On Friday, January 31, 2014, artnaseef <[email protected]> wrote:
Another thing - "22Mb legacy turd" is not a technical argument (at least, I
don't recognize it as one).
Memory usage is important to a message broker - which has to spool to disk
as soon as it's out of RAM - which drastically affects performance. BTW
that 22mb turd is just the compressed disk size of the code, never mind the
runtime overhead
I'm disappointed.
If there are concerns with maintenance, what are they?
No one wants to maintain it for one; it's been dormant for years; plus it's
kinda crappy.
Second there's a much better solution now. Though It'll probably annoy you
if I mention it out loud.
Third, jolokia is probably enough these days for runtime sevices (nice,
lean REST/JSON API to the mbeans). Any devops can knock
themselves out with any script/tool/web page with that.
BTW before the Savoir/Talend zealots jump in with further conspiracy
theories; jolokia isn't a Fuse/Red Hat project at all, we've no committers.
It's just a great, lean solution to the management issue.
I believe there are
currently only 3 outstanding Jira entries for the console.
Look closer. But to be honest the code's been neglected with little
community for so long, folks probably stopped raising anything but bugs &
security issues many years ago
Right? It's old
- so what, it's not older than ActiveMQ ;-).
It's not far off really - but key pieces of ActiveMQ get rewritten &
improved all the time (eg level db). The web console is the same old crap
it's always been. I wrote quite a bit of it many years ago; I apologise for
it profusely- but it still deserves a sympathetic burial.
Maybe struts 1.0 is up for a comeback too?
I love that so many people are passionate about ActiveMQ.
Me too!
I wish that
passion were being put into making it better and moving it forward
Didn't you spot that quite a few of us have been putting our passion into a
new amazing console for ActiveMQ, based on modern lightweight
technology - that's
not 21Mb of compressed turd? Or do you just discount all open source
projects without an Apache PMC in principle without even looking?
rather
than making arguments without merit and laying out criticism - very
disappointing.
I'm disappointed you're disappointed
So, back to defining the problem. All I've seen so far is the list of
security concerns from Hiram - thank you Hiram. Anything else? I do
believe I've read comments about difficulty maintaining it. Is that true,
or just an exaggerated expression of frustration?
When code is dying and losing its utility & before it's buried in the
Attic, the compassionate thing to do is see if it can survive without life
support. Moving it to a sub project seems the right thing to do. If you can
attract a community around it - great, fair play to you. If not, the attic
is ok too (most code will end up there one day, it's just a question of
time).
Apache is all about communty and right now I don't see any around the
old web console code. Moving it to a sub project will settle the argument
once and for all in a fair, Apache Way. Whatever the outcome, the community
wins
--
View this message in context:
http://activemq.2283324.n4.nabble.com/ActiveMQ-Console-let-s-get-the-problem-defined-tp4677105p4677224.html
Sent from the ActiveMQ - Dev mailing list archive at Nabble.com.
