Skip to content

Comments

Upgrade chart to latest#43

Closed
mihaelabalas84 wants to merge 163 commits intofm-devopsfrom
master
Closed

Upgrade chart to latest#43
mihaelabalas84 wants to merge 163 commits intofm-devopsfrom
master

Conversation

@mihaelabalas84
Copy link

No description provided.

tsurdilo and others added 30 commits April 8, 2024 09:50
Signed-off-by: Tihomir Surdilovic <tihomir@temporal.io>
This is useful for patch releases on older release lines.
…alio#488)

* Bumps server version to the specified appVersion
* Use chart.appversion instead of image.tag from values in deployment spec
* Allow overriding deployment spec with image.tag

---------

Co-authored-by: Kshitij <kshitij.tulsyan@observe.ai>
…io#394)

1. remove web ui config in values.yaml
2. remove web config volume in web-deployment.yaml
3. remove web-config config map
4. remove line 280 since the install bash didn't configure web ui auth
5. update the document for web ui configuration with env variable

Co-authored-by: Rob Holland <rob.holland@gmail.com>
* Require tags for server, admintools and ui.

Don't use server tag for admintools, it's versioned separately now.
Allow templated values inside the values.yaml for web annotations
Signed-off-by: Valentin Zayash <valioozz@gmail.com>
…poralio#485)

* Update Cassandra host URLs to remove the ".cluster.local" suffix
* Configure SQL_TLS environment variables in server-job
abarganier and others added 29 commits April 17, 2025 18:00
* [server-job] update TEMPORAL_ADDRESS

Signed-off-by: luis.costa <luis.costa@talkdesk.com>
Temporal core allows for the registration and configuration of custom data stores.
However, there is a disparity in how the helm chart translates persistence
configuration between helm values and the resulting configmap.

As an example, the temporal config for a custom driver for the default store may look
as so:

```
persistence:
  defaultStore: my-custom-store
  visibilityStore: es-visibility
  numHistoryShards: 4
  datastores:
    my-custom-store:
      customDatastore:
        name: "my-custom"
        options:
          hosts: "mycustom.example.com"
          keyspace: "temporal"
```

Translating to helm values would look like

```
server:
  config:
    persistence:
      defaultStore: my-custom-store
      additionalStores:
        my-custom-store:
          customDatastore:
            name: "my-custom"
            options:
              hosts: "mycustom.example.com"
              keyspace: "temporal"
```

This gets us most of the way there, but there are a few problems:

1. The charts always assume defaultStore = "default" and thus have opinions
   about subordinate structures such as default.driver.  This results in objects
   such as Secrets and Jobs being incorrectly emitted for default.driver = "cassandra" even
   though we are trying to specify an alternate store.  This logic should probably
   be $defaultStore.driver and $visibilityStore.driver instead.
2. The additionalStores configuration does not lend itself to first-class
   representation in conditional statements, even though it can be used to
   emit the resulting configmap correctly.

One could argue that we should clean up the helm representation to more closely align
with the temporal-server yaml, but this is a bigger task (and a breaking change).  Instead,
this patch ties into the existing model of using $store.driver = {cassandra, sql, elasticsearch}
and adds a 4th first-class type "custom."

The user may configure this as so:

```
server:
  config:
    persistence:
      default:
        driver: custom
      visibility:
        driver: custom
```

And in doing so, we instruct the various templates to elide the objects (Jobs, Secrets, etc.)
related to that store, leaving it to the user to supply the rest of the required configuration
using a combination of additionalStores, additionalEnv, additionalVolumeMounts, etc., to inject
configuration and secrets.  They would also, of course, be responsible for any DDL setup or
migration.

Signed-off-by: Greg Haskins <greg@manetu.com>
This change was made by an automated process to ensure all GitHub Actions workflows have explicitly defined permissions as per best practices.
* Update the contributing docs for clarity.

Help to guide users on what kind of changes we will accept.

---------

Co-authored-by: Tom Wheeler <tom@temporal.io>
what: fixed a very minor typo in README.md
* rough starting point

* give users a few toggles

* no init container startup

* Modify server config t

Updated server configuration to support sprig

* legacy -> modern

* misc fixes and tests

* use cassandra too if it exists and a few bug fixes with tests

* few bug fixes

* fix a few unit tests

* addressing pr comments. not yet tested

* updating unit tests

* update go tests

* rm cassandra check changes they use the cassandra image

* Update charts/temporal/templates/server-deployment.yaml

* address pr feedback and add some unit tests
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.