Re: Time for another ASF board report

[Daniel Kulp]

Bruce,

> On Oct 18, 2017, at 2:04 PM, Bruce Snyder  wrote:
> 
> There was a comment regarding Artemis added to the board report that
> mentioned the following:
> 
> 'An embedded web management console was added'
> 
> What embedded web management console is this? Was it built from the ground
> up or was it adopted from something else?

It uses hawt.io as a framework, but there were extensive requirements set forth 
that we able to be met with the newer setup.   See the summary at:

https://lists.apache.org/thread.html/ea23d51f254690a905caf064cbd11610773dc3dc9015c10319b7d826@%3Cdev.activemq.apache.org%3E

And the links in that email.   Basically, all the hawt.io branding is removed 
and replaced with ActiveMQ branding, all of stuff not related to ActiveMQ is 
removed/hidden, and all of the ActiveMQ/Artemis functionally is provided by 
plugins created and maintained here at Apache.  



Dan



> Bruce
> 
> 
> On Tue, Oct 10, 2017 at 9:28 AM, Bruce Snyder 
> wrote:
> 
>> Thank you for taking the time, Robbie.
>> 
>> Bruce
>> 
>> On Fri, Oct 6, 2017 at 9:18 AM, Robbie Gemmell 
>> wrote:
>> 
>>> I've added some initial notes.
>>> 
>>> Robbie
>>> 
>>> On 5 October 2017 at 15:54, Bruce Snyder  wrote:
 It is time to create a report for the ASF board for October. Please
>>> take 5
 minutes and contribute to the report available at the following URL:
 
 https://cwiki.apache.org/confluence/pages/viewpage.action?
>>> pageId=74681966
 
 Please contribute any project activity from any of the ActiveMQ
>>> subprojects.
 
 This report will be submitted on 10 Oct.
 
 Bruce
 --
 perl -e 'print
 unpack("u30","D0G)U8V4\@4VYY9&5R\"F)R=6-E+G-N>61E>> );'
 
 ActiveMQ in Action: http://bit.ly/2je6cQ
 Blog: http://bsnyder.org/ 
 Twitter: http://twitter.com/brucesnyder
>>> 
>> 
>> 
>> 
>> --
>> perl -e 'print unpack("u30","D0G)U8V4\@4VYY9&
>> 5R\"F)R=6-E+G-N>61E> 
>> ActiveMQ in Action: http://bit.ly/2je6cQ
>> Blog: http://bsnyder.org/ 
>> Twitter: http://twitter.com/brucesnyder
>> 
> 
> 
> 
> -- 
> perl -e 'print
> unpack("u30","D0G)U8V4\@4VYY9&5R\"F)R=6-E+G-N>61E 
> ActiveMQ in Action: http://bit.ly/2je6cQ
> Blog: http://bsnyder.org/ 
> Twitter: http://twitter.com/brucesnyder

-- 
Daniel Kulp
dk...@apache.org - http://dankulp.com/blog
Talend Community Coder - http://coders.talend.com

Re: [VOTE] Apache ActiveMQ 5.15.2

[W B D]

+1 (non-binding)

Upgraded from 5.15.1 to 5.15.2 in our staging environment; no issues, but
then again 5.15.1 was also working fine for us. Using the auto transport.
Clients tested were an ActiveMQ 5.14.1 Java client and a C# client using
NMS ActiveMQ 1.7.2. The NMS client sometimes takes up to 30 seconds to
connect, but I don't think that's new, and is not specific to the auto
transport. Also tested with hawtio (dropping in the latest 1.5.4, default
offline version) - all in all looks good.

Regards,
Bruce D

On Wed, Oct 18, 2017 at 3:57 AM, Jean-Baptiste Onofré 
wrote:

> +1 (binding)
>
> Regards
> JB
>
> On Oct 18, 2017, 12:56, at 12:56, Robbie Gemmell 
> wrote:
> >On 17 October 2017 at 15:37, Christopher Shannon
> > wrote:
> >> Hi Everyone,
> >>
> >> I have created the ActiveMQ 5.15.2 release and it's ready for a vote.
> >>
> >> The list of resolved issues is here:
> >>
> >https://issues.apache.org/jira/secure/ReleaseNote.jspa?
> projectId=12311210=12341669
> >>
> >> You can get the release artifacts here:
> >> https://dist.apache.org/repos/dist/dev/activemq/activemq/5.15.2/
> >>
> >> Maven repository is at:
> >>
> >https://repository.apache.org/content/repositories/
> orgapacheactivemq-1151/
> >>
> >> Source tag:
> >https://git-wip-us.apache.org/repos/asf?p=activemq.git;a=commit;h=
> 9e595d8674456810ee3bd2d4a9920ed8b680937b
> >>
> >> Please vote to approve this release.  The vote will remain open for
> >72 hours.
> >>
> >> [ ] +1 Release the binary as Apache ActiveMQ 5.15.2
> >> [ ] -1 (provide specific comments)
> >>
> >> Here's my +1
> >
> >+1 (non-binding)
> >
> >I checked things over as follows:
> >- Verified the signature and checksum files.
> >- Checked licence and notice files are present in the archives.
> >- Ran the source build and tested with the sanity test profile.
> >- Used the broker from the tar.gz binary to run some AMQP client
> >examples.
> >- Used the staging repo to run the Qpid JMS master tests against the
> >broker.
> >
> >Robbie
>

[GitHub] activemq-artemis issue #1594: ARTEMIS-1467 example showing the recommended a...

[jbertram]

Github user jbertram commented on the issue:

https://github.com/apache/activemq-artemis/pull/1594
  
I believe Clebert is referring to [this 
tool](https://github.com/apache/activemq-cli-tools) which can be used to export 
from 5.x and then the resulting XML file can be imported via the Artemis CLI.  
Using these tools might be a lot simpler than the method outlined in the 
current PR as it can be used for all messages in all queues simultaneously 
rather than having to create a bridge for each queue.


---

[GitHub] activemq-artemis pull request #1599: ARTEMIS-1470 - fix console move message

[asfgit]

Github user asfgit closed the pull request at:

https://github.com/apache/activemq-artemis/pull/1599


---

[GitHub] activemq-artemis pull request #1593: ARTEMIS-1463 - added SSL example + docs

[asfgit]

Github user asfgit closed the pull request at:

https://github.com/apache/activemq-artemis/pull/1593


---

[GitHub] activemq-artemis issue #1598: ARTEMIS-1469 - Add support for trusting any cl...

