Errata

Monitoring with Ganglia

Errata for Monitoring with Ganglia

Submit your own errata for this product.

The errata list is a list of errors and their corrections that were found after the product was released.

The following errata were submitted by our customers and have not yet been approved or disproved by the author or editor. They solely represent the opinion of the customer.

Color Key: Serious technical mistake Minor technical mistake Language or formatting error Typo Question Note Update

Version Location Description Submitted by Date submitted
Printed Page 24
max_udp_msg_len paragraph

The text states that "it is not a good idea to change this value" from the default however it doesn't explain why.

Nick Satterly  Jan 13, 2013 
Printed Page 100
Example code

Text following the example should refer the reader to the heartbeat description text in Table 5-2 that explains why gmetric is called first with "-H" before submitting the custom metric.

Nick Satterly  Jan 13, 2013 
Printed Page 131,132
Two code examples

The two places where gmetric is called in the code examples has the variable "${CMDNAME}" where "PING_MS" or "PING_LOSS" should be and the "-t" option should have a valid type such as "uint16" or "float".

Nick Satterly  Jan 13, 2013 
Other Digital Version 814
2nd paragraph

This is for the Kindle version. It doesn't give page numbers so the Kindle location is 814 out of 6553. It's in chapter 2 in the section on the gmond configuration file. Under the description for host_tmax it states "gmond will consider the host as being down" yet in the actual configuration file for gmetad (not gmond) a comment states that it's the web frontend which considers the host as being down (and note that the comment in the actual configuration file for gmetad is exactly backwards according to the code - it's not TN < 4*TMAX it's actually TN > 4*TMAX according to ganglia.php. All this is very confusiing - there's no clear explanation in the book about where the TN value is found or generated during the monitoring process which is not helpful if one is trying to scale ganglia for hundreds of nodes which may require judging how to minimize network traffic by lengthening the time between collections of statistics without running afoul of the TN and TMAX calculations. The scaling chapter should cover this stuff in detail.

Michael Durket  Dec 18, 2012