You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@activemq.apache.org by "prateekjainaa@gmail.com" <pr...@gmail.com> on 2023/07/07 09:47:22 UTC
activemq | cluster | metrics
Hi All,
I am working on an application which is right now running a single
instance of activemq. Now, due to increase in load and to improve
availability; I am testing active/active cluster topology of activemq. I
am successfully running a cluster (active/active) now.
My question is, what are the key metrics that can be looked at from the
observability perspective of this cluster. Basically, I am looking at areas
that can be looked at, if I suspect that it is clustering which is causing
the issue or, to prove that clustering functionality is not at fault. Any
suggestions are welcome.
Regards,
Prateek Jain
--------------------------------------------------------------
EXPECTATION : Causes all troubles......
--------------------------------------------------------------
Re: activemq | cluster | metrics
Posted by "prateekjainaa@gmail.com" <pr...@gmail.com>.
Thanks Justin for the feedback. I will keep that in mind.
On Fri 7 Jul 2023, 20:15 Justin Bertram, <jb...@apache.org> wrote:
> The first thing I would recommend is to read the "Performance
> Considerations" [1] section of the "Clusters" chapter in the ActiveMQ
> Artemis User Manual. If you want to ensure that performance is close to
> optimal (i.e. messages are not being moved around the cluster very often)
> then you can monitor the internal store-and-forward queues which are used
> for clustering. Essentially you just want to ensure that the MessagesAdded
> metric for those queues is not continually increasing. Ideally it would
> remain at 0.
>
> Aside from that I'm not sure I can give you any other advice on
> specifically what to look for if you suspect clustering is causing a
> problem. You'll just have to deal with specific use-cases as they arise.
>
>
> Justin
>
> [1]
>
> https://activemq.apache.org/components/artemis/documentation/latest/clusters.html#performance-considerations
>
> On Fri, Jul 7, 2023 at 4:47 AM prateekjainaa@gmail.com <
> prateekjainaa@gmail.com> wrote:
>
> > Hi All,
> >
> > I am working on an application which is right now running a single
> > instance of activemq. Now, due to increase in load and to improve
> > availability; I am testing active/active cluster topology of activemq. I
> > am successfully running a cluster (active/active) now.
> >
> > My question is, what are the key metrics that can be looked at from the
> > observability perspective of this cluster. Basically, I am looking at
> areas
> > that can be looked at, if I suspect that it is clustering which is
> causing
> > the issue or, to prove that clustering functionality is not at fault. Any
> > suggestions are welcome.
> >
> >
> > Regards,
> > Prateek Jain
> >
> > --------------------------------------------------------------
> > EXPECTATION : Causes all troubles......
> > --------------------------------------------------------------
> >
>
Re: activemq | cluster | metrics
Posted by Justin Bertram <jb...@apache.org>.
The first thing I would recommend is to read the "Performance
Considerations" [1] section of the "Clusters" chapter in the ActiveMQ
Artemis User Manual. If you want to ensure that performance is close to
optimal (i.e. messages are not being moved around the cluster very often)
then you can monitor the internal store-and-forward queues which are used
for clustering. Essentially you just want to ensure that the MessagesAdded
metric for those queues is not continually increasing. Ideally it would
remain at 0.
Aside from that I'm not sure I can give you any other advice on
specifically what to look for if you suspect clustering is causing a
problem. You'll just have to deal with specific use-cases as they arise.
Justin
[1]
https://activemq.apache.org/components/artemis/documentation/latest/clusters.html#performance-considerations
On Fri, Jul 7, 2023 at 4:47 AM prateekjainaa@gmail.com <
prateekjainaa@gmail.com> wrote:
> Hi All,
>
> I am working on an application which is right now running a single
> instance of activemq. Now, due to increase in load and to improve
> availability; I am testing active/active cluster topology of activemq. I
> am successfully running a cluster (active/active) now.
>
> My question is, what are the key metrics that can be looked at from the
> observability perspective of this cluster. Basically, I am looking at areas
> that can be looked at, if I suspect that it is clustering which is causing
> the issue or, to prove that clustering functionality is not at fault. Any
> suggestions are welcome.
>
>
> Regards,
> Prateek Jain
>
> --------------------------------------------------------------
> EXPECTATION : Causes all troubles......
