Details
-
Improvement
-
Status: Resolved
-
Minor
-
Resolution: Fixed
-
None
-
None
-
Minor
-
Not applicable
Description
The current Cluster Disk usage report under the Activity tab in ECL Watch allows a user to assess cluster disk usage.
However, a detailed description of what the numbers and the color code scheme displayed on the speedometers represent appears to be currently not documented in the HPCC Systems documents. This Jira ticket is for adding a more detailed description of the "Disk Usage" report in the "Using ECL Watch" document. Few suggested items to be included are:
- Some comments about how MAX and MEAN values displayed on the speedometers are calculated. For instance: per my initial assessments, it looks like that MAX represents the value of the single filesystem of a slave node showing largest occupation within the cluster, whereas the MEAN represents and average of the occupation across all filesystems of all slave nodes within the cluster.
- Some comments about what are the thresholds behind the color scheme. Per my initial assessments, it looks like that green color is displayed until usage is around 85%, then orange color is displayed until about 95% and then red color up to 100% of usage.