[jbertram]

Github user jbertram commented on the issue:

https://github.com/apache/activemq-artemis/pull/1598
  
Even though it's only for testing it would still be good to have a little 
documentation on the option.


---

[GitHub] activemq pull request #262: AMQ-6828: Improvement to python AMQP example.

[asfgit]

Github user asfgit closed the pull request at:

https://github.com/apache/activemq/pull/262


---

Re: Time for another ASF board report

[Bruce Snyder]

There was a comment regarding Artemis added to the board report that
mentioned the following:

'An embedded web management console was added'

What embedded web management console is this? Was it built from the ground
up or was it adopted from something else?

Bruce


On Tue, Oct 10, 2017 at 9:28 AM, Bruce Snyder 
wrote:

> Thank you for taking the time, Robbie.
>
> Bruce
>
> On Fri, Oct 6, 2017 at 9:18 AM, Robbie Gemmell 
> wrote:
>
>> I've added some initial notes.
>>
>> Robbie
>>
>> On 5 October 2017 at 15:54, Bruce Snyder  wrote:
>> > It is time to create a report for the ASF board for October. Please
>> take 5
>> > minutes and contribute to the report available at the following URL:
>> >
>> > https://cwiki.apache.org/confluence/pages/viewpage.action?
>> pageId=74681966
>> >
>> > Please contribute any project activity from any of the ActiveMQ
>> subprojects.
>> >
>> > This report will be submitted on 10 Oct.
>> >
>> > Bruce
>> > --
>> > perl -e 'print
>> > unpack("u30","D0G)U8V4\@4VYY9&5R\"F)R=6-E+G-N>61E> );'
>> >
>> > ActiveMQ in Action: http://bit.ly/2je6cQ
>> > Blog: http://bsnyder.org/ 
>> > Twitter: http://twitter.com/brucesnyder
>>
>
>
>
> --
> perl -e 'print unpack("u30","D0G)U8V4\@4VYY9&
> 5R\"F)R=6-E+G-N>61E
> ActiveMQ in Action: http://bit.ly/2je6cQ
> Blog: http://bsnyder.org/ 
> Twitter: http://twitter.com/brucesnyder
>



-- 
perl -e 'print
unpack("u30","D0G)U8V4\@4VYY9&5R\"F)R=6-E+G-N>61E
Twitter: http://twitter.com/brucesnyder

[GitHub] activemq-artemis issue #1592: NO-JIRA improve MQTT protocol logging

[jbertram]

Github user jbertram commented on the issue:

https://github.com/apache/activemq-artemis/pull/1592
  
I just pushed the latest iteration with the suggestion from @franz1981 
about using a new method as well as removing the system property and enforcing 
a hard limit of 256 characters when printing the payload.


---

[GitHub] activemq-artemis issue #1596: ARTEMIS-450 Deadlocked queues over AMQP after ...

[clebertsuconic]

Github user clebertsuconic commented on the issue:

https://github.com/apache/activemq-artemis/pull/1596
  
@franz1981 lets not waste time.. this is such a small component... you 
added a new parameter that was clashing with previous usage.. all I'm doing is 
fix it accordingly to the previous version after the tests you broke.


---

[GitHub] activemq-artemis pull request #1596: ARTEMIS-450 Deadlocked queues over AMQP...

[asfgit]

Github user asfgit closed the pull request at:

https://github.com/apache/activemq-artemis/pull/1596


---

[GitHub] activemq-artemis issue #1596: ARTEMIS-450 Deadlocked queues over AMQP after ...

[clebertsuconic]

Github user clebertsuconic commented on the issue:

https://github.com/apache/activemq-artemis/pull/1596
  
@franz1981 nope.. you broke a test Franz.


---

[GitHub] activemq-artemis issue #1596: ARTEMIS-450 Deadlocked queues over AMQP after ...

[franz1981]

Github user franz1981 commented on the issue:

https://github.com/apache/activemq-artemis/pull/1596
  
@clebertsuconic I do not agree with that: the change I did was validated 
with tests and the documentation says that when the initial delay isn't 
specified is defaulted to period, so the expectation on it is that it will be 
defaulted to period in a `getInitialDelay` too.
As I said, for me it's the same but IMO if you want to add a new semantic 
over a null initialDelay you need to specify it in some way (ie docs/tests 
and/or even better, on the API) in order to avoid subtle bugs and/or to avoid 
to change tests that are only checking a get after a set.



---

[GitHub] activemq-artemis issue #1596: ARTEMIS-450 Deadlocked queues over AMQP after ...

[clebertsuconic]

Github user clebertsuconic commented on the issue:

https://github.com/apache/activemq-artemis/pull/1596
  
using your own words.. the least astonishment way would be to have the 
initialDelay nullable (or -1 on this case) I'm fixing what was broken 
before.


---

[GitHub] activemq-artemis issue #1596: ARTEMIS-450 Deadlocked queues over AMQP after ...

[clebertsuconic]

Github user clebertsuconic commented on the issue:

https://github.com/apache/activemq-artemis/pull/1596
  
@franz1981 It's just that I changed the meaning of getInitialDelay... -1 
meaning unset.. and use the previous semantic...  the change you did broke some 
usage.. so I had to have a null state for initialDelay where I would get the 
previous semantic... that's all


---

[GitHub] activemq-artemis issue #1596: ARTEMIS-450 Deadlocked queues over AMQP after ...

[franz1981]

Github user franz1981 commented on the issue:

https://github.com/apache/activemq-artemis/pull/1596
  
@clebertsuconic It is not following the [principle of least 
astonishment](https://en.wikipedia.org/wiki/Principle_of_least_astonishment) 
from the API (and doc) perspective: if you remove the `getInitialDelay`probably 
you could use safetly that change, but considering that all the other 
properties are exposed, probably it is not the right choice.
That's my 2 cents on it: I'm happy anyway that the functionality isn't 
broken (for my usage I mean) :+1: 


---

[GitHub] activemq-artemis issue #1596: ARTEMIS-450 Deadlocked queues over AMQP after ...

