I like the on by default in docker, off by default in helm compromise -
that said it still makes sense to split it for prod if desired cause it
will not use the same service/httproute at all IMHO, one will be public
like versus the other can stay internal most of the time.

Romain Manni-Bucau
@rmannibucau <https://x.com/rmannibucau> | .NET Blog
<https://dotnetbirdie.github.io/> | Blog <https://rmannibucau.github.io/> | Old
Blog <http://rmannibucau.wordpress.com> | Github
<https://github.com/rmannibucau> | LinkedIn
<https://www.linkedin.com/in/rmannibucau> | Book
<https://www.packtpub.com/en-us/product/java-ee-8-high-performance-9781788473064>
Javaccino founder (Java/.NET service - contact via linkedin)


Le lun. 1 juin 2026 à 20:14, Yong Zheng <[email protected]> a écrit :

> Hello,
>
> Why not put them in the same repo but keep them separate (similar to spark
> plugin which doesn’t get include automatically). Same apply to helm as
> people can decide on what to deploy and avoid UI component if desired.
>
> As the UI is SPA, the actually calls are from client browser as opposed to
> UI pod, the usability may be limited there in certain use cases. I would
> start a different ML for this topic.
>
> Thanks,
> Yong Zheng
>
> > On Jun 1, 2026, at 11:52 AM, Dmitri Bourlatchkov <[email protected]>
> wrote:
> >
> > Hi All,
> >
> > The idea of moving the UI to the main repo SGTM.
> >
> > I tend to agree with Romain, that the UI should be easy to _disable_
> > because in production environments it may not be intended to run at all,
> or
> > perhaps not on the same nodes as the main Polaris server. The simplicity
> of
> > the UI (static pages + REST API via JS) may not justify always running it
> > from the SRE perspective, I guess.
> >
> > I'm sure there are multiple ways to tackle this in helm. For a start, a
> > simple on/off helm value might be sufficient, though.
> >
> > Regarding defaults, how about ON by default in plain docker (for local
> > runs), but OFF by default in helm (mostly for "prod" environments)?
> >
> > Cheers,
> > Dmitri.
> >
> >> On Mon, Jun 1, 2026 at 11:44 AM Romain Manni-Bucau <
> [email protected]>
> >> wrote:
> >>
> >>> Le lun. 1 juin 2026 à 17:35, Robert Stupp <[email protected]> a écrit :
> >>>
> >>> But isn't the concern solved with the configuration option "enable web
> >> UI =
> >>> true|false"?
> >>>
> >>
> >> if default is false it does but breaks the easy by default no?
> >>
> >>
> >>>
> >>> With a standalone web UI, as it exists today, everybody can run it and
> >>> perform "click-ops."
> >>> Similar for custom built web UIs.
> >>> Or if some user extracts the static files and serves their own instance
> >>> using a local http daemon.
> >>>
> >>> The web UI doesn't allow anything that cannot be done via another
> >>> interface, such as a terminal and `curl`.
> >>>
> >>
> >> yes but it is less tempting enough to be sufficient for me
> >>
> >>
> >>>
> >>> With that, I'm not sure maintaining multiple different distributions
> >> really
> >>> helps lock people out of web UIs ("click-ops").
> >>>
> >>
> >> from my windows integrating the ui is not helping since you need the
> >> database anyway so maybe the ui outside is sufficient too else the all
> in
> >> one with h2 or pg will help, i'm not clear on it
> >>
> >>
> >>>
> >>> Robert
> >>>
> >>>
> >>> On Mon, Jun 1, 2026 at 5:15 PM Romain Manni-Bucau <
> [email protected]
> >>>
> >>> wrote:
> >>>
> >>>> Le lun. 1 juin 2026 à 17:12, Robert Stupp <[email protected]> a écrit :
> >>>>
> >>>>> Hi Romain,
> >>>>>
> >>>>> Disabling the console (web UI) is easy once it becomes part of the
> >>>>> polaris-server.
> >>>>>
> >>>>> To clarify, the web UI is a static web resource, it doesn't execute
> >>>> web-UI
> >>>>> code in Quarkus.
> >>>>> The UI uses the same authentication and authorization mechanisms that
> >>> any
> >>>>> REST client uses; it is just a UI on top of that.
> >>>>>
> >>>>
> >>>>
> >>>> Yep, this is my issue, it is trivial to use and break the "all as
> code"
> >>>> pattern and get back to click-ops pattern.
> >>>> Being API only is less tempting so I'm trying to stick to that.
