[DISCUSS] Removing the Web Console

classic Classic list List threaded Threaded
53 messages Options
123
Reply | Threaded
Open this post in threaded view
|

[DISCUSS] Removing the Web Console

christopher.l.shannon
First, I know this topic was brought up back in January 2014 and there were
a lot of discussions about what to do about it  and ultimately nothing
happened.  However, it has been nearly 3 years since the last time this
subject was brought up and absolutely nothing has changed so I think it is
time to bring it up again and see what people's current opinions are.

The Web Console is extremely out of date and since the last discussions on
the subject is still completely un-maintained.  It is buggy and has had
many security vulnerabilities that keep popping up including several that
have been reported over the past year.  In the past 3 years no one has
shown any interest in contributing fixes to the console to maintain it.
Essentially no work has gone into the console except for security fixes.

Also, I know there was talk about moving it into a sub project however I
don't think that really solves anything.  The code would just be moved to a
new location and still be un-maintained and full of potential security
vulnerabilities.

So my preference would be just to EOL the console and remove it form future
versions. However, if there are people who really still want to keep it
then at the very least I think it should go into a sub project along with
some sort of warning that says it is deprecated and to use at your own
risk, etc.

Thoughts?
Reply | Threaded
Open this post in threaded view
|

Re: [DISCUSS] Removing the Web Console

Jim Gomes
Thanks for getting the discussion going again.  You bring up some
interesting points.  As stale as the console may be, I still find it
incredibly useful, and would hope that it will remain until a replacement
option is available.  I have found moving to Apache Artemis to feel like
I'm moving backwards because there is no admin console for it.

For those that may view it as a security risk, it is a simple matter to
disable it.  If it were to be replaced, what would be some potential
replacements? Could the most vulnerable parts of it be removed while still
remaining useful?  I mostly use it for knowing what clients are connected,
how many messages have been sent to destinations, and things like that. I
can't see how those limited functions would be difficult to keep, nor how
they could be a security issue.

On Wed, Sep 28, 2016 at 8:18 AM Christopher Shannon <
[hidden email]> wrote:

> First, I know this topic was brought up back in January 2014 and there were
> a lot of discussions about what to do about it  and ultimately nothing
> happened.  However, it has been nearly 3 years since the last time this
> subject was brought up and absolutely nothing has changed so I think it is
> time to bring it up again and see what people's current opinions are.
>
> The Web Console is extremely out of date and since the last discussions on
> the subject is still completely un-maintained.  It is buggy and has had
> many security vulnerabilities that keep popping up including several that
> have been reported over the past year.  In the past 3 years no one has
> shown any interest in contributing fixes to the console to maintain it.
> Essentially no work has gone into the console except for security fixes.
>
> Also, I know there was talk about moving it into a sub project however I
> don't think that really solves anything.  The code would just be moved to a
> new location and still be un-maintained and full of potential security
> vulnerabilities.
>
> So my preference would be just to EOL the console and remove it form future
> versions. However, if there are people who really still want to keep it
> then at the very least I think it should go into a sub project along with
> some sort of warning that says it is deprecated and to use at your own
> risk, etc.
>
> Thoughts?
>
Reply | Threaded
Open this post in threaded view
|

Re: [DISCUSS] Removing the Web Console

clebertsuconic
> For those that may view it as a security risk, it is a simple matter to disable it

I'm not sure I agree with this..
from my experience most users will just ignore or not read the docs at all..

 the most exploited security breaches I know happened around
documented recommendations. (Like removing the user admin/admin from
your whatever configuration for example). It's a dumb example but I
hope it stress what I mean.

the ones usually bad intentioned (or ill intentioned if you are
British)  will actually read the manual. just what I have seen as a
common pattern.
Reply | Threaded
Open this post in threaded view
|

Re: [DISCUSS] Removing the Web Console

John D. Ament-2
In reply to this post by Jim Gomes
Just wondering - considering where a number of committers work.  Why not
leverage hawt.io as a new console?

John

On Thu, Sep 29, 2016 at 7:45 PM Jim Gomes <[hidden email]> wrote:

> Thanks for getting the discussion going again.  You bring up some
> interesting points.  As stale as the console may be, I still find it
> incredibly useful, and would hope that it will remain until a replacement
> option is available.  I have found moving to Apache Artemis to feel like
> I'm moving backwards because there is no admin console for it.
>
> For those that may view it as a security risk, it is a simple matter to
> disable it.  If it were to be replaced, what would be some potential
> replacements? Could the most vulnerable parts of it be removed while still
> remaining useful?  I mostly use it for knowing what clients are connected,
> how many messages have been sent to destinations, and things like that. I
> can't see how those limited functions would be difficult to keep, nor how
> they could be a security issue.
>
> On Wed, Sep 28, 2016 at 8:18 AM Christopher Shannon <
> [hidden email]> wrote:
>
> > First, I know this topic was brought up back in January 2014 and there
> were
> > a lot of discussions about what to do about it  and ultimately nothing
> > happened.  However, it has been nearly 3 years since the last time this
> > subject was brought up and absolutely nothing has changed so I think it
> is
> > time to bring it up again and see what people's current opinions are.
> >
> > The Web Console is extremely out of date and since the last discussions
> on
> > the subject is still completely un-maintained.  It is buggy and has had
> > many security vulnerabilities that keep popping up including several that
> > have been reported over the past year.  In the past 3 years no one has
> > shown any interest in contributing fixes to the console to maintain it.
> > Essentially no work has gone into the console except for security fixes.
> >
> > Also, I know there was talk about moving it into a sub project however I
> > don't think that really solves anything.  The code would just be moved
> to a
> > new location and still be un-maintained and full of potential security
> > vulnerabilities.
> >
> > So my preference would be just to EOL the console and remove it form
> future
> > versions. However, if there are people who really still want to keep it
> > then at the very least I think it should go into a sub project along with
> > some sort of warning that says it is deprecated and to use at your own
> > risk, etc.
> >
> > Thoughts?
> >
>
Reply | Threaded
Open this post in threaded view
|

Re: [DISCUSS] Removing the Web Console

christopher.l.shannon
John, you'd have to search the archives but if you look at the history
Hawt.io was brought in as the console but backed out.  I think there was
disagreement about having a RH product as part of the distribution or
something like that with branding.  This was before my time contributing so
someone who was involved with the discussions back then can probably chime
in.  I know there was quite a large discussion on the issue.