> --------------------------------------------------------------
>
Re: activemq | cluster | metrics
Posted by "prateekjainaa@gmail.com" <pr...@gmail.com>.
Hi Justin,
I am using a slightly older version of artemis (2.9.0). This is the
version which is used under jboss EAP version 7.3.
Regards,
Prateek Jain
--------------------------------------------------------------
EXPECTATION : Causes all troubles......
--------------------------------------------------------------
On Fri, Jul 7, 2023 at 1:28 PM Justin Bertram <jb...@apache.org> wrote:
> What version of ActiveMQ are you using?
>
>
> Justin
>
> On Fri, Jul 7, 2023 at 4:47 AM prateekjainaa@gmail.com <
> prateekjainaa@gmail.com> wrote:
>
> > Hi All,
> >
> > I am working on an application which is right now running a single
> > instance of activemq. Now, due to increase in load and to improve
> > availability; I am testing active/active cluster topology of activemq. I
> > am successfully running a cluster (active/active) now.
> >
> > My question is, what are the key metrics that can be looked at from the
> > observability perspective of this cluster. Basically, I am looking at
> areas
> > that can be looked at, if I suspect that it is clustering which is
> causing
> > the issue or, to prove that clustering functionality is not at fault. Any
> > suggestions are welcome.
> >
> >
> > Regards,
> > Prateek Jain
> >
> > --------------------------------------------------------------
> > EXPECTATION : Causes all troubles......
> > --------------------------------------------------------------
> >
>
Re: activemq | cluster | metrics
Posted by Justin Bertram <jb...@apache.org>.
What version of ActiveMQ are you using?
Justin
On Fri, Jul 7, 2023 at 4:47 AM prateekjainaa@gmail.com <
prateekjainaa@gmail.com> wrote:
> Hi All,
>
> I am working on an application which is right now running a single
> instance of activemq. Now, due to increase in load and to improve
> availability; I am testing active/active cluster topology of activemq. I
> am successfully running a cluster (active/active) now.
>
> My question is, what are the key metrics that can be looked at from the
> observability perspective of this cluster. Basically, I am looking at areas
> that can be looked at, if I suspect that it is clustering which is causing
> the issue or, to prove that clustering functionality is not at fault. Any
> suggestions are welcome.
>
>
> Regards,
> Prateek Jain
>
> --------------------------------------------------------------
> EXPECTATION : Causes all troubles......
> --------------------------------------------------------------
>
Re: activemq | cluster | metrics
Posted by "prateekjainaa@gmail.com" <pr...@gmail.com>.
Thanks a lot, Erwin! Yes, we do have a JMX based metrics setup. I will
compare and see if we are missing anything from a cluster metric
perspective.
Regards,
Prateek Jain
--------------------------------------------------------------
EXPECTATION : Causes all troubles......
--------------------------------------------------------------
On Fri, Jul 7, 2023 at 12:35 PM Dondorp, Erwin
<er...@cgi.com.invalid> wrote:
> Prateek,
>
> Due to the design of the Artemis GUI, all data that you can see there can
> be retrieved using:
> * Java JMX
> * Jolokia (a REST interface on Java JMX)
>
> Additionally, you can use a Prometheus plugin to allow data collection.
> Many other applications (e.g. all java applications) use that, or have a
> plugin to use it.
> You need a data collector application to pick up all that data and store
> it in a (Prometheus) database.
> Then use Grafana dashboards for visualization. And you can choose to build
> alerting rules on this all.
> This allows a more common approach in your landscape.
>
> https://github.com/rh-messaging/artemis-prometheus-metrics-plugin
> my system just returned the following metrics for Prometheus:
> * artemis_address_memory_usage
> * artemis_address_memory_usage_percentage
> * artemis_address_size
> * artemis_connection_count
> * artemis_consumer_count
> * artemis_delivering_durable_message_count
> * artemis_delivering_durable_persistent_size
> * artemis_delivering_message_count
> * artemis_delivering_persistent_size
> * artemis_disk_store_usage
> * artemis_durable_message_count
> * artemis_durable_persistent_size
> * artemis_message_count
> * artemis_messages_acknowledged
> * artemis_messages_added
> * artemis_messages_expired
> * artemis_messages_killed
> * artemis_number_of_pages
> * artemis_persistent_size
> * artemis_routed_message_count
> * artemis_scheduled_durable_message_count
> * artemis_scheduled_durable_persistent_size
> * artemis_scheduled_message_count
> * artemis_scheduled_persistent_size
> * artemis_total_connection_count
> * artemis_unrouted_message_count
>
> Erwin
>
> -----Original Message-----
> From: prateekjainaa@gmail.com <pr...@gmail.com>
> Sent: vrijdag 7 juli 2023 11:47
> To: users@activemq.apache.org
> Subject: activemq | cluster | metrics
>
>
> EXTERNAL SENDER: Do not click any links or open any attachments unless
> you trust the sender and know the content is safe.
