You are viewing a plain text version of this content. The canonical link for it is here.
Posted to solr-dev@lucene.apache.org by Mark Miller <ma...@gmail.com> on 2008/07/23 19:38:29 UTC
Pulling statistics from solr for different time slices
Before I create an issue, is it even possible / desirable to add
handlerStart and totaltime to the stats that the RequestHandlerBase reports?
With the current info, I can't see that the qps dropped to 0.03 at 3 am
on sunday (that is, unless the cumulative average dropped to that), or
that every Friday after 12, my average response time jumps to 10 seconds.
I'd like to be able to extract that type of info. I know using the
RequestHandlerBase is not perfect, you don't have to extend it and you
could override the stats, but it seems to be the best most universal
method I've found to get this type of info.
I am fairly certain that you can get by without the handlerStart info,
but totaltime would help a lot. Would it violate anything to add it? Is
there another way to get the info I want in a way that works across most
handlers out there?
- Mark
Re: Pulling statistics from solr for different time slices
Posted by Mark Miller <ma...@gmail.com>.
Chris Hostetter wrote:
> : Before I create an issue, is it even possible / desirable to add handlerStart
> : and totaltime to the stats that the RequestHandlerBase reports?
>
> are you talking about the aggregate stats reported by
> RequestHandlerBase.getStatistics() ?
>
> I see no reason not to ...
> Committed revision 680553.
>
>
>
> -Hoss
>
>
I am, thanks Hoss. The issue was that I don't want the full average, I
want the average over a given sample period, and including that info
lets you get it.
Re: Pulling statistics from solr for different time slices
Posted by Chris Hostetter <ho...@fucit.org>.
: Before I create an issue, is it even possible / desirable to add handlerStart
: and totaltime to the stats that the RequestHandlerBase reports?
are you talking about the aggregate stats reported by
RequestHandlerBase.getStatistics() ?
I see no reason not to ...
Committed revision 680553.
-Hoss