I discovered, while testing the output from the new python rewrite of gmetad, that there where EXTRA_DATA_ELEMENT tags missing from the XML output of the original gmetad. The problem turned out to be that the summary metric data was overwritting the actual metric data during the evaluation of the METRIC tag from gmond. This meant that when the EXTRA_DATA_ELEMENT tags where later read and added to the metric data, the original metric data no longer existed and instead was replaced with the summary metric data. Since some of the summary data no longer lined up with the original metric data, some of the extra data was being dropped.
I have added a patch (r1302) to trunk to fix this problem however the patch touches enough code that it really needs some review before being back ported. I have added a back port proposal to the 3.1.x STATUS file. Please review the patch and vote. I consider this bug to be a showstopper for 3.1 Brad ------------------------------------------------------------------------- This SF.net email is sponsored by the 2008 JavaOne(SM) Conference Don't miss this year's exciting event. There's still time to save $100. Use priority code J8TL2D2. http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone _______________________________________________ Ganglia-developers mailing list Ganglia-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/ganglia-developers
