panel: Application Instances list for Operator #1256
Labels
No labels
ambition
application-offering
ambition
configure-applications
ambition
front-end
ambition/install-applications
ambition
security
ambition
switch-host
ambition
update-applications
ambition
user-management
blocked
component
api-service
component
fediversity-panel
component
nixops4
documentation
points
0
points
0.5
points
1
points
13
points
2
points
21
points
3
points
34
points
5
points
55
points
8
points
infinite
productisation
project-management
question
role
application-developer
role
application-operator
role
hosting-provider
role
maintainer
security
technical debt
testing
type
bug
type
deliverable
type
key-result
type
objective
type
task
type
unclear
type
user-story
user experience
No milestone
No project
No assignees
2 participants
Notifications
Due date
No due date set.
Dependencies
No dependencies set
Reference
fediversity/fediversity#1256
Loading…
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
As an operator using the panel
I want to view all the fediversity application instances that I operate
so that I can
I do not want to
This is related to
note that, so far, instances have not been a billing unit: as open-source software fediversity itself technically has no notion of billing, and procolix's productization effort similarly intends to have things as a package deal spanning a single set of application instances (i.e. one instance per application, without even necessarily an option to opt out of individual applications).
this would imply to me applications have no dependencies (ancilliary nodes), or alternatively, that one would be free to reconfigure instances into deployment groups on the same set of ancilliary nodes.
that maybe raises the question: in what ways does the current model not suffice? i'm not so much claiming that it would, but if we could understand the operator needs not sufficiently serviced by this model better, then that could perhaps help better inform how to approach things here.
Good to know.
I don't understand this. I'm not trying to change the model here, I'm trying to build an interface for the current model.
nodes in an (abstract) group are currently tracked mostly in
core/setups/groups/<group>.nix, which should be exposed thrucore#lib.baseGroups.core/setups/hosts-common.nixthen uses this to deploy these.such deployed machines may be tracked mostly in terraform state, tho telemetry may also give a sense of the nodes in a set.
that TF state may be queried using some of the TF verbs exposed by
core/.to set expectations, there is no such notion of instances at:
core/lib/data-model.nixcore/setups/hosts-common.nixImo applications do not have other applications as dependencies. I just see no example of where they would. Please provide an example. SSO is not an example of inter-applications dependency, as SSO relations aren't configured through nix, at least I haven't seen a case yet where Nix Code provisions e.g. an OIDC client.
These are all concepts we wanna completely hide from the operator, right?
panel: application instances list for operatorto panel: Application Instances list for Operatorthis sounds like a matter of semantics to me, so i've filed #1280 to try and address that - because presumably we agree that applications may depend on services like databases, and in our data model those currently have ended up utilizing our
applicationdata model entity as well.we currently handle all that thru nix, see
core/resource/oidc/.i think they would facilitate exposing the info you do want exposed to them - in that sense i'm not sure i'd see them as completely separate.
From the glossary:
Then the data model needs to be adapted to the glossary. A database is not user-facing…
Good to know, I'll look at that.
i think this is implemented now ✌️