Page 1 of 1
Bandwidth usage table lags by >24 hours
Posted: Mon Aug 03, 2020 9:45 am
by GldRush98
Capture.PNG
I'm curious as to why this table data lags behind the actual data Nagios has?
You can see the Day 1 column is populated, but even though there is data for Day 2 as well as Day 3 (today) in the graph, this data is not represented in the text table. It seems this table constantly lags behind by the current day plus the previous day never being populated.
Is there a reason for this?
Is there any way to make this table accurately reflect the currently known data amounts?
Re: Bandwidth usage table lags by >24 hours
Posted: Tue Aug 04, 2020 11:22 am
by jbrunkow
Nagios spools performance data into small files which get moved around and processed. Sometimes a problem can arise which stops the files from being processed and they begin to spool up.
Troubleshooting Graphs
Can you see whether some performance data has spooled up on that server?
Code: Select all
ls /usr/local/nagios/var/spool/perfdata/ | wc -l
ls /usr/local/nagios/var/spool/xidpe/ | wc -l
You can clear this spool to see if the process speeds up thereafter.
Code: Select all
find /usr/local/nagios/var/spool/perfdata/ -type f -delete
We can change the way that Nagios processes performance data by editing the relevant configuration file.
Code: Select all
/usr/local/nagios/etc/pnp/process_perfdata.cfg
It never hurts to send (
PM ) us a profile either! We can find lots of information about your system in there. You can download the profile by navigating to
System Profile under
System Config on the left pane > and clicking the
Download Profile button on that page.
Re: Bandwidth usage table lags by >24 hours
Posted: Fri Aug 21, 2020 1:45 pm
by GldRush98
Nothing is backed up as far as perfdata goes. As you can see the perfdata graph is accurate and up to date.
Code: Select all
[root@nagios ~]# ls /usr/local/nagios/var/spool/perfdata/ | wc -l
0
[root@nagios ~]# ls /usr/local/nagios/var/spool/xidpe/ | wc -l
0
[root@nagios ~]# ll /usr/local/nagios/var/spool/perfdata/
total 0
You know what... now that I'm looking at more data, everything in the usage table (the text table) appears off a day compared to the perfdata graph itself. Assuming the graph is right, which I believe it is because it is showing up-to-current data... the data spikes clearly in the graph are NOT lining up to where the high numbers are in the text table.
Everything is shifted a day behind in the table
Capture.PNG
This text table is inaccurate. It could be the php timezone shifting things off, or it might be a full 24 hours off, I am not positive, but something is causing the data in the table to be wrong when compared to the actual perfdata represented by the highchart.
Re: Bandwidth usage table lags by >24 hours
Posted: Fri Aug 21, 2020 2:23 pm
by GldRush98
With some more experimenting I believe that the php timezone offset is throwing the text table's calculations off. There should probably be a bug report filed for this as it appears that if php's current timezone is accounted for, data looks to shift to the correct days and even the previous day's data shows up in my table. (today's data does not, but I don't necessarily expect it to until the day is "complete")
Capture2.PNG
Also, June is spelled wrong when the month array is set up:
Code: Select all
/* Convert a month number to a month name */
function monthname($no)
{
$names = array(1 => 'January', 'February', 'March', 'April', 'May', 'Junr', 'July', 'August', 'September', 'October', 'November', 'December');
return $names[$no];
} //function monthname($no)
Re: Bandwidth usage table lags by >24 hours
Posted: Mon Aug 24, 2020 10:42 am
by jbrunkow
Alright, I have submitted a bug report for this issue. I will pass that information along to our developers.
Forgive me if this is a bit obvious. But have you double checked that all of the time zone settings are correct on your
XI instance?
Changing The System Time
Re: Bandwidth usage table lags by >24 hours
Posted: Tue Aug 25, 2020 11:22 am
by GldRush98
Yes, I have double checked all of the various timezone settings and all are set and showing correctly for my timezone.
Re: Bandwidth usage table lags by >24 hours
Posted: Tue Aug 25, 2020 2:50 pm
by jbrunkow
Thank you for confirming that.
Our developers confirmed that report is working as intended. The reason this data lags behind is that our software uses a
round robin database to average this data over time, so that it does not become too large to manage.
RRDtool