I do agree with Jim that having a console is useful but I would prefer to
have something 3rd party as a console whether that is Hawt.io or something
else.  I personally would love to just use Hawt.io and I think it is a good
product (and I don't work for RH).  However even if we don't bundle Hawt.io
with ActiveMQ users can still use it, they just have to install it separate.

Ultimately just don't think we should be in the business of trying to
provide a web console (or any software for that matter) if we aren't going
to provide any support for it as it will just stay buggy and vulnerable to
exploits.

This also brings up another point...I suppose we should look into some sort
of web console support for Artemis as well (ie writing a plugin for
Hawt.io, etc)

On Thu, Sep 29, 2016 at 9:47 PM, John D. Ament <[hidden email]>
wrote:

> Just wondering - considering where a number of committers work.  Why not
> leverage hawt.io as a new console?
>
> John
>
> On Thu, Sep 29, 2016 at 7:45 PM Jim Gomes <[hidden email]> wrote:
>
> > Thanks for getting the discussion going again.  You bring up some
> > interesting points.  As stale as the console may be, I still find it
> > incredibly useful, and would hope that it will remain until a replacement
> > option is available.  I have found moving to Apache Artemis to feel like
> > I'm moving backwards because there is no admin console for it.
> >
> > For those that may view it as a security risk, it is a simple matter to
> > disable it.  If it were to be replaced, what would be some potential
> > replacements? Could the most vulnerable parts of it be removed while
> still
> > remaining useful?  I mostly use it for knowing what clients are
> connected,
> > how many messages have been sent to destinations, and things like that. I
> > can't see how those limited functions would be difficult to keep, nor how
> > they could be a security issue.
> >
> > On Wed, Sep 28, 2016 at 8:18 AM Christopher Shannon <
> > [hidden email]> wrote:
> >
> > > First, I know this topic was brought up back in January 2014 and there
> > were
> > > a lot of discussions about what to do about it  and ultimately nothing
> > > happened.  However, it has been nearly 3 years since the last time this
> > > subject was brought up and absolutely nothing has changed so I think it
> > is
> > > time to bring it up again and see what people's current opinions are.
> > >
> > > The Web Console is extremely out of date and since the last discussions
> > on
> > > the subject is still completely un-maintained.  It is buggy and has had
> > > many security vulnerabilities that keep popping up including several
> that
> > > have been reported over the past year.  In the past 3 years no one has
> > > shown any interest in contributing fixes to the console to maintain it.
> > > Essentially no work has gone into the console except for security
> fixes.
> > >
> > > Also, I know there was talk about moving it into a sub project however
> I
> > > don't think that really solves anything.  The code would just be moved
> > to a
> > > new location and still be un-maintained and full of potential security
> > > vulnerabilities.
> > >
> > > So my preference would be just to EOL the console and remove it form
> > future
> > > versions. However, if there are people who really still want to keep it
> > > then at the very least I think it should go into a sub project along
> with
> > > some sort of warning that says it is deprecated and to use at your own
> > > risk, etc.
> > >
> > > Thoughts?
> > >
> >
>
Reply | Threaded
Open this post in threaded view
|

Re: [DISCUSS] Removing the Web Console

jgenender
In reply to this post by John D. Ament-2
John D. Ament-2 wrote
Just wondering - considering where a number of committers work.  Why not
leverage hawt.io as a new console?
Oh boy... that was likely not the right thing to say... ;-)  I see Pandora's box about to open up... and surprisingly, John, your answer concerns me as you are pretty good about the Apache way. ;-)  I mean that as a compliment (mostly).

I suggest you review the very long and heated discussion surrounding hawt.io that occurred surrounding its inclusion in ActiveMQ.

Long story short, one of the main issues was what you just stated.  A) The fact that where "number of committers work" left out others from ActiveMQ's perspective for how they can contribute and making it fair.  B) There was certainly a branding issue that didn't make hawt.io a component of ActiveMQ.  

The thread regarding that has a very long discussion surrounding those items and its a very big sore spot in this community.  Suggestions ranged from templating hawt.io to allow ActiveMQ to brand it, to actually making it a project at Apache.... both of which were non negotiable from the hawt.io side.  So from that perspective... it's not going to work.  

If you are interested in having it become a part of ActiveMQ, I suggest that you review the thread and examine the proposed solutions and see if you can get your employer consider what was discussed.  I would also attempt to do it without rehashing the negative side.

As a person who works with many, many companies that use ActiveMQ... I see the included web console used nearly everywhere.  I would say the web console to hawt.io installation ratio in the wild is about 20:1.  So I just wanted to share that perspective.  Therefore, I am very heavily aligned with Jim's comments and agree with him on that point.
Reply | Threaded
Open this post in threaded view
|

Re: [DISCUSS] Removing the Web Console

clebertsuconic
I don't want to read that discussion again.. but from what I remember
of what I once read, and after I talked to some guys in person, the
issue was where the component would live.. like the plugin being
outside of AMQ5 code.

I believe that if we consumed hawt-io as a component (just like we
consume Jetty), and have the plugins, checkstyles, apache branding,
activemq5 and Artemis brand on the main repo, it shouldn't be an
issue.

hawtio would be a component, the plugin would live on our codebase and
users would experience it as an apache product.

It would be great if we could try at least with artemis, and if
everybody agrees/likes we could do the same on ActiveMQ5 branch.

On Fri, Sep 30, 2016 at 10:51 AM, jgenender <[hidden email]> wrote:

> John D. Ament-2 wrote
>> Just wondering - considering where a number of committers work.  Why not
>> leverage hawt.io as a new console?
>
> Oh boy... that was likely not the right thing to say... ;-)  I see Pandora's
> box about to open up... and surprisingly, John, your answer concerns me as
> you are pretty good about the Apache way. ;-)  I mean that as a compliment
> (mostly).
>
> I suggest you review the very long and heated discussion surrounding hawt.io
> that occurred surrounding its inclusion in ActiveMQ.
>
> Long story short, one of the main issues was what you just stated.  A) The
> fact that where "number of committers work" left out others from ActiveMQ's
> perspective for how they can contribute and making it fair.  B) There was
> certainly a branding issue that didn't make hawt.io a component of ActiveMQ.
>
> The thread regarding that has a very long discussion surrounding those items
> and its a very big sore spot in this community.  Suggestions ranged from
> templating hawt.io to allow ActiveMQ to brand it, to actually making it a
> project at Apache.... both of which were non negotiable from the hawt.io
> side.  So from that perspective... it's not going to work.
>
> If you are interested in having it become a part of ActiveMQ, I suggest that
> you review the thread and examine the proposed solutions and see if you can
> get your employer consider what was discussed.  I would also attempt to do
> it without rehashing the negative side.
>
> As a person who works with many, many companies that use ActiveMQ... I see
> the included web console used nearly everywhere.  I would say the web
> console to hawt.io installation ratio in the wild is about 20:1.  So I just
> wanted to share that perspective.  Therefore, I am very heavily aligned with
> Jim's comments and agree with him on that point.
>
>
>
>
> --
> View this message in context: http://activemq.2283324.n4.nabble.com/DISCUSS-Removing-the-Web-Console-tp4717136p4717295.html
> Sent from the ActiveMQ - Dev mailing list archive at Nabble.com.



