[EnMasse] Publishing to specific topic and subscribing with wildcard leads to growing queue/topic

[Bob Claerhout]

I think so, I've never created the address (nor the queue or the topic) except for the root topic called "gw".
The address is a result of an AMQP 1.0 connection from ditto to enmasse.


On 7/30/19 5:37 PM, Rob Godfrey wrote:


On Tue, 30 Jul 2019 at 13:06, Bob Claerhout <bob.claerhout at aloxy.io<mailto:bob.claerhout at aloxy.io>> wrote:
In eclipse-ditto I'm creating a connection which publishes the data to that address.
So it should probably not create the queue and just publish to that address instead.

Yeah - are you saying that ditto is doing this (creating the queue)?

-- Rob



On 7/30/19 11:52 AM, Rob Godfrey wrote:
So, I guess the question is, what is creating the queue with the specific device id in it (and which presumably no-one is consuming from)?

I would imagine that simply creating the topic does not create an queues... so, can you tell at what point this queue is being created, and by which process?

-- Rob

On Tue, 30 Jul 2019 at 11:33, Bob Claerhout <bob.claerhout at aloxy.io<mailto:bob.claerhout at aloxy.io>> wrote:
Rob,

Sorry, you're correct. The address is in fact a topic:
[cid:part3.1F0DDCD6.31A840F5 at aloxy.io]

Best regars,
Bob

On 7/30/19 11:28 AM, Rob Godfrey wrote:
I'm a little confused - this may just be imprecise use of language, but you say: "Eclipse-ditto is publishing messages to a *queue* called "gw/tenant/tenant:device" ", and then you talk about the backend service subscribing with a (topic) wildcard.

If there is a queue then it will receive all messages sent to that (precise) address.

A consumer subscribing to a wildcard is essentially setting up a second queue which will receive all messages that match the pattern (but will not receive messages that are already in the specific queue, prior to the subscription).

Is the address supposed to represent a queue (with the associated guaranteed delivery semantics) or a topic (in which case it is expected that messages sent when there are no consumers will be discarded)?

-- Rob

On Tue, 30 Jul 2019 at 11:18, Bob Claerhout <bob.claerhout at aloxy.io<mailto:bob.claerhout at aloxy.io>> wrote:
Hi,

While setting up prometheus to monitor our installation, I noticed the "artemis_message_count" on a specific address is growing while I am subscribed to this address using a wildcard.
To be more specific:
Eclipse-ditto is publishing messages to a queue called "gw/tenant/tenant:device". A backend service (python) is subscribed to "gw/#".
In prometheus, I'm getting following artemis_message_counts:
[cid:part5.C8680524.E8EF374A at aloxy.io]

When I look at the metric "artemis_consumer_count", following records are shown:
[cid:part6.B125E2D6.606848BC at aloxy.io]

Any idea on how to prevent the queue from building up?

Best regards
[cid:part7.39528678.D0D58938 at aloxy.io]  Bob Claerhout
Software developer
M +32 479 34 84 92
The Beacon, Sint-Pietersvliet 7, 2000
Antwerp
bob.claerhout at aloxy.io<mailto:bob.claerhout at aloxy.io> | www.aloxy.io<http://www.aloxy.io>
_______________________________________________
enmasse mailing list
enmasse at redhat.com<mailto:enmasse at redhat.com>
https://www.redhat.com/mailman/listinfo/enmasse


--
_____________________________________________________________________________

Red Hat GmbH, www.de.redhat.com<http://www.de.redhat.com/>,
Registered seat: Grasbrunn, Commercial register: Amtsgericht Muenchen, HRB 153243,
Managing Directors: ,Charles Cachera, Michael Cunningham, Michael O'Neill, Eric Shander




--
_____________________________________________________________________________

Red Hat GmbH, www.de.redhat.com<http://www.de.redhat.com/>,
Registered seat: Grasbrunn, Commercial register: Amtsgericht Muenchen, HRB 153243,
Managing Directors: ,Charles Cachera, Michael Cunningham, Michael O'Neill, Eric Shander




--
_____________________________________________________________________________

Red Hat GmbH, www.de.redhat.com<http://www.de.redhat.com/>,
Registered seat: Grasbrunn, Commercial register: Amtsgericht Muenchen, HRB 153243,
Managing Directors: ,Charles Cachera, Michael Cunningham, Michael O'Neill, Eric Shander


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/enmasse/attachments/20190730/be75821c/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: pjligbpiifmlpkpm.png
Type: image/png
Size: 3940 bytes
Desc: pjligbpiifmlpkpm.png
URL: <http://listman.redhat.com/archives/enmasse/attachments/20190730/be75821c/attachment.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: peihckpcocfdfmcj.png
Type: image/png
Size: 50736 bytes
Desc: peihckpcocfdfmcj.png
URL: <http://listman.redhat.com/archives/enmasse/attachments/20190730/be75821c/attachment-0001.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: bagmoinhepbbjljp.png
Type: image/png
Size: 26943 bytes
Desc: bagmoinhepbbjljp.png
URL: <http://listman.redhat.com/archives/enmasse/attachments/20190730/be75821c/attachment-0002.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mijgimmamimbonlh.png
Type: image/png
Size: 6330 bytes
Desc: mijgimmamimbonlh.png
URL: <http://listman.redhat.com/archives/enmasse/attachments/20190730/be75821c/attachment-0003.png>