Skip to content

Drop monasca and senlin from the schema - #112

Merged
github-actions[bot] merged 2 commits into
developfrom
jim.lin/fix/drop-monasca
Sep 14, 2026
Merged

github-actions[bot] merged 2 commits into
developfrom
jim.lin/fix/drop-monasca

Conversation

@Eandalf-Bigstack

Copy link
Copy Markdown
Contributor

What type of PR is this?

/kind cleanup


Which issue(s) this PR fixes?

Part of bigstack-oss/cubecos#772, specifically bigstack-oss/cubecos#771 (COS UI > Home > Health page after the OpenStack, InfluxDB and Ceph upgrade).


What this PR does?

Removes two services from the schema that CubeCOS no longer ships, so the API stops describing things that are gone.

monasca — removed from CubeCOS by bigstack-oss/cubecos#672 phase 4. Four places still described it:

where what it claimed
metrics module group listed monasca beside telegraf and grafana, so a client rendering that group showed a module the cluster does not have
two module-name enums GET /v1/modules and the repair endpoint both accepted "monasca" as a valid module name
monasca.debug.enabled a documented tuning for a service with nothing left to configure

senlin — removed from CubeCOS by d4550c91, back on the yoga train. The schema was only half-updated at the time: senlin came out of the module-name enums but stayed in the businessLogic module list and in a senlin.debug.enabled tuning. That asymmetry is why every generated SDK is already clean while the schema still advertised the service. CubeCOS' own cluster check now reports BusinessLogic as watcher alone.

Schema only — no endpoint, request or response shape changes.


Test results (optional)

1). docs.yaml still parses as YAML after both removals, and grep -c monasca / grep -c senlin return 0.

2). Regenerating the TypeScript SDK from this schema in cube-cos-ui produces:

  • for the monasca commit: two JSDoc @type unions and two enum members removed — the exact mirror of what a34418d added for metricsDb
  • for the senlin commit: sdk/api.ts byte-for-byte unchanged, which is the evidence that senlin was already absent from the module enums and only the non-generating parts of the schema still mentioned it

3). Regenerating api/docs.json in cube-cos-api from this schema yields zero monasca and zero senlin references. That file is embedded into the binary via api/docs.go's go:embed, so this schema change does reach the shipped artifact.


Companion PRs

This PR is the base for the other two; both pin their submodule at this branch and want re-pinning to the merged SHA before they land.

  • cube-cos-api — the Go half: the metric queries and the module list
  • cube-cos-ui — the UI half: the health-page labels and the regenerated SDK

🤖 Generated with Claude Code

@SekiXu SekiXu left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Approving. Both removals are the schema catching up to code that already moved, which is the direction that cannot break a client of a shipped build:

  • senlin — already absent from cube-cos-api (git grep -i senlin on jim.lin/fix/drop-monasca returns nothing), and senlin.debug.enabled is already gone from cubecos/develop. So the schema was the stale copy, not the leading edge.
  • monasca — matches cube-cos-api#656 and cubecos#1457 exactly: the metrics module group loses monasca and keeps telegraf/grafana, the two module-name enums drop it, and monasca.debug.enabled goes at the same time cubecos#1457 deletes the CONFIG_TUNING_BOOL that registers it.

Worth knowing for the merge: this is a clean fast-forward from develop, so a true ff merge puts develop on 3d9065db — which is what cube-cos-api#656 already pins its submodule at. That order works without a re-pin.

@Eandalf-Bigstack Eandalf-Bigstack added the done Merge the pull request label Sep 14, 2026
Eandalf-Bigstack and others added 2 commits September 14, 2026 12:52
Monasca was removed from CubeCOS by bigstack-oss/cubecos#672 phase 4. Four
places in the schema still described it as something the API can report on or
an operator can tune:

  the metrics module group   listed monasca alongside telegraf and grafana, so
                             a client rendering that group showed a module the
                             cluster no longer has. prometheus and thanos moved
                             to the metricsDb group in a34418d; this is the
                             other half of that tidy-up.
  two module name enums      GET /v1/modules and the repair endpoint both
                             accepted "monasca" as a valid module name.
  monasca.debug.enabled      a documented tuning for a service with nothing
                             left to configure.

Schema only -- no endpoint, response shape or field changes, and docs.yaml
still parses. The api/docs.json regeneration and the module list in Go live in
cube-cos-api and land with the submodule bump there.

Signed-off-by: Jim Lin <jim.lin@bigstack.co>
Co-authored-by: Eandalf <clinah@connect.ust.hk>
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Senlin was removed from CubeCOS by d4550c91, back on the yoga train. The schema
was only half-updated at the time: senlin came out of the module-name enums but
stayed in two other places, which is why every generated SDK is already clean
while the schema still describes the service.

  the businessLogic module list   the services example still showed senlin
                                  beside watcher. cubecos' cluster check now
                                  reports BusinessLogic as watcher alone.
  senlin.debug.enabled            a documented tuning for a service that has
                                  not shipped for two release trains.

Neither feeds the TypeScript generator -- one is a response example, the other
a tuning document -- so consumers see no SDK change from this, only a schema
that stops advertising something that is gone.

Same shape as the monasca removal in the previous commit, and found while
doing it.

Signed-off-by: Jim Lin <jim.lin@bigstack.co>
Co-authored-by: Eandalf <clinah@connect.ust.hk>
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
@Eandalf-Bigstack Eandalf-Bigstack added done Merge the pull request and removed done Merge the pull request labels Sep 14, 2026
@github-actions
github-actions Bot merged commit 6217dc4 into develop Sep 14, 2026
4 checks passed
@Eandalf-Bigstack
Eandalf-Bigstack deleted the jim.lin/fix/drop-monasca branch September 14, 2026 04:53
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

done Merge the pull request

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants