Highlighted
2 Bronze

Watch4Net graphs help

I have recently moved from a single-server W4N to a dual-server W4N installation (1 Collector, 1 Backend/Web).  All the data is collecting fine and there don't seem to be any errors in the logs that I can find.  The problem is the graphs seem to spike and drop almost like there is missing data.  I have attached 2 screenshots, first is from my single-server installation and the second is the dual-server install.  You can see the difference.

Screenshot 2013-10-16 17.28.05.png

Screenshot 2013-10-16 17.26.56.png

Anyone have any idea why this may be happening?  I have synced the time on both servers using NTP and checked all the logs I could find but don't see any network disconnects or errors between the Collector and Backend.

I should point out, I know these spikes aren't "true", so there seems to be a data issue somewhere.

Thanks in advance

0 Kudos
Reply
2 Replies
Highlighted
4 Ruthenium

Re: Watch4Net graphs help

Please consider moving this question as-is (no need to recreate) to the proper forum for maximum visibility.  Questions written to the users' own "Discussions" space don't get the same amount of attention and can go unanswered for a long time.

You can do so by selecting "Move" under ACTIONS along the upper-right.  Then search for and select: "Storage Resource Management Community".

0 Kudos
Reply
Highlighted
2 Bronze

Re: Watch4Net graphs help

Hello,

I'll be honest, I don't see anything wrong with that second picture. The peaky nature you are seeing is just the drawing of the graphs between the data points. From that graph it appears the data it is showing is being polled with a 5 minute polling cycle, hence the way it is being displayed. The first graph is over a much wider timescale, so it's not as noticeable, however you can still see it in some parts, particularly the top of the graph at around the 11:15 mark.

If you want more granular data you can of course change the polling frequency for the data, however it's worth keeping in mind the extra load this will generate on the systems being polled, the impact on the database size to store the extra data points, and ensuring that all components in the collector chain have the change made accordingly (e.g. if you change the polling cycle in the actual config for that collector type, make sure you change the polling frequency values in other components being used such as the availability filter which has a default of 300 as the assumed poll interval).

Benjamin Johns

iQ Consult Pty Ltd

0 Kudos
Reply