Custom Query (121 matches)

The code:

foam_slivers_rrd = '%s/foam_slivers.rrd' % (RRD_DIR)
if os.path.isfile(foam_slivers_rrd):
  foam_state = gmoc.FOAMSliverCount()
  foam_state.loadRRD(foam_slivers_rrd, start_secs, end_secs)
  aggregate.addMeasurement(foam_state)

yields:

tau-verde,[~],19:36(148)$ sudo -u ganglia ./report_relations -d
Traceback (most recent call last):
  File "./report_relations", line 311, in <module>
    aggregate.addMeasurement(foam_state)
  File "/home/chaos/gmoc.py", line 1237, in addMeasurement
    if self.type == None or not self.type.uppercase == "FOAM":
AttributeError: 'str' object has no attribute 'uppercase'

To replicate:

• Browse to [​https://gmoc-db2.grnoc.iu.edu/protected-openid/index.pl?method=aggregate_details;aggregate=foam5.gpolab.bbn.com%3A3626 the foam5.gpolab.bbn.com:3626 aggregate details page.
• Click on the "Aggregate Measurements" tab.
• Click on the "Details" link of any of the available measurements (e.g. ​Disk Utilization)
• Click on the row to bring up the graph.

The UI brings up the graph of some arbitrary unrelated measurement (a "flowvisor-slice-state" measurement in every graph i've tried so far, but that's probably coincidence), rather than the requested measurement.

When gmoc.py tries to submit relational or measurement data, it prints parts of the response it gets, if invoked in verbose mode. In addition to this, gmoc.py should also store those responses in variables belonging to the object, so that if the calling program wants to print things at a different granularity from the gmoc.py defaults, it can.