--
Clebert Suconic
Reply | Threaded
Open this post in threaded view
|

Re: [DISCUSS] Removing the Web Console

jgenender
clebertsuconic wrote
I don't want to read that discussion again.. but from what I remember
of what I once read, and after I talked to some guys in person, the
issue was where the component would live.. like the plugin being
outside of AMQ5 code.

I believe that if we consumed hawt-io as a component (just like we
consume Jetty), and have the plugins, checkstyles, apache branding,
activemq5 and Artemis brand on the main repo, it shouldn't be an
issue.
I wouldn't speculate.  I wouldn't even attempt it unless you have examined the issues and do a 5 minute perusal on the thread.  I won't argue what it was because I, and some other non-Red Hat folks were central to that discussion.

My recommendation... don't rehash that.  Look at the primary problems were (tl;dr; I mentioned them previously).  Come up with a reasonable community based solution to the issues and present them.

That said, I think branding would help significantly as long as any other concerns are resolved.  I do know that templating it was certainly one of the offered solutions.
Reply | Threaded
Open this post in threaded view
|

Re: [DISCUSS] Removing the Web Console

clebertsuconic
That's nice.. so I guess we could now reboot the conversation around
hawtIO/webconsole... lets forget the past. and do it from scratch

My feeling around this was.... it's a pandora box.. don't touch it..
@jgenender used that same word here on this thread.


so... IMO, the Web console needs to look an apache product..
regardless of what components you use. if someone can provide a clean
and nice implementation.. using whatever frameworks or components that
are apache (or compatible) license, I think that's a reasonable
consideration.

And with git / github, we can first propose how it will look like, and
merge when it's pretty and ready. That's also a difference from 2 or 3
years ago when these discussions were taking place, where even if git
was being used the workflow was pretty much the same svn style.

I won't be able myself to work on this for a few weeks as I am working
on a few improvements around replication, that I want to do for 1.5.0.
but I think this would open the possibility of someone else looking
into that.. both from AMQ5 and/or Artemis perspective.

so if you (anyone) start working around this give us a sign here, so
there wouldn't be two people working on the same task.


A request I make is.. lets start fresh and do something cool and nice... ;)

On Fri, Sep 30, 2016 at 11:14 AM, jgenender <[hidden email]> wrote:

> clebertsuconic wrote
>> I don't want to read that discussion again.. but from what I remember
>> of what I once read, and after I talked to some guys in person, the
>> issue was where the component would live.. like the plugin being
>> outside of AMQ5 code.
>>
>> I believe that if we consumed hawt-io as a component (just like we
>> consume Jetty), and have the plugins, checkstyles, apache branding,
>> activemq5 and Artemis brand on the main repo, it shouldn't be an
>> issue.
>
> I wouldn't speculate.  I wouldn't even attempt it unless you have examined
> the issues and do a 5 minute perusal on the thread.  I won't argue what it
> was because I, and some other non-Red Hat folks were central to that
> discussion.
>
> My recommendation... don't rehash that.  Look at the primary problems were
> (tl;dr; I mentioned them previously).  Come up with a reasonable community
> based solution to the issues and present them.
>
> That said, I think branding would help significantly as long as any other
> concerns are resolved.  I do know that templating it was certainly one of
> the offered solutions.
>
>
>
>
> --
> View this message in context: http://activemq.2283324.n4.nabble.com/DISCUSS-Removing-the-Web-Console-tp4717136p4717302.html
> Sent from the ActiveMQ - Dev mailing list archive at Nabble.com.



--
Clebert Suconic
Reply | Threaded
Open this post in threaded view
|

Re: [DISCUSS] Removing the Web Console

dkulp

> On Sep 30, 2016, at 2:31 PM, Clebert Suconic <[hidden email]> wrote:
> so... IMO, the Web console needs to look an apache product..
> regardless of what components you use. if someone can provide a clean
> and nice implementation.. using whatever frameworks or components that
> are apache (or compatible) license, I think that's a reasonable
> consideration.

It’s a bit more than that….    It cannot be used to promote other 3rd party things.   Thus, other than a small “powered by” logo or similar in a non-prominant place, no other links out.   Also, all the non-ActiveMQ-essential things need to be able to be stripped out.  

Second, all the code related to interfacing and interacting with ActiveMQ/Artemis needs to be part of the ActiveMQ community.  This goes beyond branding.   Using the current ActiveMQ “plugin" from Hawt io is NOT ok unless all of that can be moved into the Apache community (which the developers did NOT want to do last time this was discussed).   Basically, how ActiveMQ is presented to the user MUST be completely under the control of the ActiveMQ community, not some other community.