[franz1981]

Github user franz1981 commented on the issue:

https://github.com/apache/activemq-artemis/pull/1596
  
@clebertsuconic It is a get of field after a set of the same that is not 
getting the configured/expected value: it is not what a user could expect.


---

[GitHub] activemq-artemis issue #1596: ARTEMIS-450 Deadlocked queues over AMQP after ...

[clebertsuconic]

Github user clebertsuconic commented on the issue:

https://github.com/apache/activemq-artemis/pull/1596
  
your test wasn’t validating semantics.  The change is good. 


---

[GitHub] activemq-artemis pull request #1599: ARTEMIS-1470 - fix console move message

[andytaylor]

GitHub user andytaylor opened a pull request:

https://github.com/apache/activemq-artemis/pull/1599

ARTEMIS-1470 - fix console move message

https://issues.apache.org/jira/browse/ARTEMIS-1470

You can merge this pull request into a Git repository by running:

$ git pull https://github.com/andytaylor/activemq-artemis movemessage

Alternatively you can review and apply these changes as the patch at:

https://github.com/apache/activemq-artemis/pull/1599.patch

To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:

This closes #1599


commit a36459a283b7763f0bed670bd48da23a846312de
Author: Andy Taylor 
Date:   2017-10-18T15:41:38Z

ARTEMIS-1470 - fix console move message

https://issues.apache.org/jira/browse/ARTEMIS-1470




---

[GitHub] activemq-artemis pull request #1596: ARTEMIS-450 Deadlocked queues over AMQP...

[franz1981]

Github user franz1981 commented on a diff in the pull request:

https://github.com/apache/activemq-artemis/pull/1596#discussion_r145455314
  
--- Diff: 
artemis-commons/src/main/java/org/apache/activemq/artemis/core/server/ActiveMQScheduledComponent.java
 ---
@@ -90,7 +90,7 @@ public 
ActiveMQScheduledComponent(ScheduledExecutorService scheduledExecutorServ
  long checkPeriod,
  TimeUnit timeUnit,
  boolean onDemand) {
-  this(scheduledExecutorService, executor, checkPeriod, checkPeriod, 
timeUnit, onDemand);
+  this(scheduledExecutorService, executor, -1, checkPeriod, timeUnit, 
onDemand);
--- End diff --

I'm receiving an error in the test `testVerifyDefaultInitialDelay`:
```
java.lang.AssertionError: The initial delay must be defaulted to the period 
Expected :100
Actual   :-1
```
Modifying things like this (and leaving the constructor as it is) doesn't 
break any tests:
```
   // this will restart the scheduled component upon changes
   private void restartIfNeeded() {
  if (isStarted()) {
 stop();
 //do not need to start with the initialDelay: the component was 
already running
 start(this.period);
  }
   }

   private void start(final long initialDelay) {
  if (future != null) {
 // already started
 return;
  }

  if (scheduledExecutorService == null) {
 scheduledExecutorService = new ScheduledThreadPoolExecutor(1, 
getThreadFactory());
 startedOwnScheduler = true;

  }

  if (onDemand) {
 return;
  }

  this.millisecondsPeriod = timeUnit.convert(period, 
TimeUnit.MILLISECONDS);

  if (period >= 0) {
 future = 
scheduledExecutorService.scheduleWithFixedDelay(runForScheduler, initialDelay, 
period, timeUnit);
  } else {
 logger.tracef("did not start scheduled executor on %s because 
period was configured as %d", this, period);
  }
   }

   @Override
   public synchronized void start() {
  start(this.initialDelay);
   }
```


---

[GitHub] activemq-artemis pull request #1592: NO-JIRA improve MQTT protocol logging

[ppatierno]

Github user ppatierno commented on a diff in the pull request:

https://github.com/apache/activemq-artemis/pull/1592#discussion_r145441606
  
--- Diff: 
artemis-protocols/artemis-mqtt-protocol/src/main/java/org/apache/activemq/artemis/core/protocol/mqtt/MQTTUtil.java
 ---
@@ -67,6 +76,8 @@
 
public static final int DEFAULT_KEEP_ALIVE_FREQUENCY = 5000;
 
+   public static final String logUTF8PublishPayload = 
System.getProperty("org.apache.activemq.artemis.core.protocol.mqtt.logUTF8PublishPayload");
--- End diff --

So for people who need to have the entire payload because it's in a 
reasonable range size, they need to use Wireshark ?


---

[GitHub] activemq-artemis pull request #1592: NO-JIRA improve MQTT protocol logging

[tabish121]

Github user tabish121 commented on a diff in the pull request:

https://github.com/apache/activemq-artemis/pull/1592#discussion_r145438163
  
--- Diff: 
artemis-protocols/artemis-mqtt-protocol/src/main/java/org/apache/activemq/artemis/core/protocol/mqtt/MQTTUtil.java
 ---
@@ -67,6 +76,8 @@
 
public static final int DEFAULT_KEEP_ALIVE_FREQUENCY = 5000;
 
+   public static final String logUTF8PublishPayload = 
System.getProperty("org.apache.activemq.artemis.core.protocol.mqtt.logUTF8PublishPayload");
--- End diff --

We do that in 5.x for things like logs on STOMP messages where you don't 
want to dump the full content portion of the frame but a small bit can help, I 
can't recall what the default is but something like 60 chars or the like is 
used.  You end up with a content string of "ABCDZYZ"


---

[GitHub] activemq-artemis pull request #1592: NO-JIRA improve MQTT protocol logging

[jbertram]

Github user jbertram commented on a diff in the pull request:

https://github.com/apache/activemq-artemis/pull/1592#discussion_r145437240
  
--- Diff: 
artemis-protocols/artemis-mqtt-protocol/src/main/java/org/apache/activemq/artemis/core/protocol/mqtt/MQTTUtil.java
 ---
@@ -67,6 +76,8 @@
 
public static final int DEFAULT_KEEP_ALIVE_FREQUENCY = 5000;
 
+   public static final String logUTF8PublishPayload = 
System.getProperty("org.apache.activemq.artemis.core.protocol.mqtt.logUTF8PublishPayload");
--- End diff --

To keep this as simple as possible I'll get rid of the system property and 
always log the payload as UTF-8, but also always limit the output to a certain 
number of characters so we don't get hundreds of megs of data (potentially).


