<gbudb/>

This element controls the generation of a GBUdb analysis X-header.

The content of the X-header will be the GBUdb information for the apparent source IP of the message at the time of the scan. (GBUdb data changes with each new scan)

<gbudb on-off='on'>X-GBUdb-Analysis</gbudb>

on-off='on' turns on the gbudb X-Header, on-off='off' turns it off.

X-GBUdb-Analysis is the default name of the X-header. You can change this text to customize the name of the header.

The contents of the header generally look something like this:

X-GBUdb-Analysis: 0, 12.34.56.78, Ugly c=0.071429 p=-1 Source Normal

This X-GBUdb-Analsys header would be interpreted as follows:

0 - The ordinal of the Received header containing the source IP (in this case the TOP Received header).

12.34.56.78 - The apparent source IP for this message.

Ugly - The GBUdb record type. Ugly means the IP is evaluated based on statistics. It DOES NOT mean that the IP is bad in any way-- just that it must always be examined. It is not administratively Good, Bad, or Ignored (infrastructure). As a mnemoic device, "U" also means Uspecified, Usual process, or "keep an eye on this Ugly mug."

c=0.071429 - The confidence figure is an indicator of how much experience this SNF node has with this message source.

p=-1 - The probability figure is an indicator of how likely it is for this SNF node to see spam from this message source. Negative numbers indicate spam is less likely, positive numbers indicate spam is more likely, and zero is indeterminate (50/50 chance). In this case (probability figure = -1) there is no expectation of spam from this IP source.

Source - Indicates that this analysis is for the apparrent source IP for this message.

Normal - Indicates that this IP's statistics fall in the "Normal" range as defined by this node's GBUdb range envelope settings. For more information about this see our page on GBUdb Evaluation Ranges.

For more information on how to interpret GBUdb analysis data see our page on What GBUdb Learns.

Please email support@armresearch.com with any questions.

Related Topics