Splunk for *nix technology add-on intervals

[rainhailrob]

Has anyone seen an issue increasing the intervals for TA running SUSE? When I increase the intervals, the *nix 4.5 app stops updating the graphs.

Here are the intervals I changed:
vmstat- 300
iostat- 600
ps- 300
top- 300
protocol- 300
lsof- 600
df- 600
cpu- 300
fschange- 600

However CPU Overview, Memory Overview, Disk Overview don't work for last 15 mins or last 60 mins. I don't know if there isn't enough data points for the graph? I edited the CPU overview "CPU consumption by command" graph to 2 hours and the graph works. Anything less than 120 mins (even 90 mins) and the graph does not fill in.

[araitz · Accepted solution]

For short time ranges, the CPU graphs sometimes don't seem to populate as expected. The main reason is that the views were created in Splunk's Simple XML and sparse points are not connected such as to make the timecharts visible. We have just fixed that issue in preparing for the next release of the app. Setting the intervals as long as you did will almost certainly exacerbate the problem.