ActiveMQ health state check

[Ephemeris Lappis <ep...@gmail.com>]

Hello.

I'm looking for the best ways to check our ActiveMQ instances.

Trying the jolokia "health" service, I thought I had a nice way to do
it, but I see that the status is ... lying.

We use a jdbc persistence configuration, and when the database
crashes, the CurrentStatus is still "Good"... I'm not sure that
messages could be processed in this situation.

What are the realm conditions that should set a distinct status ?
And why a my missing database is not detected as a probably failure
cause of the broker ?

Thanks for your help.

Regards.

Re: ActiveMQ health state check

[Ephemeris Lappis <ep...@gmail.com>]

Hello.

I'm going to create a JIRA.

I'm also going to find some other points, if any, to check the actual
state, with other Jolokia requests.

Thanks for your help.

Regards.

Le lun. 4 mars 2024 à 07:18, Jean-Baptiste Onofré <jb...@nanthrax.net> a écrit :
>
> Hi
>
> Jolokia health service only considers the BrokerService status, it
> doesn't check all layers (like persistent store or transports).
>
> I propose to create a Jira as it's possible to improve this.
> In the meantime, you can implement your own check including the
> different layers.
>
> Regards
> JB
>
> On Thu, Feb 29, 2024 at 12:53 PM Ephemeris Lappis
> <ep...@gmail.com> wrote:
> >
> > Hello.
> >
> > I'm looking for the best ways to check our ActiveMQ instances.
> >
> > Trying the jolokia "health" service, I thought I had a nice way to do
> > it, but I see that the status is ... lying.
> >
> > We use a jdbc persistence configuration, and when the database
> > crashes, the CurrentStatus is still "Good"... I'm not sure that
> > messages could be processed in this situation.
> >
> > What are the realm conditions that should set a distinct status ?
> > And why a my missing database is not detected as a probably failure
> > cause of the broker ?
> >
> > Thanks for your help.
> >
> > Regards.

Re: ActiveMQ health state check

[Jean-Baptiste Onofré <jb...@nanthrax.net>]

Hi

Jolokia health service only considers the BrokerService status, it
doesn't check all layers (like persistent store or transports).

I propose to create a Jira as it's possible to improve this.
In the meantime, you can implement your own check including the
different layers.

Regards
JB

On Thu, Feb 29, 2024 at 12:53 PM Ephemeris Lappis
<ep...@gmail.com> wrote:
>
> Hello.
>
> I'm looking for the best ways to check our ActiveMQ instances.
>
> Trying the jolokia "health" service, I thought I had a nice way to do
> it, but I see that the status is ... lying.
>
> We use a jdbc persistence configuration, and when the database
> crashes, the CurrentStatus is still "Good"... I'm not sure that
> messages could be processed in this situation.
>
> What are the realm conditions that should set a distinct status ?
> And why a my missing database is not detected as a probably failure
> cause of the broker ?
>
> Thanks for your help.
>
> Regards.