Quoting [EMAIL PROTECTED]:

> Hi again,
>
> I have been trying to figure out SLA's for some hours now and start to
> understand them.
>
> I noticed in the SLA Report overview of some of my Storage Interfaces (which
> were 'failing the SLA') that there are lots of events, a selection:
>
> Start          Stop            Duration        Type
> 2005-01-21 00:00:02   2005-01-21 00:30:32     00:30:30 hs     SLA X
> 2005-01-21 00:30:02   2005-01-21 01:00:32     00:30:30 hs     SLA X
> 2005-01-21 01:00:02   2005-01-21 01:30:32     00:30:30 hs     SLA X
>
> For some reason they overlap, I guess there is a reason for that. But now one
> sees multiple SLA 'failures', while it is simply a large one. This also
> results
> in a longer Unavailable time than view range and a negative Availability
> percentage.
> This was for a harddrive, but my Linux Memory is using Storage polling too,
> and
> that is more dynamic. Because of all these overlapping events, you don't see
> when the % storage is not too high.

I still don't understand the above behaviour. I do have some more information
though. I now know these alarms are generated by the rrd_analyzer. But what I
don't understand is this:

If at 13:11 an SLA is failed, and then the rrd_analyzer runs, it adds an event
with start time 13:11, but it also gets an end time exactly 30:30 minutes
later.
Now by default the rrd_analyzer runs every 30 minutes (as cronjob), so that is
where the 30 minutes plus an extra 30 seconds might come from.
But it must be hardcoded, since when I run the rrd_analyzer every minute i get
an alarm every minute too, both 30:30 long and thus overlapping

13:11:00 -> 13:41:30
13:12:00 -> 13:42:30

What I don't understand is why it:
- gives an alarm that ends in the future
- and thus gives false information, or at least an alarm for a time the analyzer
knows nothing about.
- the minimum period is 30:30 minutes; what if I want to know a more detailed
time period of the SLA failure?

At this moment the alarm information for this storage device is useless to me. I
could not test yet whether this is only for 'Real Memory', for the 'Storage'
type or for all SLA's.

Alex

----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.


-------------------------------------------------------
This SF.Net email is sponsored by: IntelliVIEW -- Interactive Reporting
Tool for open source databases. Create drag-&-drop reports. Save time
by over 75%! Publish reports on the web. Export to DOC, XLS, RTF, etc.
Download a FREE copy at http://www.intelliview.com/go/osdn_nl
_______________________________________________
jffnms-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/jffnms-users

Reply via email to