---

[GitHub] activemq-artemis pull request #1592: NO-JIRA improve MQTT protocol logging

[ppatierno]

Github user ppatierno commented on a diff in the pull request:

https://github.com/apache/activemq-artemis/pull/1592#discussion_r145434452
  
--- Diff: 
artemis-protocols/artemis-mqtt-protocol/src/main/java/org/apache/activemq/artemis/core/protocol/mqtt/MQTTUtil.java
 ---
@@ -67,6 +76,8 @@
 
public static final int DEFAULT_KEEP_ALIVE_FREQUENCY = 5000;
 
+   public static final String logUTF8PublishPayload = 
System.getProperty("org.apache.activemq.artemis.core.protocol.mqtt.logUTF8PublishPayload");
--- End diff --

My 2 cents. The PUBLISH payload has a maximum size of 256 MB. I could have 
a use case where I need tracing packets but not printing 256 MB of bytes on the 
console. Use case where could be a problem in the messages flow that is not 
dependent on the payload. In this moment using just TRACE enabled or not it's 
just about having or not the logging. Maybe we could use DEBUG for normal 
packets logging and then TRACE for printing the payload as well.


---

[GitHub] activemq-artemis issue #1596: ARTEMIS-450 Deadlocked queues over AMQP after ...

[clebertsuconic]

Github user clebertsuconic commented on the issue:

https://github.com/apache/activemq-artemis/pull/1596
  
@franz1981 I'm using an initialDelay = -1 by default now, and added some 
code around on start.


---

[GitHub] activemq-artemis pull request #1598: ARTEMIS-1469 - Add support for trusting...

[cshannon]

GitHub user cshannon opened a pull request:

https://github.com/apache/activemq-artemis/pull/1598

ARTEMIS-1469 - Add support for trusting any client certificate

Added a new trustAll flag which will support trusting any client
keystore when doing testing against a broker.  This setting should not
be used in production and is strictly for testing.

You can merge this pull request into a Git repository by running:

$ git pull https://github.com/cshannon/activemq-artemis ARTEMIS-1469

Alternatively you can review and apply these changes as the patch at:

https://github.com/apache/activemq-artemis/pull/1598.patch

To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:

This closes #1598


commit 24519bbcff5e2b9a9542934696c4034ceea32790
Author: Christopher L. Shannon (cshannon) 
Date:   2017-10-18T14:07:23Z

ARTEMIS-1469 - Add support for trusting any client certificate

Added a new trustAll flag which will support trusting any client
keystore when doing testing against a broker.  This setting should not
be used in production and is strictly for testing.




---

[GitHub] activemq-artemis pull request #1596: ARTEMIS-450 Deadlocked queues over AMQP...

[franz1981]

Github user franz1981 commented on a diff in the pull request:

https://github.com/apache/activemq-artemis/pull/1596#discussion_r145425664
  
--- Diff: 
artemis-commons/src/main/java/org/apache/activemq/artemis/core/server/ActiveMQScheduledComponent.java
 ---
@@ -251,6 +251,9 @@ public synchronized boolean isStarted() {
// this will restart the scheduled component upon changes
private void restartIfNeeded() {
   if (isStarted()) {
+ // it has already been through an initial delay,
+ // now we just use the next interval
+ this.initialDelay = period;
--- End diff --

@clebertsuconic When I have a very big period but I need the first call to 
happen right after the start of the component. IMO a user expectation is that 
the component start will honour the configuration provided if the componet was 
in a stopped state...


---

[GitHub] activemq-artemis pull request #1596: ARTEMIS-450 Deadlocked queues over AMQP...

[clebertsuconic]

Github user clebertsuconic commented on a diff in the pull request:

https://github.com/apache/activemq-artemis/pull/1596#discussion_r145424777
  
--- Diff: 
artemis-commons/src/main/java/org/apache/activemq/artemis/core/server/ActiveMQScheduledComponent.java
 ---
@@ -251,6 +251,9 @@ public synchronized boolean isStarted() {
// this will restart the scheduled component upon changes
private void restartIfNeeded() {
   if (isStarted()) {
+ // it has already been through an initial delay,
+ // now we just use the next interval
+ this.initialDelay = period;
--- End diff --

What do you need an initialDelay different from the frequency anyways? What 
is the usecase?


---

[GitHub] activemq-artemis pull request #1596: ARTEMIS-450 Deadlocked queues over AMQP...

[franz1981]

Github user franz1981 commented on a diff in the pull request:

https://github.com/apache/activemq-artemis/pull/1596#discussion_r145417351
  
--- Diff: 
artemis-commons/src/main/java/org/apache/activemq/artemis/core/server/ActiveMQScheduledComponent.java
 ---
@@ -251,6 +251,9 @@ public synchronized boolean isStarted() {
// this will restart the scheduled component upon changes
private void restartIfNeeded() {
   if (isStarted()) {
+ // it has already been through an initial delay,
+ // now we just use the next interval
+ this.initialDelay = period;
--- End diff --

@clebertsuconic it is "initial" on each start after a stop (eg how i use it 
of JDBC)


---

[GitHub] activemq-artemis pull request #1596: ARTEMIS-450 Deadlocked queues over AMQP...

[franz1981]

Github user franz1981 commented on a diff in the pull request:

https://github.com/apache/activemq-artemis/pull/1596#discussion_r145416958
  
--- Diff: 
artemis-commons/src/main/java/org/apache/activemq/artemis/core/server/ActiveMQScheduledComponent.java
 ---
@@ -251,6 +251,9 @@ public synchronized boolean isStarted() {
// this will restart the scheduled component upon changes
private void restartIfNeeded() {
   if (isStarted()) {
+ // it has already been through an initial delay,
+ // now we just use the next interval
+ this.initialDelay = period;
--- End diff --

@clebertsuconic But if (at the end of the test) I will start the component, 
it will have a changed configuration...it can lead to subtle bugs...


