You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@nifi.apache.org by Russell Bateman <ru...@windofkeltia.com> on 2017/07/31 16:45:49 UTC
New "hierarchical" controller service system confusing
Friends,
I find the new (well, since 0.7.x -> 1.x) way of associating controller
services based on the process group a bit disorienting. When one follows
the right arrow from a consuming processor and lands on the Process
Group Configurationpage where the services are configured, one is
obliged to click on the Generaltab to figure out whether the controller
service is at the root (the case for services whose flows were upgraded
from the 0.7.x world) or for a specific process group (the case for
configuration accomplished post-upgrade/native 1.x).
I realize that this will be a dwindling problem, but there's enough
white space in this dialog to justify putting the Process Group Name to
the right of the dialog title (Process Group Configuration) while on the
Controller Servicestab, even without the momentary confusion of origin
(that is, a flow from 0.7.x vs. a "native" 1.x flow). So, I consider
that this enhancement holds regardless of whether upgraded flows are
present or not.
At the same time, I'm tempted to ask how, when looking on the Controller
Services tab/page, one figures out which processor instance(s) is (are)
consuming a particular controller service I'm looking at?
Russ
Re: New "hierarchical" controller service system confusing
Posted by Russell Bateman <ru...@windofkeltia.com>.
Thanks for the suggestion, Matt. I created NIFI-4251.
On 08/01/2017 09:39 AM, Matt Gilman wrote:
> Russell,
>
> Thanks for the suggestion on the improved Controller Service UX. Would
> you mind filing a JIRA for this improvement?
>
> In 1.x, the user can see the components referencing a Controller
> Service in 1 of 3 places. The references are shown in the read-only
> details dialog, the service configuration dialog, and the
> enable/disable service dialog. The references will include Processors,
> Reporting Tasks, and Controller Services (and components that
> reference those services and so on).
>
> Thanks
>
> Matt
>
>
> On Mon, Jul 31, 2017 at 12:45 PM, Russell Bateman
> <russ@windofkeltia.com <ma...@windofkeltia.com>> wrote:
>
> Friends,
>
> I find the new (well, since 0.7.x -> 1.x) way of associating
> controller services based on the process group a bit disorienting.
> When one follows the right arrow from a consuming processor and
> lands on the Process Group Configurationpage where the services
> are configured, one is obliged to click on the Generaltab to
> figure out whether the controller service is at the root (the case
> for services whose flows were upgraded from the 0.7.x world) or
> for a specific process group (the case for configuration
> accomplished post-upgrade/native 1.x).
>
> I realize that this will be a dwindling problem, but there's
> enough white space in this dialog to justify putting the Process
> Group Name to the right of the dialog title (Process Group
> Configuration) while on the Controller Servicestab, even without
> the momentary confusion of origin (that is, a flow from 0.7.x vs.
> a "native" 1.x flow). So, I consider that this enhancement holds
> regardless of whether upgraded flows are present or not.
>
> At the same time, I'm tempted to ask how, when looking on the
> Controller Services tab/page, one figures out which processor
> instance(s) is (are) consuming a particular controller service I'm
> looking at?
>
> Russ
>
>
Re: New "hierarchical" controller service system confusing
Posted by Matt Gilman <ma...@gmail.com>.
Russell,
Thanks for the suggestion on the improved Controller Service UX. Would you
mind filing a JIRA for this improvement?
In 1.x, the user can see the components referencing a Controller Service in
1 of 3 places. The references are shown in the read-only details dialog,
the service configuration dialog, and the enable/disable service dialog.
The references will include Processors, Reporting Tasks, and Controller
Services (and components that reference those services and so on).
Thanks
Matt
On Mon, Jul 31, 2017 at 12:45 PM, Russell Bateman <ru...@windofkeltia.com>
wrote:
> Friends,
>
> I find the new (well, since 0.7.x -> 1.x) way of associating controller
> services based on the process group a bit disorienting. When one follows
> the right arrow from a consuming processor and lands on the Process Group
> Configuration page where the services are configured, one is obliged to
> click on the General tab to figure out whether the controller service is
> at the root (the case for services whose flows were upgraded from the 0.7.x
> world) or for a specific process group (the case for configuration
> accomplished post-upgrade/native 1.x).
>
> I realize that this will be a dwindling problem, but there's enough white
> space in this dialog to justify putting the Process Group Name to the right
> of the dialog title (Process Group Configuration) while on the Controller
> Services tab, even without the momentary confusion of origin (that is, a
> flow from 0.7.x vs. a "native" 1.x flow). So, I consider that this
> enhancement holds regardless of whether upgraded flows are present or not.
>
> At the same time, I'm tempted to ask how, when looking on the Controller
> Services tab/page, one figures out which processor instance(s) is (are)
> consuming a particular controller service I'm looking at?
>
> Russ
>