Dan



>
> And with git / github, we can first propose how it will look like, and
> merge when it's pretty and ready. That's also a difference from 2 or 3
> years ago when these discussions were taking place, where even if git
> was being used the workflow was pretty much the same svn style.
>
> I won't be able myself to work on this for a few weeks as I am working
> on a few improvements around replication, that I want to do for 1.5.0.
> but I think this would open the possibility of someone else looking
> into that.. both from AMQ5 and/or Artemis perspective.
>
> so if you (anyone) start working around this give us a sign here, so
> there wouldn't be two people working on the same task.
>
>
> A request I make is.. lets start fresh and do something cool and nice... ;)
>
> On Fri, Sep 30, 2016 at 11:14 AM, jgenender <[hidden email]> wrote:
>> clebertsuconic wrote
>>> I don't want to read that discussion again.. but from what I remember
>>> of what I once read, and after I talked to some guys in person, the
>>> issue was where the component would live.. like the plugin being
>>> outside of AMQ5 code.
>>>
>>> I believe that if we consumed hawt-io as a component (just like we
>>> consume Jetty), and have the plugins, checkstyles, apache branding,
>>> activemq5 and Artemis brand on the main repo, it shouldn't be an
>>> issue.
>>
>> I wouldn't speculate.  I wouldn't even attempt it unless you have examined
>> the issues and do a 5 minute perusal on the thread.  I won't argue what it
>> was because I, and some other non-Red Hat folks were central to that
>> discussion.
>>
>> My recommendation... don't rehash that.  Look at the primary problems were
>> (tl;dr; I mentioned them previously).  Come up with a reasonable community
>> based solution to the issues and present them.
>>
>> That said, I think branding would help significantly as long as any other
>> concerns are resolved.  I do know that templating it was certainly one of
>> the offered solutions.
>>
>>
>>
>>
>> --
>> View this message in context: http://activemq.2283324.n4.nabble.com/DISCUSS-Removing-the-Web-Console-tp4717136p4717302.html
>> Sent from the ActiveMQ - Dev mailing list archive at Nabble.com.
>
>
>
> --
> Clebert Suconic

--
Daniel Kulp
[hidden email] <mailto:[hidden email]> - http://dankulp.com/blog <http://dankulp.com/blog>
Talend Community Coder - http://coders.talend.com <http://coders.talend.com/>
Reply | Threaded
Open this post in threaded view
|

Re: [DISCUSS] Removing the Web Console

christopher.l.shannon
So if we created a new plugin for Hawt.io that we hosted and controlled
would that be acceptable?  (Assuming the branding issues and skinning
issues are taken care of).

At this point I don't know that it would ever happen for 5.x but since
there's nothing currently available for Artemis then it might make sense to
do it since I think we need a web console.

On Fri, Sep 30, 2016 at 2:51 PM, Daniel Kulp <[hidden email]> wrote:

>
> > On Sep 30, 2016, at 2:31 PM, Clebert Suconic <[hidden email]>
> wrote:
> > so... IMO, the Web console needs to look an apache product..
> > regardless of what components you use. if someone can provide a clean
> > and nice implementation.. using whatever frameworks or components that
> > are apache (or compatible) license, I think that's a reasonable
> > consideration.
>
> It’s a bit more than that….    It cannot be used to promote other 3rd
> party things.   Thus, other than a small “powered by” logo or similar in a
> non-prominant place, no other links out.   Also, all the
> non-ActiveMQ-essential things need to be able to be stripped out.
>
> Second, all the code related to interfacing and interacting with
> ActiveMQ/Artemis needs to be part of the ActiveMQ community.  This goes
> beyond branding.   Using the current ActiveMQ “plugin" from Hawt io is NOT
> ok unless all of that can be moved into the Apache community (which the
> developers did NOT want to do last time this was discussed).   Basically,
> how ActiveMQ is presented to the user MUST be completely under the control
> of the ActiveMQ community, not some other community.
>
> Dan
>
>
>
> >
> > And with git / github, we can first propose how it will look like, and
> > merge when it's pretty and ready. That's also a difference from 2 or 3
> > years ago when these discussions were taking place, where even if git
> > was being used the workflow was pretty much the same svn style.
> >
> > I won't be able myself to work on this for a few weeks as I am working
> > on a few improvements around replication, that I want to do for 1.5.0.
> > but I think this would open the possibility of someone else looking
> > into that.. both from AMQ5 and/or Artemis perspective.
> >
> > so if you (anyone) start working around this give us a sign here, so
> > there wouldn't be two people working on the same task.
> >
> >
> > A request I make is.. lets start fresh and do something cool and nice...
> ;)
> >
> > On Fri, Sep 30, 2016 at 11:14 AM, jgenender <[hidden email]>
> wrote:
> >> clebertsuconic wrote
> >>> I don't want to read that discussion again.. but from what I remember
> >>> of what I once read, and after I talked to some guys in person, the
> >>> issue was where the component would live.. like the plugin being
> >>> outside of AMQ5 code.
> >>>
> >>> I believe that if we consumed hawt-io as a component (just like we
> >>> consume Jetty), and have the plugins, checkstyles, apache branding,
> >>> activemq5 and Artemis brand on the main repo, it shouldn't be an
> >>> issue.
> >>
> >> I wouldn't speculate.  I wouldn't even attempt it unless you have
> examined
> >> the issues and do a 5 minute perusal on the thread.  I won't argue what
> it
> >> was because I, and some other non-Red Hat folks were central to that
> >> discussion.
> >>
> >> My recommendation... don't rehash that.  Look at the primary problems
> were
> >> (tl;dr; I mentioned them previously).  Come up with a reasonable
> community
> >> based solution to the issues and present them.
> >>
> >> That said, I think branding would help significantly as long as any
> other
> >> concerns are resolved.  I do know that templating it was certainly one
> of
> >> the offered solutions.
> >>
> >>
> >>
> >>
> >> --
> >> View this message in context: http://activemq.2283324.n4.
> nabble.com/DISCUSS-Removing-the-Web-Console-tp4717136p4717302.html
> >> Sent from the ActiveMQ - Dev mailing list archive at Nabble.com.
> >
> >
> >
> > --
> > Clebert Suconic
>
> --
> Daniel Kulp
> [hidden email] <mailto:[hidden email]> - http://dankulp.com/blog <
> http://dankulp.com/blog>
> Talend Community Coder - http://coders.talend.com <
> http://coders.talend.com/>
>
Reply | Threaded
Open this post in threaded view
|

Re: [DISCUSS] Removing the Web Console

clebertsuconic
In reply to this post by dkulp
@Daniel +1... I share the same view... Lets just move forward.. and
forget the past thread.

