Hi everyone, A came across a lot of these errors in /var/log/messages when I turned "scalability" off in gmetad.conf...
RRD_update (/var/lib/ganglia/rrds/unspecified/__SummaryInfo__/contexts.rrd): /var/lib/ganglia/rrds/unspecified/__SummaryInfo__/contexts.rrd: illegal attempt to update using time 1297207085 when last update time is 1297207085 (minimum one second step) A bit of searching on bugzilla found this related bug report ... "Illegal attempt to update RRDs with multiple gmetad" (see http://bugzilla.ganglia.info/cgi-bin/bugzilla/show_bug.cgi?id=96 ) I discovered that the error is generated when the last cluster listed in the XML reported by the downstream gmetad is written to disk a second time. This happens because the section of code to write the cluster summary when it parses a </CLUSTER> end tag also gets triggered by a </GRID> end tag. In process_xml.c there is a switch statement for GRID and CLUSTER tags where I believe there is a "break" missing -- exactly where the "No break" comment is! switch ( xt->tag ) { case GRID_TAG: rc = endElement_GRID(data, el); /* No break. */ case CLUSTER_TAG: rc = endElement_CLUSTER(data, el); break; default: break; } When I put a break exactly where the "No break" comment is and recompile gmetad I don't get any more RRD update errors. Does anyone know why there is no break ? It would seem a deliberate choice seeing that someone has explicitly added a comment that a break should not be present. Thanks, Nick ------------------------------------------------------------------------------ The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE: Pinpoint memory and threading errors before they happen. Find and fix more than 250 security defects in the development cycle. Locate bottlenecks in serial and parallel code that limit performance. http://p.sf.net/sfu/intel-dev2devfeb _______________________________________________ Ganglia-developers mailing list Ganglia-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/ganglia-developers
