Re: [DISCUSS] Removing Rest from ActiveMQ Artemis

2022-09-22 Thread Clebert Suconic
As I said on the Voting Thread. Rest is an independent maven component. Users have to consume it from maven on their own Rest application. So, they can still keep doing it from a previous release until they decide to not use it any more: https://github.com/apache/activemq-artemis/pull/4229

Re: [DISCUSS] Removing Rest from ActiveMQ Artemis

2022-09-22 Thread Clebert Suconic
>> Although I don't think anybody wants to invest the resources to bring all the dependencies up-to-date. << until someone does that, I think it should be out. this is pretty much abandonware. On Thu, Sep 22, 2022 at 2:04 PM Justin Bertram wrote: > > I've been meaning to get to this for

Re: [DISCUSS] Removing Rest from ActiveMQ Artemis

2022-09-22 Thread Justin Bertram
I've been meaning to get to this for awhile. Better late than never, right? Generally speaking, I agree that REST is not a great fit for messaging. Whenever folks ask about it on the mailing lists, Slack, Stack Overflow, etc. I strongly encourage them for two main reasons: - The STOMP protocol

Re: [DISCUSS] Removing Rest from ActiveMQ Artemis

2022-09-13 Thread Clebert Suconic
@Matt no problem at all.. just making it clear... And I am actually conflicted if there would be value in bringing it to artemis. If there would be any extra features to be developed? I don't honestly "believe" in rest over messaging the API gets too stateful to my taste.that's why I'm

Re: [DISCUSS] Removing Rest from ActiveMQ Artemis

2022-09-13 Thread Matt Pavlovich
Hey Clebert- Yeah, I was just addressing Lucas’ question from earlier in the thread re removing from ActiveMQ 5. -Matt > On Sep 13, 2022, at 9:27 AM, Clebert Suconic > wrote: > > I'm not suggesting we remove it.. just if there is value in bringing > it to artemis. > > On Tue, Sep 13, 2022

Re: [DISCUSS] Removing Rest from ActiveMQ Artemis

2022-09-13 Thread Clebert Suconic
I'm not suggesting we remove it.. just if there is value in bringing it to artemis. On Tue, Sep 13, 2022 at 10:24 AM Matt Pavlovich wrote: > > The REST service in ActiveMQ 5 is actively used, and should not be removed. > If there are bugs or issues, please file a JIRA. > > Thanks, > Matt

Re: [DISCUSS] Removing Rest from ActiveMQ Artemis

2022-09-13 Thread Matt Pavlovich
The REST service in ActiveMQ 5 is actively used, and should not be removed. If there are bugs or issues, please file a JIRA. Thanks, Matt Pavlovich > On Sep 12, 2022, at 3:58 PM, Tetreault, Lucas > wrote: > > Should we remove REST from ActiveMQ "Classic" as well? I think the same >

Re: [DISCUSS] Removing Rest from ActiveMQ Artemis

2022-09-13 Thread Clebert Suconic
Question now: should we have someone bringing the rest from AMQ5 into Artemis? Is there any value on converting it to artemis? On Mon, Sep 12, 2022 at 9:53 PM Clebert Suconic wrote: > > If there is value on the rest from activemq 5. I would rather bring that one > to artemis instead of keeping

Re: [DISCUSS] Removing Rest from ActiveMQ Artemis

2022-09-12 Thread Clebert Suconic
If there is value on the rest from activemq 5. I would rather bring that one to artemis instead of keeping the one we had. On Mon, Sep 12, 2022 at 8:12 PM Tetreault, Lucas wrote: > > Do the reasons for removal are the same as those presented at the > beginning of this thread? > No, probably not

Re: [DISCUSS] Removing Rest from ActiveMQ Artemis

2022-09-12 Thread Cesar Hernandez
> > How can we analyze usage? Agree, that's an interesting question. On the top of my head, and without knowing activemq.apache.org website infrastructure details, it would be interesting to get overall traffic stats from the page https://activemq.apache.org/rest . That at least will give us

Re: [DISCUSS] Removing Rest from ActiveMQ Artemis

2022-09-12 Thread Tetreault, Lucas
> Do the reasons for removal are the same as those presented at the beginning > of this thread? No, probably not all the same reasons but it seems to me that there is some value in keeping the two brokers in sync. Especially since some of the arguments in support of removing it are "I don't

Re: [DISCUSS] Removing Rest from ActiveMQ Artemis

2022-09-12 Thread Cesar Hernandez
> > Should we remove REST from ActiveMQ "Classic" as well? Do the reasons for removal are the same as those presented at the beginning of this thread? As far as I understand, REST from ActiveMQ Classis it´s working. My two cents is that the usage of the https://activemq.apache.org/rest should

Re: [DISCUSS] Removing Rest from ActiveMQ Artemis

2022-09-12 Thread Tetreault, Lucas
Should we remove REST from ActiveMQ "Classic" as well? I think the same arguments apply about it being abandonware, etc. We could deprecate it in the upcoming 5.18.0 release and use this as incentive to cut a 6.0.0 release? __ That would be exciting! - Lucas On 2022-09-12, 7:41 AM,

Re: [DISCUSS] Removing Rest from ActiveMQ Artemis

2022-09-12 Thread Christopher Shannon
Yeah that's the tricky part, I'm not sure how doable it is but figured if it was ever going to be looked at then a major version change would be the time to re-visit. On Mon, Sep 12, 2022 at 1:03 PM Clebert Suconic wrote: > We will have to find a way to comply with older clients though. We can

Re: [DISCUSS] Removing Rest from ActiveMQ Artemis

2022-09-12 Thread Clebert Suconic
We will have to find a way to comply with older clients though. We can break the API but compatibility with older clients has always been respected. On Mon, Sep 12, 2022 at 10:54 AM Christopher Shannon wrote: > > If you are going to move to version 3.0 it would be a nice time to look at > things

Re: [DISCUSS] Removing Rest from ActiveMQ Artemis

2022-09-12 Thread Christopher Shannon
If you are going to move to version 3.0 it would be a nice time to look at things to fix that are breaking changes like the spec violations I pointed out in https://issues.apache.org/jira/browse/ARTEMIS-1262 . The issue has been around for at least 5 years when I opened the Jira. I didn't see a

Re: [DISCUSS] Removing Rest from ActiveMQ Artemis

2022-09-12 Thread Clebert Suconic
I will go ahead and remove it... I will also bump upstream/main as 3.0 as part of the removal. On Thu, Sep 8, 2022 at 2:24 PM Clebert Suconic wrote: > > I didn’t meant to imply I was going to remove it now > > I intend to do it on Monday. If no objection. > > > Although keeping it means we

Re: [DISCUSS] Removing Rest from ActiveMQ Artemis

2022-09-08 Thread Clebert Suconic
I didn’t meant to imply I was going to remove it now I intend to do it on Monday. If no objection. Although keeping it means we would have to fix it. I honestly don’t see many options to keep it. On Thu, Sep 8, 2022 at 1:09 PM Robbie Gemmell wrote: > While I'm fine with it going I'd maybe

Re: [DISCUSS] Removing Rest from ActiveMQ Artemis

2022-09-08 Thread Robbie Gemmell
While I'm fine with it going I'd maybe give other folks more of chance to reply...or at least use lay concensus style 'ill do it at if noone objects' :) On Thu, 8 Sept 2022 at 17:48, Clebert Suconic wrote: > > I will go ahead and remove it... > > > to be honest I don't believe much in rest from

Re: [DISCUSS] Removing Rest from ActiveMQ Artemis

2022-09-08 Thread Clebert Suconic
I will go ahead and remove it... to be honest I don't believe much in rest from JMS due to the session and stateful nature. But if we were to provide REST for our users, I would rather bring the servlet from AMQ5. it would be a major task anyway... and this module has to go for sure. On

Re: [DISCUSS] Removing Rest from ActiveMQ Artemis

2022-09-08 Thread Robbie Gemmell
It has looked to be rotting for a long time, and requires various user hoop jumping I dont expect many/any folks are interested in doingI think removing it makes sense. On Thu, 8 Sept 2022 at 14:54, Clebert Suconic wrote: > > I'm not sure if there's much to discuss here. Rest in Artemis has

[DISCUSS] Removing Rest from ActiveMQ Artemis

2022-09-08 Thread Clebert Suconic
I'm not sure if there's much to discuss here. Rest in Artemis has been abandonware for a while (like 5 years)... The jboss-rest interface is a few major releases behind, the module compiles but it's not functional, and any time someone ask questions we just mention don't use it... (favoring stomp