> >>>>
> >>>> *Still thinking out loud* but having 3 images sounds the best of all
> >>>> worlds:
> >>>>
> >>>> * server
> >>>> * frontend/ui
> >>>> * all-in-one (== demo)
> >>>>
> >>>> It is the pattern used by most products (all in one can even embed
> >>> postgres
> >>>> to really be ready to use) and everybody will be happy at a very low
> >> cost
> >>>> in terms of dev/maintenance and infra IMHO.
> >>>>
> >>>> Hope it makes sense.
> >>>>
> >>>>
> >>>>>
> >>>>> Hope this helps,
> >>>>>
> >>>>> Robert
> >>>>>
> >>>>>
> >>>>> On Mon, Jun 1, 2026 at 4:33 PM Romain Manni-Bucau <
> >>> [email protected]
> >>>>>
> >>>>> wrote:
> >>>>>
> >>>>>> Hi,
> >>>>>>
> >>>>>> as an user I have a small request: can it be disabled by default
> >> (or
> >>>> made
> >>>>>> unusable by disabling the authn) or at least be made easy to
> >> disable
> >>>>>> (corrollar being: can we get a headless docker image, solves my
> >>> concern
> >>>>>> too)?
> >>>>>> rational is that when you have everything automated around polaris
> >>> you
> >>>>>> don't want the UI to be able to mess up things nor render a
> >> different
> >>>>> view.
> >>>>>> not my usage but an in between readonly enforced mode - even for
> >>> admin
> >>>>>> users - can be useful I think so maybe 3 modes in terms of toggle:
> >>>> Full,
> >>>>>> ReadOnly, Off or alike.
> >>>>>>
> >>>>>> Shouldn't be crazy to do with quarkus but think it is an important
> >>>>> feature
> >>>>>> to integrate it cleanly as a global solution.
> >>>>>>
> >>>>>> the tricky part is to make the prod and dev defaults converging but
> >>>>> quarkus
> >>>>>> has a profile option by default which can be the way to solve it.
> >>>>>>
> >>>>>> hope it makes sense.
> >>>>>>
> >>>>>> Romain Manni-Bucau
> >>>>>> @rmannibucau <https://x.com/rmannibucau> | .NET Blog
> >>>>>> <https://dotnetbirdie.github.io/> | Blog <
> >>>> https://rmannibucau.github.io/
> >>>>>>
> >>>>>> | Old
> >>>>>> Blog <http://rmannibucau.wordpress.com> | Github
> >>>>>> <https://github.com/rmannibucau> | LinkedIn
> >>>>>> <https://www.linkedin.com/in/rmannibucau> | Book
> >>>>>> <
> >>>>>>
> >>>>>
> >>>>
> >>>
> >>
> https://www.packtpub.com/en-us/product/java-ee-8-high-performance-9781788473064
> >>>>>>>
> >>>>>> Javaccino founder (Java/.NET service - contact via linkedin)
> >>>>>>
> >>>>>>
> >>>>>> Le lun. 1 juin 2026 à 15:28, Jean-Baptiste Onofré <[email protected]
> >>>
> >>> a
> >>>>>> écrit :
> >>>>>>
> >>>>>>> Hi,
> >>>>>>>
> >>>>>>> Yes, that is indeed part of the proposal.
> >>>>>>>
> >>>>>>> The goal is to provide the UI as soon as the Polaris server
> >> starts
> >>> to
> >>>>>>> ensure a better experience for our users.
> >>>>>>>
> >>>>>>> Regards,
> >>>>>>> JB
> >>>>>>>
> >>>>>>> On Mon, Jun 1, 2026 at 3:12 PM Robert Stupp <[email protected]>
> >>> wrote:
> >>>>>>>>
> >>>>>>>> +1
> >>>>>>>>
> >>>>>>>> Could we then bundle the Console with the Polaris server?
> >>>>>>>>
> >>>>>>>> On Mon, Jun 1, 2026 at 8:01 AM Jean-Baptiste Onofré <
> >>>> [email protected]
> >>>>>>
> >>>>>>> wrote:
> >>>>>>>>
> >>>>>>>>> Hi everyone,
> >>>>>>>>>
> >>>>>>>>> As mentioned on the dev mailing list a few days ago, I am
> >>>> currently
> >>>>>>>>> preparing the first release of the Polaris Console. There is
> >>>>>>>>> significant interest in the Console, and it is a key
> >> component
> >>> of
> >>>>> our
> >>>>>>>>> efforts to improve the onboarding experience.
> >>>>>>>>>
> >>>>>>>>> After discussing this with Yong, who started to help with the
> >>>>>> console,
> >>>>>>>>> we believe that moving the Console into the main Polaris
> >>>> repository
> >>>>>>>>> would provide better visibility and encourage further
> >>>>> contributions.
> >>>>>>>>> This would also allow it to be released as part of the main
> >>>> Polaris
> >>>>>>>>> release cycle. Additionally, the Console could be included as
> >>> an
> >>>>>>>>> optional component in the main Helm chart, allowing users to
> >>>> choose
> >>>>>>>>> whether to enable it.
> >>>>>>>>>
> >>>>>>>>> Thoughts?
> >>>>>>>>>
> >>>>>>>>> Regards,
> >>>>>>>>> JB
> >>>>>>>>>
> >>>>>>>
> >>>>>>
> >>>>>
> >>>>
> >>>
> >>
>

Reply via email to