter and it is easiest to add a new sensor.

If the historic data needs to be bound to the new sensor in order to retain the monitoring structure please refer to our knoweldge base article Binding historic datasets to new sensors and recalculating historic data for an individual sensor.

Last night we have celebrated our new office space with a housewarming party for our team and the local network that keeps the company going. Thanks to everyone who joined us and enjoyed the party!

(Click the photo to zoom in)

Recently we came across the following three pretty notable pages provided by Cisco, allowing users to find Cisco MIBs and to discern OID segment almost at a glance.

(more…)

We have had multiple queries regarding strange, respectively wrong readouts when monitoring certain ASA equipment. As we found out today, the problem is a bug in the 7.2 OS causing the SNMP iftable to forward erroneous return values to monitoring software, such as PRTG. Instead of graphing PER vlan values, this bug results in graphing the total number of untagged vlan packets for each vlan.

(more…)