@Christopher Shannon.. in regard to your other email.. +1

On Fri, Sep 30, 2016 at 2:51 PM, Daniel Kulp <[hidden email]> wrote:

>
>> On Sep 30, 2016, at 2:31 PM, Clebert Suconic <[hidden email]> wrote:
>> so... IMO, the Web console needs to look an apache product..
>> regardless of what components you use. if someone can provide a clean
>> and nice implementation.. using whatever frameworks or components that
>> are apache (or compatible) license, I think that's a reasonable
>> consideration.
>
> It’s a bit more than that….    It cannot be used to promote other 3rd party things.   Thus, other than a small “powered by” logo or similar in a non-prominant place, no other links out.   Also, all the non-ActiveMQ-essential things need to be able to be stripped out.
>
> Second, all the code related to interfacing and interacting with ActiveMQ/Artemis needs to be part of the ActiveMQ community.  This goes beyond branding.   Using the current ActiveMQ “plugin" from Hawt io is NOT ok unless all of that can be moved into the Apache community (which the developers did NOT want to do last time this was discussed).   Basically, how ActiveMQ is presented to the user MUST be completely under the control of the ActiveMQ community, not some other community.
>
> Dan
>
>
>
>>
>> And with git / github, we can first propose how it will look like, and
>> merge when it's pretty and ready. That's also a difference from 2 or 3
>> years ago when these discussions were taking place, where even if git
>> was being used the workflow was pretty much the same svn style.
>>
>> I won't be able myself to work on this for a few weeks as I am working
>> on a few improvements around replication, that I want to do for 1.5.0.
>> but I think this would open the possibility of someone else looking
>> into that.. both from AMQ5 and/or Artemis perspective.
>>
>> so if you (anyone) start working around this give us a sign here, so
>> there wouldn't be two people working on the same task.
>>
>>
>> A request I make is.. lets start fresh and do something cool and nice... ;)
>>
>> On Fri, Sep 30, 2016 at 11:14 AM, jgenender <[hidden email]> wrote:
>>> clebertsuconic wrote
>>>> I don't want to read that discussion again.. but from what I remember
>>>> of what I once read, and after I talked to some guys in person, the
>>>> issue was where the component would live.. like the plugin being
>>>> outside of AMQ5 code.
>>>>
>>>> I believe that if we consumed hawt-io as a component (just like we
>>>> consume Jetty), and have the plugins, checkstyles, apache branding,
>>>> activemq5 and Artemis brand on the main repo, it shouldn't be an
>>>> issue.
>>>
>>> I wouldn't speculate.  I wouldn't even attempt it unless you have examined
>>> the issues and do a 5 minute perusal on the thread.  I won't argue what it
>>> was because I, and some other non-Red Hat folks were central to that
>>> discussion.
>>>
>>> My recommendation... don't rehash that.  Look at the primary problems were
>>> (tl;dr; I mentioned them previously).  Come up with a reasonable community
>>> based solution to the issues and present them.
>>>
>>> That said, I think branding would help significantly as long as any other
>>> concerns are resolved.  I do know that templating it was certainly one of
>>> the offered solutions.
>>>
>>>
>>>
>>>
>>> --
>>> View this message in context: http://activemq.2283324.n4.nabble.com/DISCUSS-Removing-the-Web-Console-tp4717136p4717302.html
>>> Sent from the ActiveMQ - Dev mailing list archive at Nabble.com.
>>
>>
>>
>> --
>> Clebert Suconic
>
> --
> Daniel Kulp
> [hidden email] <mailto:[hidden email]> - http://dankulp.com/blog <http://dankulp.com/blog>
> Talend Community Coder - http://coders.talend.com <http://coders.talend.com/>



--
Clebert Suconic
Reply | Threaded
Open this post in threaded view
|

Re: [DISCUSS] Removing the Web Console

chirino
I'd like to point out that HawtIO 2.0 is different from 1.x.  Is been
modularized using bower components and any project can build a customized
HawtIO based console very easily with user specified plugins and branding.
Also please note, HawtIO is not a Red Hat product, it's an open source
project which Red Hat heavily contributes to.

On Mon, Oct 3, 2016 at 11:14 AM, Clebert Suconic <[hidden email]>
wrote:

> @Daniel +1... I share the same view... Lets just move forward.. and
> forget the past thread.
>
> @Christopher Shannon.. in regard to your other email.. +1
>
> On Fri, Sep 30, 2016 at 2:51 PM, Daniel Kulp <[hidden email]> wrote:
> >
> >> On Sep 30, 2016, at 2:31 PM, Clebert Suconic <[hidden email]>
> wrote:
> >> so... IMO, the Web console needs to look an apache product..
> >> regardless of what components you use. if someone can provide a clean
> >> and nice implementation.. using whatever frameworks or components that
> >> are apache (or compatible) license, I think that's a reasonable
> >> consideration.
> >
> > It’s a bit more than that….    It cannot be used to promote other 3rd
> party things.   Thus, other than a small “powered by” logo or similar in a
> non-prominant place, no other links out.   Also, all the
> non-ActiveMQ-essential things need to be able to be stripped out.
> >
> > Second, all the code related to interfacing and interacting with
> ActiveMQ/Artemis needs to be part of the ActiveMQ community.  This goes
> beyond branding.   Using the current ActiveMQ “plugin" from Hawt io is NOT
> ok unless all of that can be moved into the Apache community (which the
> developers did NOT want to do last time this was discussed).   Basically,
> how ActiveMQ is presented to the user MUST be completely under the control
> of the ActiveMQ community, not some other community.
> >
> > Dan
> >
> >
> >
> >>
> >> And with git / github, we can first propose how it will look like, and
> >> merge when it's pretty and ready. That's also a difference from 2 or 3
> >> years ago when these discussions were taking place, where even if git
> >> was being used the workflow was pretty much the same svn style.
> >>
> >> I won't be able myself to work on this for a few weeks as I am working
> >> on a few improvements around replication, that I want to do for 1.5.0.
> >> but I think this would open the possibility of someone else looking
> >> into that.. both from AMQ5 and/or Artemis perspective.
> >>
> >> so if you (anyone) start working around this give us a sign here, so
> >> there wouldn't be two people working on the same task.
> >>
> >>
> >> A request I make is.. lets start fresh and do something cool and
> nice... ;)
> >>
> >> On Fri, Sep 30, 2016 at 11:14 AM, jgenender <[hidden email]>
> wrote:
> >>> clebertsuconic wrote
> >>>> I don't want to read that discussion again.. but from what I remember
> >>>> of what I once read, and after I talked to some guys in person, the
> >>>> issue was where the component would live.. like the plugin being
> >>>> outside of AMQ5 code.
> >>>>
> >>>> I believe that if we consumed hawt-io as a component (just like we
> >>>> consume Jetty), and have the plugins, checkstyles, apache branding,
> >>>> activemq5 and Artemis brand on the main repo, it shouldn't be an
> >>>> issue.
> >>>
> >>> I wouldn't speculate.  I wouldn't even attempt it unless you have
> examined
> >>> the issues and do a 5 minute perusal on the thread.  I won't argue
> what it
> >>> was because I, and some other non-Red Hat folks were central to that
> >>> discussion.
> >>>
> >>> My recommendation... don't rehash that.  Look at the primary problems
> were
> >>> (tl;dr; I mentioned them previously).  Come up with a reasonable
> community
> >>> based solution to the issues and present them.
> >>>
> >>> That said, I think branding would help significantly as long as any
> other
> >>> concerns are resolved.  I do know that templating it was certainly one
> of
> >>> the offered solutions.
> >>>
> >>>
> >>>
> >>>
> >>> --
> >>> View this message in context: http://activemq.2283324.n4.
> nabble.com/DISCUSS-Removing-the-Web-Console-tp4717136p4717302.html
> >>> Sent from the ActiveMQ - Dev mailing list archive at Nabble.com.
> >>
> >>
> >>
> >> --
> >> Clebert Suconic
> >
> > --
> > Daniel Kulp
> > [hidden email] <mailto:[hidden email]> - http://dankulp.com/blog <
> http://dankulp.com/blog>
> > Talend Community Coder - http://coders.talend.com <
> http://coders.talend.com/>
>
>
>
> --
> Clebert Suconic
>



--
Hiram Chirino
Engineering | Red Hat, Inc.
[hidden email] | fusesource.com | redhat.com
skype: hiramchirino | twitter: @hiramchirino
Reply | Threaded
Open this post in threaded view
|

Re: [DISCUSS] Removing the Web Console

John D. Ament-2
@Hiram

The website branding says otherwise (take a look at the top right corner).

I like hawt... and using it in a modular fashion embedded within activemq
and artemis would be a great solution.  Forgetting any relationships.

John

On Tue, Oct 4, 2016 at 11:46 AM Hiram Chirino <[hidden email]>
wrote:

> I'd like to point out that HawtIO 2.0 is different from 1.x.  Is been
> modularized using bower components and any project can build a customized
> HawtIO based console very easily with user specified plugins and branding.
> Also please note, HawtIO is not a Red Hat product, it's an open source
> project which Red Hat heavily contributes to.
>
> On Mon, Oct 3, 2016 at 11:14 AM, Clebert Suconic <
> [hidden email]>
> wrote:
>
> > @Daniel +1... I share the same view... Lets just move forward.. and
> > forget the past thread.
> >
> > @Christopher Shannon.. in regard to your other email.. +1
> >
> > On Fri, Sep 30, 2016 at 2:51 PM, Daniel Kulp <[hidden email]> wrote:
> > >
> > >> On Sep 30, 2016, at 2:31 PM, Clebert Suconic <
> [hidden email]>
> > wrote:
> > >> so... IMO, the Web console needs to look an apache product..
> > >> regardless of what components you use. if someone can provide a clean
> > >> and nice implementation.. using whatever frameworks or components that
> > >> are apache (or compatible) license, I think that's a reasonable
> > >> consideration.
> > >
> > > It’s a bit more than that….    It cannot be used to promote other 3rd
> > party things.   Thus, other than a small “powered by” logo or similar in
> a
> > non-prominant place, no other links out.   Also, all the
> > non-ActiveMQ-essential things need to be able to be stripped out.
> > >
> > > Second, all the code related to interfacing and interacting with
> > ActiveMQ/Artemis needs to be part of the ActiveMQ community.  This goes
> > beyond branding.   Using the current ActiveMQ “plugin" from Hawt io is
> NOT
> > ok unless all of that can be moved into the Apache community (which the
> > developers did NOT want to do last time this was discussed).   Basically,
> > how ActiveMQ is presented to the user MUST be completely under the
> control
> > of the ActiveMQ community, not some other community.
> > >
> > > Dan
> > >
> > >
> > >
> > >>
> > >> And with git / github, we can first propose how it will look like, and
> > >> merge when it's pretty and ready. That's also a difference from 2 or 3
> > >> years ago when these discussions were taking place, where even if git
> > >> was being used the workflow was pretty much the same svn style.
> > >>
> > >> I won't be able myself to work on this for a few weeks as I am working
> > >> on a few improvements around replication, that I want to do for 1.5.0.
> > >> but I think this would open the possibility of someone else looking
> > >> into that.. both from AMQ5 and/or Artemis perspective.
> > >>
> > >> so if you (anyone) start working around this give us a sign here, so
> > >> there wouldn't be two people working on the same task.
> > >>
> > >>
> > >> A request I make is.. lets start fresh and do something cool and
> > nice... ;)
> > >>
> > >> On Fri, Sep 30, 2016 at 11:14 AM, jgenender <[hidden email]>
> > wrote:
> > >>> clebertsuconic wrote
> > >>>> I don't want to read that discussion again.. but from what I
> remember
> > >>>> of what I once read, and after I talked to some guys in person, the
> > >>>> issue was where the component would live.. like the plugin being
> > >>>> outside of AMQ5 code.
> > >>>>
> > >>>> I believe that if we consumed hawt-io as a component (just like we
> > >>>> consume Jetty), and have the plugins, checkstyles, apache branding,
> > >>>> activemq5 and Artemis brand on the main repo, it shouldn't be an
> > >>>> issue.
> > >>>
> > >>> I wouldn't speculate.  I wouldn't even attempt it unless you have
> > examined
> > >>> the issues and do a 5 minute perusal on the thread.  I won't argue
> > what it
> > >>> was because I, and some other non-Red Hat folks were central to that
> > >>> discussion.
> > >>>
> > >>> My recommendation... don't rehash that.  Look at the primary problems
> > were
> > >>> (tl;dr; I mentioned them previously).  Come up with a reasonable
> > community
> > >>> based solution to the issues and present them.
> > >>>
> > >>> That said, I think branding would help significantly as long as any
> > other
> > >>> concerns are resolved.  I do know that templating it was certainly
> one
> > of
> > >>> the offered solutions.
> > >>>
> > >>>
> > >>>
> > >>>
> > >>> --
> > >>> View this message in context: http://activemq.2283324.n4.
> > nabble.com/DISCUSS-Removing-the-Web-Console-tp4717136p4717302.html
> > >>> Sent from the ActiveMQ - Dev mailing list archive at Nabble.com.
> > >>
> > >>
> > >>
> > >> --
> > >> Clebert Suconic
> > >
> > > --
> > > Daniel Kulp
> > > [hidden email] <mailto:[hidden email]> - http://dankulp.com/blog <
> > http://dankulp.com/blog>
> > > Talend Community Coder - http://coders.talend.com <
> > http://coders.talend.com/>
> >
> >
> >
> > --
> > Clebert Suconic
> >
>
>
>
> --
> Hiram Chirino
> Engineering | Red Hat, Inc.
> [hidden email] | fusesource.com | redhat.com
> skype: hiramchirino | twitter: @hiramchirino
>
Reply | Threaded
Open this post in threaded view
|

