Free performance monitoring for VMware™ and IBM Power Systems™

3.20

It contains these major enhancements:
  • Custom groups You can group arbitrary lpars or CPU pools together and then get on 2 clicks required graphs. It is already running on Live demo site.
  • Favourites You can define your favourite graphs (lpars or pools) and then you can easilly find them under new "Favourites" menu. This also already running on Live demo site.
  • Alerting. It allows you define alerts based on CPU utilization of lpars or CPU pools.
More also on page 3.20 announcement page

Another changes and enhancements:
  • lpar aggregated and 'total cpu per HMC" graphs have been extended about MAX column to better identified peeks
  • all graphs are now created with "slope-mode" rrdtool function
  • real time fix, for IVM/SDMC fixed download time range from 86400 to 1440 minutes
  • repaired Excel export of aggregated graphs "ERROR: Attempting to reuse 'utiltotu0' at ... "
  • implemented CPW benchmark (iSeries based benchmark) for CPU workload estimator
  • changed directory structure of lpar2rrd home


Knows issues which might appear after 3.20 upgrade.
Please check below before going to support. Apologize for inconvenience.
  • Does not work CPU pool alerting due to a bug in update_cfg_alert.sh. Download and replace scripts/update_cfg_alert.sh.
    Remove etc/alert.cfg if you have already created it and create it once more via new script: "$ scripts/update_cfg_alert.sh"
  • If you have any problem then at first look at etc/lpar2rd.cfg if SSH directive has a space between "ssh" and "-q" : SSH="ssh -q"
    Best is remove "-q" from there.
  • Update or install process fails for creating favourites, alerts and custom group cfg files.
    Run failed cmds manually.
    Note that for alerting use this script scripts/update_cfg_alert.sh as is explained above.
  • Physical and Logical configuration web page is completely blank
    Solution is clearing browser cache/restart browser/press shift during reload or all together.
  • There has been found a bug in alerting, fix it manualy before is released 3.21 where it will be fixed.
    Just go to LPAR2RRD home dir and copy&pasted text started with "ed " and ended "EOF"
    $ cd $LPAR2RRD_HOME$
    $ ed bin/alrt.pl << EOF
    284s/managedname/managedname_prev/g
    w
    q
    EOF