---

[GitHub] activemq-artemis pull request #1593: ARTEMIS-1463 - added SSL example + docs

[andytaylor]

Github user andytaylor commented on a diff in the pull request:

https://github.com/apache/activemq-artemis/pull/1593#discussion_r145416142
  
--- Diff: 
examples/features/standard/message-counters/src/main/java/org/apache/activemq/artemis/jms/example/MessageCounterExample.java
 ---
@@ -75,9 +75,11 @@ public static void main(final String[] args) throws 
Exception {
 
  // Step 7. Use JMX to retrieve the message counters using the 
JMSQueueControl
  ObjectName on = 
ObjectNameBuilder.DEFAULT.getQueueObjectName(SimpleString.toSimpleString(queue.getQueueName()),
 SimpleString.toSimpleString(queue.getQueueName()), RoutingType.ANYCAST);
+ //we dont actually need credentials as the guest login i sused 
but this is how its done
  HashMap env = new HashMap();
  String[] creds = {"guest", "guest"};
  env.put(JMXConnector.CREDENTIALS, creds);
+ env.put(JMXConnector.CREDENTIALS, creds);
--- End diff --

fixed


---

[GitHub] activemq-artemis pull request #1596: ARTEMIS-450 Deadlocked queues over AMQP...

[clebertsuconic]

Github user clebertsuconic commented on a diff in the pull request:

https://github.com/apache/activemq-artemis/pull/1596#discussion_r145416018
  
--- Diff: 
artemis-commons/src/main/java/org/apache/activemq/artemis/core/server/ActiveMQScheduledComponent.java
 ---
@@ -251,6 +251,9 @@ public synchronized boolean isStarted() {
// this will restart the scheduled component upon changes
private void restartIfNeeded() {
   if (isStarted()) {
+ // it has already been through an initial delay,
+ // now we just use the next interval
+ this.initialDelay = period;
--- End diff --

Nope. Initial delay should be at the beginnnjng only.  Why would you want a 
further initial delay when it’s not initial any longer.  Just to keep it 
simple.  


---

[GitHub] activemq-artemis pull request #1593: ARTEMIS-1463 - added SSL example + docs

[andytaylor]

Github user andytaylor commented on a diff in the pull request:

https://github.com/apache/activemq-artemis/pull/1593#discussion_r145415855
  
--- Diff: 
examples/features/standard/message-counters/src/main/java/org/apache/activemq/artemis/jms/example/MessageCounterExample.java
 ---
@@ -75,9 +75,11 @@ public static void main(final String[] args) throws 
Exception {
 
  // Step 7. Use JMX to retrieve the message counters using the 
JMSQueueControl
  ObjectName on = 
ObjectNameBuilder.DEFAULT.getQueueObjectName(SimpleString.toSimpleString(queue.getQueueName()),
 SimpleString.toSimpleString(queue.getQueueName()), RoutingType.ANYCAST);
+ //we dont actually need credentials as the guest login i sused 
but this is how its done
  HashMap env = new HashMap();
  String[] creds = {"guest", "guest"};
  env.put(JMXConnector.CREDENTIALS, creds);
+ env.put(JMXConnector.CREDENTIALS, creds);
--- End diff --

yes :)


---

[GitHub] activemq-artemis pull request #1596: ARTEMIS-450 Deadlocked queues over AMQP...

[franz1981]

Github user franz1981 commented on a diff in the pull request:

https://github.com/apache/activemq-artemis/pull/1596#discussion_r145411963
  
--- Diff: 
artemis-commons/src/main/java/org/apache/activemq/artemis/core/server/ActiveMQScheduledComponent.java
 ---
@@ -251,6 +251,9 @@ public synchronized boolean isStarted() {
// this will restart the scheduled component upon changes
private void restartIfNeeded() {
   if (isStarted()) {
+ // it has already been through an initial delay,
+ // now we just use the next interval
+ this.initialDelay = period;
--- End diff --

I've built up this test:
```
   @Test
   public void testVerifyInitialDelayChanged() {
  final long initialDelay = 10;
  final long period = 100;
  final ActiveMQScheduledComponent local = new 
ActiveMQScheduledComponent(scheduledExecutorService, executorService, 
initialDelay, period, TimeUnit.MILLISECONDS, false) {
 @Override
 public void run() {

 }
  };
  local.start();
  final long newInitialDelay = 1000;
  //the parameters are valid?
  assert initialDelay != newInitialDelay && newInitialDelay != period;
  local.setInitialDelay(newInitialDelay);
  local.stop();
  Assert.assertEquals("the initial dalay can't change", 
newInitialDelay, local.getInitialDelay());
   }
```
And it fail: maybe is better to create a `start(long forceInitialDelay)` 
method and call it into the restart, wdyt?



---

[GitHub] activemq-artemis pull request #1593: ARTEMIS-1463 - added SSL example + docs

[clebertsuconic]

Github user clebertsuconic commented on a diff in the pull request:

https://github.com/apache/activemq-artemis/pull/1593#discussion_r145409876
  
--- Diff: 
examples/features/standard/message-counters/src/main/java/org/apache/activemq/artemis/jms/example/MessageCounterExample.java
 ---
@@ -75,9 +75,11 @@ public static void main(final String[] args) throws 
Exception {
 
  // Step 7. Use JMX to retrieve the message counters using the 
JMSQueueControl
  ObjectName on = 
ObjectNameBuilder.DEFAULT.getQueueObjectName(SimpleString.toSimpleString(queue.getQueueName()),
 SimpleString.toSimpleString(queue.getQueueName()), RoutingType.ANYCAST);
+ //we dont actually need credentials as the guest login i sused 
but this is how its done
  HashMap env = new HashMap();
  String[] creds = {"guest", "guest"};
  env.put(JMXConnector.CREDENTIALS, creds);
+ env.put(JMXConnector.CREDENTIALS, creds);
--- End diff --

Was this a mistake? the double line here.


---

[GitHub] activemq-artemis pull request #1597: ARTEMIS-1468 Fix delete address in cons...

[asfgit]

Github user asfgit closed the pull request at:

https://github.com/apache/activemq-artemis/pull/1597


---

[GitHub] activemq-artemis pull request #1596: ARTEMIS-450 Deadlocked queues over AMQP...

[franz1981]

Github user franz1981 commented on a diff in the pull request:

https://github.com/apache/activemq-artemis/pull/1596#discussion_r145408498
  
--- Diff: 
artemis-commons/src/main/java/org/apache/activemq/artemis/core/server/ActiveMQScheduledComponent.java
 ---
@@ -251,6 +251,9 @@ public synchronized boolean isStarted() {
// this will restart the scheduled component upon changes
private void restartIfNeeded() {
   if (isStarted()) {
+ // it has already been through an initial delay,
+ // now we just use the next interval
+ this.initialDelay = period;
--- End diff --

nice! :+1: 


---

[GitHub] activemq-artemis pull request #1592: NO-JIRA improve MQTT protocol logging

[clebertsuconic]

Github user clebertsuconic commented on a diff in the pull request:

https://github.com/apache/activemq-artemis/pull/1592#discussion_r145407743
  
--- Diff: 
artemis-protocols/artemis-mqtt-protocol/src/main/java/org/apache/activemq/artemis/core/protocol/mqtt/MQTTUtil.java
 ---
@@ -67,6 +76,8 @@
 
public static final int DEFAULT_KEEP_ALIVE_FREQUENCY = 5000;
 
+   public static final String logUTF8PublishPayload = 
System.getProperty("org.apache.activemq.artemis.core.protocol.mqtt.logUTF8PublishPayload");
--- End diff --

this is about tracing.. why don't you just always log the UTF?
if you don't need that level of information, just take trace out.


of you could do something like.. have this whole thing at debug... and if 
trace, throw the body on log.trace.

The system property is a bit odd here. 


---

[GitHub] activemq-artemis pull request #1592: NO-JIRA improve MQTT protocol logging

[clebertsuconic]

Github user clebertsuconic commented on a diff in the pull request:

https://github.com/apache/activemq-artemis/pull/1592#discussion_r145407343
  
--- Diff: 
artemis-protocols/artemis-mqtt-protocol/src/main/java/org/apache/activemq/artemis/core/protocol/mqtt/MQTTUtil.java
 ---
@@ -149,16 +160,62 @@ public static void logMessage(MQTTSessionState state, 
MqttMessage message, boole
  if (message.fixedHeader() != null) {
 log.append(message.fixedHeader().messageType().toString());
 
-if (message.variableHeader() instanceof 
MqttPublishVariableHeader) {
-   log.append("(" + ((MqttPublishVariableHeader) 
message.variableHeader()).messageId() + ") " + 
message.fixedHeader().qosLevel());
-} else if (message.variableHeader() instanceof 
MqttMessageIdVariableHeader) {
+if (message.variableHeader() instanceof 
MqttMessageIdVariableHeader) {
log.append("(" + ((MqttMessageIdVariableHeader) 
message.variableHeader()).messageId() + ")");
 }
 
-if (message.fixedHeader().messageType() == 
MqttMessageType.SUBSCRIBE) {
-   for (MqttTopicSubscription sub : ((MqttSubscribeMessage) 
message).payload().topicSubscriptions()) {
-  log.append("\n\t" + sub.topicName() + " : " + 
sub.qualityOfService());
-   }
+switch (message.fixedHeader().messageType()) {
+   case PUBLISH:
+  MqttPublishVariableHeader publishHeader = 
(MqttPublishVariableHeader) message.variableHeader();
+  log.append("(" + publishHeader.packetId() + ")")
+ .append(" topic=" + publishHeader.topicName())
+ .append(", qos=" + message.fixedHeader().qosLevel())
+ .append(", retain=" + 
message.fixedHeader().isRetain())
+ .append(", dup=" + message.fixedHeader().isDup());
+  if (logUTF8PublishPayload != null) {
+ log.append(" payload=" + 
((MqttPublishMessage)message).payload().toString(StandardCharsets.UTF_8));
+  }
+  break;
+   case CONNECT:
+  MqttConnectVariableHeader connectHeader = 
(MqttConnectVariableHeader) message.variableHeader();
+  MqttConnectPayload payload = 
((MqttConnectMessage)message).payload();
+  log.append(" name=").append(connectHeader.name())
+ .append(", version=").append(connectHeader.version())
--- End diff --

@jbertram I would create the logMessage @franz1981 is suggesting, but my 
concern here is not much about perf (since this method is already only called 
upon logging).. but out of making the code concise. Just a suggestion though.


---

[GitHub] activemq-artemis pull request #1597: ARTEMIS-1468 Fix delete address in cons...

[mtaylor]

GitHub user mtaylor opened a pull request:

https://github.com/apache/activemq-artemis/pull/1597

ARTEMIS-1468 Fix delete address in console



You can merge this pull request into a Git repository by running:

$ git pull https://github.com/mtaylor/activemq-artemis ARTEMIS-1468

Alternatively you can review and apply these changes as the patch at:

https://github.com/apache/activemq-artemis/pull/1597.patch

To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:

This closes #1597


commit c15742150ae5691fe2f057132412356983064c1c
Author: Martyn Taylor 
Date:   2017-10-18T12:28:27Z

ARTEMIS-1468 Fix delete address in console




---

[GitHub] activemq-artemis issue #1594: ARTEMIS-1467 example showing the recommended a...

[clebertsuconic]

Github user clebertsuconic commented on the issue:

https://github.com/apache/activemq-artemis/pull/1594
  
you could also use export, import?


---

[GitHub] activemq-artemis pull request #1595: ARTEMIS-1466 Upgrade Node.js version to...

[asfgit]

Github user asfgit closed the pull request at:

https://github.com/apache/activemq-artemis/pull/1595


---

[GitHub] activemq-artemis pull request #1596: ARTEMIS-450 Deadlocked queues over AMQP...