Re: [DISCUSS] Removing the Web Console

clebertsuconic
On Tue, Oct 4, 2016 at 11:57 AM, John D. Ament <[hidden email]> wrote:
> @Hiram
>
> The website branding says otherwise (take a look at the top right corner).


That symbol on the top is just a link to the following:

"Like hawtio? It’s part of a community of Red Hat projects. Learn more
about Red Hat and our open source communities:"



 No other implications from what I see.
Reply | Threaded
Open this post in threaded view
|

Re: [DISCUSS] Removing the Web Console

Martyn Taylor
+1 on improving/adding a console.  Providing a console out of the box is a
massive win for user experience imo and something that I feel Artemis would
greatly benefit from.  Whether it's HawtIO or something else we should make
every effort to standardise across both 5.x and Artemis.

On Tue, Oct 4, 2016 at 5:11 PM, Clebert Suconic <[hidden email]>
wrote:

> On Tue, Oct 4, 2016 at 11:57 AM, John D. Ament <[hidden email]>
> wrote:
> > @Hiram
> >
> > The website branding says otherwise (take a look at the top right
> corner).
>
>
> That symbol on the top is just a link to the following:
>
> "Like hawtio? It’s part of a community of Red Hat projects. Learn more
> about Red Hat and our open source communities:"
>
>
>
>  No other implications from what I see.
>
Reply | Threaded
Open this post in threaded view
|

Re: [DISCUSS] Removing the Web Console

Hadrian Zbarcea
I would suggest discussing the goals for such a console first. Is it
intended to monitory just one broker instance or a whole network of
brokers? Should it manage just the brokers or other services? Should it
rely on JMX or something else?

Then one can think about reusing and/or improving something that's
available or some other solution.

The way this discuss goes, sounds to me like trying to push again for
something that was rejected in the past and I suspect will not go anywhere.

My $0.02,
Hadrian

On 10/07/2016 06:41 AM, Martyn Taylor wrote:

> +1 on improving/adding a console.  Providing a console out of the box is a
> massive win for user experience imo and something that I feel Artemis would
> greatly benefit from.  Whether it's HawtIO or something else we should make
> every effort to standardise across both 5.x and Artemis.
>
> On Tue, Oct 4, 2016 at 5:11 PM, Clebert Suconic <[hidden email]>
> wrote:
>
>> On Tue, Oct 4, 2016 at 11:57 AM, John D. Ament <[hidden email]>
>> wrote:
>>> @Hiram
>>>
>>> The website branding says otherwise (take a look at the top right
>> corner).
>>
>>
>> That symbol on the top is just a link to the following:
>>
>> "Like hawtio? It’s part of a community of Red Hat projects. Learn more
>> about Red Hat and our open source communities:"
>>
>>
>>
>>  No other implications from what I see.
>>
>
Reply | Threaded
Open this post in threaded view
|

Re: [DISCUSS] Removing the Web Console

clebertsuconic
I think this is pretty straightforward:

i - it should relay on JMX or jolokia.. A common thing between both
Artemis and ActiveMQ

ii - it should manage at least a single broker.. with some metrics...


iii - anything beyond that will just be a collaboration over the code.


The best way to discuss this IMO would be through a Pull Request..
someone send an initial draft.. we can have some **technical**
discussion over of PR, and commit it as version 1... then
collaboratively this could be increased just as with anything else.



I think we are clear from the previous discussions... and that's a
request I am making here, probably the third time... lets CTRL-Alt-Del
and start fresh... The issues we had are clear... and I see everybody
with a single goal here.. to have an integrated console that looks
like an Apache project, pretty and neat.


Once someone put a first version, we can only improve it from there.

On Fri, Oct 7, 2016 at 12:17 PM, Hadrian Zbarcea <[hidden email]> wrote:

> I would suggest discussing the goals for such a console first. Is it
> intended to monitory just one broker instance or a whole network of brokers?
> Should it manage just the brokers or other services? Should it rely on JMX
> or something else?
>
> Then one can think about reusing and/or improving something that's available
> or some other solution.
>
> The way this discuss goes, sounds to me like trying to push again for
> something that was rejected in the past and I suspect will not go anywhere.
>
> My $0.02,
> Hadrian
>
>
> On 10/07/2016 06:41 AM, Martyn Taylor wrote:
>>
>> +1 on improving/adding a console.  Providing a console out of the box is a
>> massive win for user experience imo and something that I feel Artemis
>> would
>> greatly benefit from.  Whether it's HawtIO or something else we should
>> make
>> every effort to standardise across both 5.x and Artemis.
>>
>> On Tue, Oct 4, 2016 at 5:11 PM, Clebert Suconic
>> <[hidden email]>
>> wrote:
>>
>>> On Tue, Oct 4, 2016 at 11:57 AM, John D. Ament <[hidden email]>
>>> wrote:
>>>>
>>>> @Hiram
>>>>
>>>> The website branding says otherwise (take a look at the top right
>>>
>>> corner).
>>>
>>>
>>> That symbol on the top is just a link to the following:
>>>
>>> "Like hawtio? It’s part of a community of Red Hat projects. Learn more
>>> about Red Hat and our open source communities:"
>>>
>>>
>>>
>>>  No other implications from what I see.
>>>
>>
>



--
Clebert Suconic
Reply | Threaded
Open this post in threaded view
|

Re: [DISCUSS] Removing the Web Console

Jim Gomes
I was under the impression that all Apache development discussions occur on
the email list. This is for legal and policy reasons. Please correct me if
I'm wrong.

