Explicit documentation regarding all JMX metrics?

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

Explicit documentation regarding all JMX metrics?

Koen De Groote
Greetings,

I've enabled JMX metrics on an install of mine and am going through the
available beans.

I've google for an explanation of what the beans all mean, but I'm coming
up empty.

For instance, these:

:category=ADMIN,...,name=clientErrors,scope=/admin/file
:category=ADMIN,...,name=clientErrors,scope=/admin/logging
:category=ADMIN,...,name=clientErrors,scope=/admin/luke
:category=ADMIN,...,name=clientErrors,scope=/admin/mbeans
:category=ADMIN,...,name=clientErrors,scope=/admin/ping
:category=ADMIN,...,name=clientErrors,scope=/admin/plugins
:category=ADMIN,...,name=clientErrors,scope=/admin/properties
:category=ADMIN,...,name=clientErrors,scope=/admin/segments
:category=ADMIN,...,name=clientErrors,scope=/admin/system
:category=ADMIN,...,name=clientErrors,scope=/admin/threads
:category=ADMIN,...,name=clientErrors,scope=/config
:category=ADMIN,...,name=clientErrors,scope=/schema

What am I looking at here?

I can't find a single page explaining what the significance of these
metrics is.

Nor can I find a page that basically says "You'll want to monitor X, Y and
Z get a good view of your cluster".

Can anyone provide an example?

Kind regards,
Koen De Groote
Reply | Threaded
Open this post in threaded view
|

Re: Explicit documentation regarding all JMX metrics?

Erick Erickson
Here’s something that might be useful:
https://support.lucidworks.com/s/article/What-should-you-monitor

What version of Solr? For Solr 6.4+, you may want to use the /admin/metrics endpoint, see the reference guide for metrics. WARNING: there are a ton of metrics. The ref guide has information on now to trim down what you’re looking at.

Best,
Erick

> On Nov 19, 2019, at 8:28 AM, Koen De Groote <[hidden email]> wrote:
>
> Greetings,
>
> I've enabled JMX metrics on an install of mine and am going through the
> available beans.
>
> I've google for an explanation of what the beans all mean, but I'm coming
> up empty.
>
> For instance, these:
>
> :category=ADMIN,...,name=clientErrors,scope=/admin/file
> :category=ADMIN,...,name=clientErrors,scope=/admin/logging
> :category=ADMIN,...,name=clientErrors,scope=/admin/luke
> :category=ADMIN,...,name=clientErrors,scope=/admin/mbeans
> :category=ADMIN,...,name=clientErrors,scope=/admin/ping
> :category=ADMIN,...,name=clientErrors,scope=/admin/plugins
> :category=ADMIN,...,name=clientErrors,scope=/admin/properties
> :category=ADMIN,...,name=clientErrors,scope=/admin/segments
> :category=ADMIN,...,name=clientErrors,scope=/admin/system
> :category=ADMIN,...,name=clientErrors,scope=/admin/threads
> :category=ADMIN,...,name=clientErrors,scope=/config
> :category=ADMIN,...,name=clientErrors,scope=/schema
>
> What am I looking at here?
>
> I can't find a single page explaining what the significance of these
> metrics is.
>
> Nor can I find a page that basically says "You'll want to monitor X, Y and
> Z get a good view of your cluster".
>
> Can anyone provide an example?
>
> Kind regards,
> Koen De Groote

Reply | Threaded
Open this post in threaded view
|

Re: Explicit documentation regarding all JMX metrics?

Koen De Groote
Thanks. I also just now found
https://lucene.apache.org/solr/guide/7_6/metrics-reporting.html

I'm currently sticking to version 7.6

I've enabled JMX metrics and am accessing them via JmxTerm. There are
indeed a lot, though filtering by key=value is easy enough.

That being said, finding out what they mean is a bit harder.

Also, it seems that even the most recent documentation is out of date.

For instance here:
https://lucene.apache.org/solr/guide/8_2/performance-statistics-reference.html#commonly-used-stats-for-request-handlers

At the bottom there is mention of "avgRequestsPerSecond" and
"avgTimePerRequest", which in fact do not exist anymore. I can see them on
older versions of solr, but not on 7.6

And it seems nothing with identical meaning has taken its place.

I'll make a separate question for that.




On Tue, Nov 19, 2019 at 2:40 PM Erick Erickson <[hidden email]>
wrote:

> Here’s something that might be useful:
> https://support.lucidworks.com/s/article/What-should-you-monitor
>
> What version of Solr? For Solr 6.4+, you may want to use the
> /admin/metrics endpoint, see the reference guide for metrics. WARNING:
> there are a ton of metrics. The ref guide has information on now to trim
> down what you’re looking at.
>
> Best,
> Erick
>
> > On Nov 19, 2019, at 8:28 AM, Koen De Groote <[hidden email]>
> wrote:
> >
> > Greetings,
> >
> > I've enabled JMX metrics on an install of mine and am going through the
> > available beans.
> >
> > I've google for an explanation of what the beans all mean, but I'm coming
> > up empty.
> >
> > For instance, these:
> >
> > :category=ADMIN,...,name=clientErrors,scope=/admin/file
> > :category=ADMIN,...,name=clientErrors,scope=/admin/logging
> > :category=ADMIN,...,name=clientErrors,scope=/admin/luke
> > :category=ADMIN,...,name=clientErrors,scope=/admin/mbeans
> > :category=ADMIN,...,name=clientErrors,scope=/admin/ping
> > :category=ADMIN,...,name=clientErrors,scope=/admin/plugins
> > :category=ADMIN,...,name=clientErrors,scope=/admin/properties
> > :category=ADMIN,...,name=clientErrors,scope=/admin/segments
> > :category=ADMIN,...,name=clientErrors,scope=/admin/system
> > :category=ADMIN,...,name=clientErrors,scope=/admin/threads
> > :category=ADMIN,...,name=clientErrors,scope=/config
> > :category=ADMIN,...,name=clientErrors,scope=/schema
> >
> > What am I looking at here?
> >
> > I can't find a single page explaining what the significance of these
> > metrics is.
> >
> > Nor can I find a page that basically says "You'll want to monitor X, Y
> and
> > Z get a good view of your cluster".
> >
> > Can anyone provide an example?
> >
> > Kind regards,
> > Koen De Groote
>
>