[clebertsuconic]

GitHub user clebertsuconic opened a pull request:

https://github.com/apache/activemq-artemis/pull/1596

ARTEMIS-450 Deadlocked queues over AMQP after too many cancellations



You can merge this pull request into a Git repository by running:

$ git pull https://github.com/clebertsuconic/activemq-artemis amqp-deadlock

Alternatively you can review and apply these changes as the patch at:

https://github.com/apache/activemq-artemis/pull/1596.patch

To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:

This closes #1596


commit cdcb7fc0a9408e8fbf506da7d9d64a1e52541267
Author: Clebert Suconic 
Date:   2017-10-17T19:54:49Z

ARTEMIS-450 Fixing deadlock over lots of rollbacks and Queue.addHead

commit 1a8887a9c40477663ece7d4c87af72bd5f3fdefb
Author: Clebert Suconic 
Date:   2017-10-18T02:30:59Z

ARTEMIS-1462 Fixing QueueControlTest




---

Re: [VOTE] Apache ActiveMQ 5.15.2

[Jean-Baptiste Onofré]

+1 (binding)

Regards
JB

On Oct 18, 2017, 12:56, at 12:56, Robbie Gemmell  
wrote:
>On 17 October 2017 at 15:37, Christopher Shannon
> wrote:
>> Hi Everyone,
>>
>> I have created the ActiveMQ 5.15.2 release and it's ready for a vote.
>>
>> The list of resolved issues is here:
>>
>https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12311210=12341669
>>
>> You can get the release artifacts here:
>> https://dist.apache.org/repos/dist/dev/activemq/activemq/5.15.2/
>>
>> Maven repository is at:
>>
>https://repository.apache.org/content/repositories/orgapacheactivemq-1151/
>>
>> Source tag:
>https://git-wip-us.apache.org/repos/asf?p=activemq.git;a=commit;h=9e595d8674456810ee3bd2d4a9920ed8b680937b
>>
>> Please vote to approve this release.  The vote will remain open for
>72 hours.
>>
>> [ ] +1 Release the binary as Apache ActiveMQ 5.15.2
>> [ ] -1 (provide specific comments)
>>
>> Here's my +1
>
>+1 (non-binding)
>
>I checked things over as follows:
>- Verified the signature and checksum files.
>- Checked licence and notice files are present in the archives.
>- Ran the source build and tested with the sanity test profile.
>- Used the broker from the tar.gz binary to run some AMQP client
>examples.
>- Used the staging repo to run the Qpid JMS master tests against the
>broker.
>
>Robbie

Re: [VOTE] Apache ActiveMQ 5.15.2

[Robbie Gemmell]

On 17 October 2017 at 15:37, Christopher Shannon
 wrote:
> Hi Everyone,
>
> I have created the ActiveMQ 5.15.2 release and it's ready for a vote.
>
> The list of resolved issues is here:
> https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12311210=12341669
>
> You can get the release artifacts here:
> https://dist.apache.org/repos/dist/dev/activemq/activemq/5.15.2/
>
> Maven repository is at:
> https://repository.apache.org/content/repositories/orgapacheactivemq-1151/
>
> Source tag: 
> https://git-wip-us.apache.org/repos/asf?p=activemq.git;a=commit;h=9e595d8674456810ee3bd2d4a9920ed8b680937b
>
> Please vote to approve this release.  The vote will remain open for 72 hours.
>
> [ ] +1 Release the binary as Apache ActiveMQ 5.15.2
> [ ] -1 (provide specific comments)
>
> Here's my +1

+1 (non-binding)

I checked things over as follows:
- Verified the signature and checksum files.
- Checked licence and notice files are present in the archives.
- Ran the source build and tested with the sanity test profile.
- Used the broker from the tar.gz binary to run some AMQP client examples.
- Used the staging repo to run the Qpid JMS master tests against the broker.

Robbie

[GitHub] activemq-artemis issue #1534: ARTEMIS-1420 limit non-ssl connection hangs up...

[gemmellr]

Github user gemmellr commented on the issue:

https://github.com/apache/activemq-artemis/pull/1534
  
> "Connection readTimeout has occurred.",
NETTY_READ_TIMEOUT = "read-timeout-seconds";
private int readTimeout;

Would a more descriptive name be better here for the configuration etc, 
something that conveys its a one-time protocol header timeout at connection 
startup? Using simply 'read timeout' seems like it could be confused with being 
related to other more general ongoing TCP read timeout handling and 
heartbeating etc, whereas it actually wont have effect beyond the first 8 
bytes. Maybe 'protocol-header-timeout' or something like that?

The test was added in the amqp test package as opposed to somewhere more 
general, which would seem nicer given it is entirely protocol-agnostic 
behaviour under test. Overlooking that though it was added to 
"AmqpSendReceiveTest" which seems odd given it obviously doesn't establish an 
AMQP connection and send or receive anything. Adding it there also means the 
test had to start->stop->reconfigure->restart the broker to perform the 
checking. Adding a new test class would be nicer, preferably outside the amqp 
package, but even if shoehorning into an existing amqp test there are perhaps 
better options like AmqpProtocolHeaderHandlingTest where it could be taken as a 
'doesnt send header' test.


---

[GitHub] activemq-artemis pull request #1595: ARTEMIS-1466 Upgrade Node.js version to...

[pkralik]

GitHub user pkralik opened a pull request:

https://github.com/apache/activemq-artemis/pull/1595

ARTEMIS-1466 Upgrade Node.js version to comply with GitBook 3.2.3



You can merge this pull request into a Git repository by running:

$ git pull https://github.com/pkralik/activemq-artemis-1 ARTEMIS-1466

Alternatively you can review and apply these changes as the patch at:

https://github.com/apache/activemq-artemis/pull/1595.patch

To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:

This closes #1595


commit b3a3057c56fcb2961593b0e071626183e51effe0
Author: Pavel Kralik 
Date:   2017-10-18T09:19:21Z

Node.JS version upgrade to fix GitBook 3.2.3 crash




---

[GitHub] activemq-artemis issue #1293: ARTEMIS-1194: SOCKS proxy support

[adagys]

Github user adagys commented on the issue:

https://github.com/apache/activemq-artemis/pull/1293
  
Rebased 👍 


---

Re: [VOTE] Apache ActiveMQ 5.15.2

[Gary Tully]

+1

On Tue, 17 Oct 2017 at 21:49 Timothy Bish  wrote:

> +1
>
> * Checked each archive to License and Notice files
> * Validated signatures and checksums
> * Ran the binary broker build on linux, checked web console and sent
> some messages using Qpid JMS example
> * Built from source and ran the sanity test profile tests
> * Built Qpid JMS using the staged bits and ran it's tests against ActiveMQ
> * Checked using the maven apache-rat:check goal for files missing any
> license headers.
>
> On 10/17/2017 10:37 AM, Christopher Shannon wrote:
> > Hi Everyone,
> >
> > I have created the ActiveMQ 5.15.2 release and it's ready for a vote.
> >
> > The list of resolved issues is here:
> >
> https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12311210=12341669
> >
> > You can get the release artifacts here:
> > https://dist.apache.org/repos/dist/dev/activemq/activemq/5.15.2/
> >
> > Maven repository is at:
> >
> https://repository.apache.org/content/repositories/orgapacheactivemq-1151/
> >
> > Source tag:
> https://git-wip-us.apache.org/repos/asf?p=activemq.git;a=commit;h=9e595d8674456810ee3bd2d4a9920ed8b680937b
> >
> > Please vote to approve this release.  The vote will remain open for 72
> hours.
> >
> > [ ] +1 Release the binary as Apache ActiveMQ 5.15.2
> > [ ] -1 (provide specific comments)
> >
> > Here's my +1
> >
>
> --
> Tim Bish
> twitter: @tabish121
> blog: http://timbish.blogspot.com/
>
>

[GitHub] activemq-artemis issue #1534: ARTEMIS-1420 limit non-ssl connection hangs up...

[stanlyDoge]

Github user stanlyDoge commented on the issue:

https://github.com/apache/activemq-artemis/pull/1534
  
@mtaylor Thanks, changed.


---

[GitHub] activemq-artemis pull request #1594: ARTEMIS-1467 example showing the recomm...

[stanlyDoge]

GitHub user stanlyDoge opened a pull request:

https://github.com/apache/activemq-artemis/pull/1594

ARTEMIS-1467 example showing the recommended approach for migrating

example showing the recommended approach for migrating messages from 
ActiveMQ to Artemis


You can merge this pull request into a Git repository by running:

$ git pull https://github.com/stanlyDoge/activemq-artemis-1 ARTEMIS-1467

Alternatively you can review and apply these changes as the patch at:

https://github.com/apache/activemq-artemis/pull/1594.patch

To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:

This closes #1594


commit 685de050aedaf567e88da9140d87d0d2ecd7c2c1
Author: Stanislav Knot 
Date:   2017-10-18T08:47:59Z

ARTEMIS-1467 example showing the recommended approach for migrating messages




---

[GitHub] activemq-artemis pull request #1592: NO-JIRA improve MQTT protocol logging

[franz1981]

Github user franz1981 commented on a diff in the pull request:

https://github.com/apache/activemq-artemis/pull/1592#discussion_r145358049
  
--- Diff: 
artemis-protocols/artemis-mqtt-protocol/src/main/java/org/apache/activemq/artemis/core/protocol/mqtt/MQTTUtil.java
 ---
@@ -149,16 +160,62 @@ public static void logMessage(MQTTSessionState state, 
MqttMessage message, boole
  if (message.fixedHeader() != null) {
 log.append(message.fixedHeader().messageType().toString());
 
-if (message.variableHeader() instanceof 
MqttPublishVariableHeader) {
-   log.append("(" + ((MqttPublishVariableHeader) 
message.variableHeader()).messageId() + ") " + 
message.fixedHeader().qosLevel());
-} else if (message.variableHeader() instanceof 
MqttMessageIdVariableHeader) {
+if (message.variableHeader() instanceof 
MqttMessageIdVariableHeader) {
log.append("(" + ((MqttMessageIdVariableHeader) 
message.variableHeader()).messageId() + ")");
 }
 
-if (message.fixedHeader().messageType() == 
MqttMessageType.SUBSCRIBE) {
-   for (MqttTopicSubscription sub : ((MqttSubscribeMessage) 
message).payload().topicSubscriptions()) {
-  log.append("\n\t" + sub.topicName() + " : " + 
sub.qualityOfService());
-   }
+switch (message.fixedHeader().messageType()) {
+   case PUBLISH:
+  MqttPublishVariableHeader publishHeader = 
(MqttPublishVariableHeader) message.variableHeader();
+  log.append("(" + publishHeader.packetId() + ")")
+ .append(" topic=" + publishHeader.topicName())
+ .append(", qos=" + message.fixedHeader().qosLevel())
+ .append(", retain=" + 
message.fixedHeader().isRetain())
+ .append(", dup=" + message.fixedHeader().isDup());
+  if (logUTF8PublishPayload != null) {
+ log.append(" payload=" + 
((MqttPublishMessage)message).payload().toString(StandardCharsets.UTF_8));
+  }
+  break;
+   case CONNECT:
+  MqttConnectVariableHeader connectHeader = 
(MqttConnectVariableHeader) message.variableHeader();
+  MqttConnectPayload payload = 
((MqttConnectMessage)message).payload();
+  log.append(" name=").append(connectHeader.name())
+ .append(", version=").append(connectHeader.version())
--- End diff --

@jbertram In order to avoid the JVM to compile `logMessage` with a big 
method (killing futher optimisations of the caller) I suggest to move the log 
message building it in a separate method (eg `private static void 
traceMessage(...)`) and use a pattern of calls like this:
```
public static void logMessage(MQTTSessionState state, MqttMessage 
message, boolean inbound) {
   if (logger.isTraceEnabled()) {
  traceMessage(state, message, inbound);
   }
}
```
This will let the JVM inline the `logMessage` call when is not needed.


---

[GitHub] activemq-artemis pull request #1593: ARTEMIS-1463 - added SSL example + docs

[andytaylor]

GitHub user andytaylor opened a pull request:

https://github.com/apache/activemq-artemis/pull/1593

ARTEMIS-1463 - added SSL example + docs

also added support for anon login

https://issues.apache.org/jira/browse/ARTEMIS-1463

You can merge this pull request into a Git repository by running:

$ git pull https://github.com/andytaylor/activemq-artemis rbac

Alternatively you can review and apply these changes as the patch at:

https://github.com/apache/activemq-artemis/pull/1593.patch

To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:

This closes #1593


commit 8dddeccca45e63c7616432a7ee4bd81327d93bb8
Author: Andy Taylor 
Date:   2017-10-18T08:58:39Z

ARTEMIS-1463 - added SSL example + docs

also added support for anon login

https://issues.apache.org/jira/browse/ARTEMIS-1463




---