On Fri, Oct 7, 2016, 9:43 AM Clebert Suconic <[hidden email]>
wrote:

> I think this is pretty straightforward:
>
> i - it should relay on JMX or jolokia.. A common thing between both
> Artemis and ActiveMQ
>
> ii - it should manage at least a single broker.. with some metrics...
>
>
> iii - anything beyond that will just be a collaboration over the code.
>
>
> The best way to discuss this IMO would be through a Pull Request..
> someone send an initial draft.. we can have some **technical**
> discussion over of PR, and commit it as version 1... then
> collaboratively this could be increased just as with anything else.
>
>
>
> I think we are clear from the previous discussions... and that's a
> request I am making here, probably the third time... lets CTRL-Alt-Del
> and start fresh... The issues we had are clear... and I see everybody
> with a single goal here.. to have an integrated console that looks
> like an Apache project, pretty and neat.
>
>
> Once someone put a first version, we can only improve it from there.
>
> On Fri, Oct 7, 2016 at 12:17 PM, Hadrian Zbarcea <[hidden email]>
> wrote:
> > I would suggest discussing the goals for such a console first. Is it
> > intended to monitory just one broker instance or a whole network of
> brokers?
> > Should it manage just the brokers or other services? Should it rely on
> JMX
> > or something else?
> >
> > Then one can think about reusing and/or improving something that's
> available
> > or some other solution.
> >
> > The way this discuss goes, sounds to me like trying to push again for
> > something that was rejected in the past and I suspect will not go
> anywhere.
> >
> > My $0.02,
> > Hadrian
> >
> >
> > On 10/07/2016 06:41 AM, Martyn Taylor wrote:
> >>
> >> +1 on improving/adding a console.  Providing a console out of the box
> is a
> >> massive win for user experience imo and something that I feel Artemis
> >> would
> >> greatly benefit from.  Whether it's HawtIO or something else we should
> >> make
> >> every effort to standardise across both 5.x and Artemis.
> >>
> >> On Tue, Oct 4, 2016 at 5:11 PM, Clebert Suconic
> >> <[hidden email]>
> >> wrote:
> >>
> >>> On Tue, Oct 4, 2016 at 11:57 AM, John D. Ament <[hidden email]>
> >>> wrote:
> >>>>
> >>>> @Hiram
> >>>>
> >>>> The website branding says otherwise (take a look at the top right
> >>>
> >>> corner).
> >>>
> >>>
> >>> That symbol on the top is just a link to the following:
> >>>
> >>> "Like hawtio? It’s part of a community of Red Hat projects. Learn more
> >>> about Red Hat and our open source communities:"
> >>>
> >>>
> >>>
> >>>  No other implications from what I see.
> >>>
> >>
> >
>
>
>
> --
> Clebert Suconic
>
Reply | Threaded
Open this post in threaded view
|

Re: [DISCUSS] Removing the Web Console

John D. Ament-2
I think I just saw two great ideas:

- A consolidated web console that is usable by both ActiveMQ and Artemis.
- It should be an independent repository/utility.

I think all of the discussions are happening on a ML, but realistically if
we're ready to start writing down some code, we need some of those
utilities in place.  Since PRs are mirrored to the MLs that would work, but
I would hate to see it baked into one of the existing repos to make that
happen.

John

On Fri, Oct 7, 2016 at 5:41 PM Jim Gomes <[hidden email]> wrote:

> I was under the impression that all Apache development discussions occur on
> the email list. This is for legal and policy reasons. Please correct me if
> I'm wrong.
>
> On Fri, Oct 7, 2016, 9:43 AM Clebert Suconic <[hidden email]>
> wrote:
>
> > I think this is pretty straightforward:
> >
> > i - it should relay on JMX or jolokia.. A common thing between both
> > Artemis and ActiveMQ
> >
> > ii - it should manage at least a single broker.. with some metrics...
> >
> >
> > iii - anything beyond that will just be a collaboration over the code.
> >
> >
> > The best way to discuss this IMO would be through a Pull Request..
> > someone send an initial draft.. we can have some **technical**
> > discussion over of PR, and commit it as version 1... then
> > collaboratively this could be increased just as with anything else.
> >
> >
> >
> > I think we are clear from the previous discussions... and that's a
> > request I am making here, probably the third time... lets CTRL-Alt-Del
> > and start fresh... The issues we had are clear... and I see everybody
> > with a single goal here.. to have an integrated console that looks
> > like an Apache project, pretty and neat.
> >
> >
> > Once someone put a first version, we can only improve it from there.
> >
> > On Fri, Oct 7, 2016 at 12:17 PM, Hadrian Zbarcea <[hidden email]>
> > wrote:
> > > I would suggest discussing the goals for such a console first. Is it
> > > intended to monitory just one broker instance or a whole network of
> > brokers?
> > > Should it manage just the brokers or other services? Should it rely on
> > JMX
> > > or something else?
> > >
> > > Then one can think about reusing and/or improving something that's
> > available
> > > or some other solution.
> > >
> > > The way this discuss goes, sounds to me like trying to push again for
> > > something that was rejected in the past and I suspect will not go
> > anywhere.
> > >
> > > My $0.02,
> > > Hadrian
> > >
> > >
> > > On 10/07/2016 06:41 AM, Martyn Taylor wrote:
> > >>
> > >> +1 on improving/adding a console.  Providing a console out of the box
> > is a
> > >> massive win for user experience imo and something that I feel Artemis
> > >> would
> > >> greatly benefit from.  Whether it's HawtIO or something else we should
> > >> make
> > >> every effort to standardise across both 5.x and Artemis.
> > >>
> > >> On Tue, Oct 4, 2016 at 5:11 PM, Clebert Suconic
> > >> <[hidden email]>
> > >> wrote:
> > >>
> > >>> On Tue, Oct 4, 2016 at 11:57 AM, John D. Ament <
> [hidden email]>
> > >>> wrote:
> > >>>>
> > >>>> @Hiram
> > >>>>
> > >>>> The website branding says otherwise (take a look at the top right
> > >>>
> > >>> corner).
> > >>>
> > >>>
> > >>> That symbol on the top is just a link to the following:
> > >>>
> > >>> "Like hawtio? It’s part of a community of Red Hat projects. Learn
> more
> > >>> about Red Hat and our open source communities:"
> > >>>
> > >>>
> > >>>
> > >>>  No other implications from what I see.
> > >>>
> > >>
> > >
> >
> >
> >
> > --
> > Clebert Suconic
> >
>
123