> EXPÉDITEUR EXTERNE: Ne cliquez sur aucun lien et n’ouvrez aucune pièce
> jointe à moins qu’ils ne proviennent d’un expéditeur fiable, ou que vous
> ayez l'assurance que le contenu provient d'une source sûre.
>
> Hi All,
>
> I am working on an application which is right now running a single
> instance of activemq. Now, due to increase in load and to improve
> availability; I am testing active/active cluster topology of activemq. I am
> successfully running a cluster (active/active) now.
>
> My question is, what are the key metrics that can be looked at from the
> observability perspective of this cluster. Basically, I am looking at areas
> that can be looked at, if I suspect that it is clustering which is causing
> the issue or, to prove that clustering functionality is not at fault. Any
> suggestions are welcome.
>
>
> Regards,
> Prateek Jain
>
> --------------------------------------------------------------
> EXPECTATION : Causes all troubles......
> --------------------------------------------------------------
>
RE: activemq | cluster | metrics
Posted by "Dondorp, Erwin" <er...@cgi.com.INVALID>.
Prateek,
Due to the design of the Artemis GUI, all data that you can see there can be retrieved using:
* Java JMX
* Jolokia (a REST interface on Java JMX)
Additionally, you can use a Prometheus plugin to allow data collection.
Many other applications (e.g. all java applications) use that, or have a plugin to use it.
You need a data collector application to pick up all that data and store it in a (Prometheus) database.
Then use Grafana dashboards for visualization. And you can choose to build alerting rules on this all.
This allows a more common approach in your landscape.
https://github.com/rh-messaging/artemis-prometheus-metrics-plugin
my system just returned the following metrics for Prometheus:
* artemis_address_memory_usage
* artemis_address_memory_usage_percentage
* artemis_address_size
* artemis_connection_count
* artemis_consumer_count
* artemis_delivering_durable_message_count
* artemis_delivering_durable_persistent_size
* artemis_delivering_message_count
* artemis_delivering_persistent_size
* artemis_disk_store_usage
* artemis_durable_message_count
* artemis_durable_persistent_size
* artemis_message_count
* artemis_messages_acknowledged
* artemis_messages_added
* artemis_messages_expired
* artemis_messages_killed
* artemis_number_of_pages
* artemis_persistent_size
* artemis_routed_message_count
* artemis_scheduled_durable_message_count
* artemis_scheduled_durable_persistent_size
* artemis_scheduled_message_count
* artemis_scheduled_persistent_size
* artemis_total_connection_count
* artemis_unrouted_message_count
Erwin
-----Original Message-----
From: prateekjainaa@gmail.com <pr...@gmail.com>
Sent: vrijdag 7 juli 2023 11:47
To: users@activemq.apache.org
Subject: activemq | cluster | metrics
EXTERNAL SENDER: Do not click any links or open any attachments unless you trust the sender and know the content is safe.
EXPÉDITEUR EXTERNE: Ne cliquez sur aucun lien et n’ouvrez aucune pièce jointe à moins qu’ils ne proviennent d’un expéditeur fiable, ou que vous ayez l'assurance que le contenu provient d'une source sûre.
Hi All,
I am working on an application which is right now running a single instance of activemq. Now, due to increase in load and to improve availability; I am testing active/active cluster topology of activemq. I am successfully running a cluster (active/active) now.
My question is, what are the key metrics that can be looked at from the observability perspective of this cluster. Basically, I am looking at areas that can be looked at, if I suspect that it is clustering which is causing the issue or, to prove that clustering functionality is not at fault. Any suggestions are welcome.
Regards,
Prateek Jain
--------------------------------------------------------------
EXPECTATION : Causes all troubles......
--------------------------------------------------------------