Justin, thx a lot. that did it!
From: Justin Bertram
Sent: Wednesday, September 13, 2023 18:02
To: users@activemq.apache.org
Subject: Re: messages not moved to DLQ
[You don't often get email from jbert...@apache.org. Learn why this is
important at https://aka.ms
We have more than 100 senders and the problem with JDBC I see always going 1
message in size, do we have any parameter of broker.xml which we can
increase to get better performance.
PFA for reference with issue highlighted.
Thanks,
Shivang
-Original Message-
From: Justin Bertram
Sent: T
To be clear, my previous explanation assumes that nothing outside the
broker is removing the persistent data. If you are operating in a
containerized/virtual/cloud environment where the persistent data is being
deleted then clearly no data will be available when the broker restarts. If
this is the
Hi Shivang,
Try creating journals/data on PVCs. It should work.
Regards,
Prateek Jain
--
EXPECTATION : Causes all troubles..
--
On Wed, Sep 13, 2023 at 5:58 PM Shivang
You might be hitting the ephemeral nature of Kubernetes on-disk files.
See Kubernetes documentation.
https://kubernetes.io/docs/concepts/storage/volumes/
Youyu
Information Classification *General
-Original Message-
From: Justin Bertram
Sent: Wednesday, September 13, 2023 1:29 PM
To: us
If you are sending one or more durable messages in your transaction then
the broker will attempt to write them to storage when they arrive. If the
broker fails to write the messages, for whatever reason, then the sender
will receive an exception when it attempts to commit the transaction. If
the br
The issue definitely seems to be in the 'large Core message'->AMQP
conversion, so for now the only way to avoid it until fixed would seem
to be to avoid that process.
For MQTT senders as you are using, it looks like perhaps the messages
are treated as 'large' if they are greater than the minimum o
Hi Justin,
We are using Artemis docker image and start kubernetes pods with it. We have
one sender which will write messages on queue and one receiver which will
read messages queue.
Now due to any reason, kubernetes queue pod gets restarted so before
restarts whatever transactions gets enqueued b
I'm not really sure what you're asking. Are you asking whether you should
use the file-based journal or a database if you have 100k transactions?
To be clear, what is "best" in one situation is often not "best" in
another. Everything depends on the specifics of your particular use-case.
Justin
If scenario is no loss transactions 100% if queue goes down whatever
transactions gets enqueued, should get dequeued once queue comes up, we have
100k transactions or more need to flow up via queue. What would be best in
such scenarios?
Thanks,
Shivang
-Original Message-
From: Justin Bert
In order to register the delivery attempt you need to acknowledge the
message, e.g.:
while(true)
{
System.out.println("waiting for message");
ClientMessage rec = consumer.receive(5000);
if(rec == null) break;
rec.acknowledge();
System.out.println(rec.getStringProp
I dont know know about that issue, no. Seems like it happens due to
converting from a 'large' Core message (which I believe is used for
MQTT too) into an AMQP message. That would happen after it was on the
queueso of course the publisher got an ack. The exception seems to
indicate the conversion fa
When deciding between the file-based journal on local storage versus a
remote database I think the three main considerations are:
- Performance
- Infrastructure
- Reliability
The file-based journal on local storage will be faster than a database for
a few reasons:
- The storage is local so th
Thank you Robbie, upgrading Camel to the latest 3.x available fixed the problem
but I experienced messages loss.
The broker log showed exactly the stack trace of this issue:
https://issues.apache.org/jira/projects/ARTEMIS/issues/ARTEMIS-4217
Unfortunately I didn’t detect a cause for this. It hap
Hi Team,
Can anyone share pros and cons in depth between both. I see only file
storage is faster than JDBC storage. Is there any disadvantage of File
Storage like losing the enqueued data or anything?
Thanks,
Shivang.
--
*This e-mail contains PRIVILEGED AND CONFIDENTIAL INFORMATION intende
Hi all.
I'm trying to test the DLQ function in artemis V 2.30.0 but i can't get it
working. My setup is like so:
running artemis inside docker container with default configuration. According
to the docs the
max-delivery-attempts is by default 10. below is a little test programm written
in java
16 matches
Mail list logo
