Advisory: Support Portal Maintenance. Login is currently unavailable, more info available here.
What feature is impacted?Device Reports - "Last 24 hours Memory" Usage Report Bubble
What is the severity of the issue? (High, medium, low, minimal)low
Summary of the issues:The Memory usage repot percentage is ~factor 10 too low (show 7% instead 70% for my appliance)
Observed behavior (What it did or didn’t do):-
Desired behavior (How is it expected to or should behave):show the correct percentage of memory usage
How do we reproduce it (Provide instructions to help us reproduce the behavior):In the firewall report dashboard in the "Last 24 hours" Network widget memory usage is shown with wrong value.Same issue if the bubble get checked to enter the full stats, report goes from 0...10 instead 0...100%
Other (Any other detail that we need to know about):-
Supporting logs, screenshots, video, etc..:-
Thanks, Sascha. I have the team looking into this one.
What we observed is that there is a behavior difference between on box vs historical reporting. The dashlet is intended to show an average utilization over the past 24 hours so that you have an understanding of how utilized an appliance on average. The report below will show utilization peaks and dips over that time range. If we include the text 'average' in the dashlet, I think that will help clear any confusion on the behavior. The dashboard also does not yet reflect "Last 24 hours" at the top, which will also serve to reinforce that point.
I may not be understanding the explanation above correctly.
I just got in the EAP and let my device get some data up into Central. I too am showing the same with respect to memory usage. I know that my memory usage has been roughly 60% since I installed v18 yet the report shows 6% and 6.0 when you click the bubble.
Are you saying this is expected? Or it should show the average over the reporting period. So it's been reporting for 2 hours but we are going to average that over a 24hr period and that's where the 6% is coming from?
The memory and CPU dashlets will show the average over the last 24 hours. If that seems off, can you share the table of the memory usage over the past 24 hours from the dashboard? Also, show the memory usage on XG over the same time range.
I think this is what you are looking for, if not, let me know.
Central Reporting CPU
Central Reporting Memory
On-Box CPU/Memory Diagnostics (48 hours)
Thank you, this is what I was looking for. It definitely looks like memory reporting is inaccurate. I've asked our QE team to look into this. Thanks again for the feedback.
We've identified the root cause of this defect. We are tracking it with 52642. Thanks for reporting it to us!
Appears to be fixed with EAP2.