Skip to content

Schema refactor#798

Merged
robholland merged 6 commits intov1from
rh-schema-refactor
Dec 10, 2025
Merged

Schema refactor#798
robholland merged 6 commits intov1from
rh-schema-refactor

Conversation

@robholland
Copy link
Copy Markdown
Contributor

What was changed

Adjusted the schema management to work per store. Use temporal-elasticsearch-tool when available.

Why?

This allows people to opt-in to schema management per store rather than on/off globally. The configuration also now better matches our behaviour. For example previously if setup was enabled it would accidentally also perform upgrades even if upgrade was disabled. Now we have just one toggle that controls whether we manage the schema. Users can disable that after initial install to prevent future upgrades.

Elasticsearch now supports upgrades like Cassandra/SQL do. Using temporal-elasticsearch-tool (on supported images after 1.30 release) will allow use of AWS authentication for Elasticsearch which the current curl based automation did not.

This uses temporal-elasticsearch-tool where available, and a shell equivalent
otherwise.

Elasticsearch will now perform schema updates like Cassandra/SQL do.

Simplify the broken setup vs upgrade toggles which did not properly control the
behaviour. We don't have a way to with fixed toggles to separate creation from
upgrade. Use a single toggle for management, the user can toggle that after
install if they want to stop automatic upgrades.
@robholland robholland requested a review from a team as a code owner December 10, 2025 14:31
@robholland robholland merged commit e405d12 into v1 Dec 10, 2025
2 checks passed
@robholland robholland deleted the rh-schema-refactor branch December 10, 2025 16:18
robholland added a commit that referenced this pull request Dec 18, 2025
* Remove prometheus dependency (#534)
* Remove prometheus from dependencies and examples
* Remove grafana dependency (#526)
* Adjust some helper wiring.
* Update Server Configmap (#794)
* .ENV to env and add config template var
* Simplify chart (#784)

Simplify our helm chart, remove sub-charts.

* Install cluster variations as part of CI. (#796)
* Skip dockerize for 1.29.* images.

This allows us to use server templated config files safely on 1.29.* and 1.30.*
so that we can rollforward/backward without having to maintain two sets of configs.

* Schema refactor (#798)
* Refactor schema management.

This uses temporal-elasticsearch-tool where available, and a shell equivalent
otherwise.

Elasticsearch will now perform schema updates like Cassandra/SQL do.

Simplify the broken setup vs upgrade toggles which did not properly control the
behaviour. We don't have a way to with fixed toggles to separate creation from
upgrade. Use a single toggle for management, the user can toggle that after
install if they want to stop automatic upgrades.

* v1 doesn't have deps. (#800)
* Push to the branch on which the action was triggered.
* Update Chart to 1.0.0-rc.1.
* Add some upgrading notes. (#805)
* Fix version specifier. (#806)
* Change image pull secrets to array instead of map (#701)
* Correct config option name. (#807)
* Pin helm unittest version due to helm 3 vs 4 compat issue. (#808)
* Switch ct to testing against main.

---------

Co-authored-by: Brian Kopp <briankopp.usa@gmail.com>
Co-authored-by: PhilHenning <phillyp.henning@gmail.com>
Co-authored-by: LeoDiazL <leo@bitovi.com>
Co-authored-by: Phil Henning <philh@bitovi.com>
Co-authored-by: Alex Stanfield <13949480+chaptersix@users.noreply.github.com>
Co-authored-by: Temporal Data <commander-data@temporal.io>
Co-authored-by: Matthew Brandman <matthb6@gmail.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants