feat!: remove distributed mode #341
Conversation
Signed-off-by: ThibaultFy <thibault.fouqueray@gmail.com>
ThibaultFy
force-pushed
the
feat/decreciate-distributed-mode
branch
from
115dc73
to
d370bf5
Compare
Signed-off-by: ThibaultFy <thibault.fouqueray@gmail.com>
Signed-off-by: ThibaultFy <thibault.fouqueray@gmail.com>
|
/e2e --tests sdk |
|
End to end tests: ✔️ SUCCESS |
Signed-off-by: ThibaultFy <thibault.fouqueray@gmail.com>
Signed-off-by: ThibaultFy <thibault.fouqueray@gmail.com>
Signed-off-by: ThibaultFy <thibault.fouqueray@gmail.com>
Signed-off-by: ThibaultFy <thibault.fouqueray@gmail.com>
Signed-off-by: ThibaultFy <thibault.fouqueray@gmail.com>
Signed-off-by: ThibaultFy <thibault.fouqueray@gmail.com>
|
/e2e --tests sdk,mnist,camelyon,substrafl |
|
End to end tests: ✔️ SUCCESS |
Signed-off-by: ThibaultFy <thibault.fouqueray@gmail.com>
| subject: | ||
| organizations: | ||
| - {{ .Values.fabric.organization | default "Owkin" }} |
There was a problem hiding this comment.
Not sure, but maybe we want to keep this only with the default value?
There was a problem hiding this comment.
Mmmmh I'm not sure what this value does. I saw fabric no I deleted it... More info on this @guilhem-barthes ?
docs/api.md
Outdated
| ## Consuming the gRPC API | ||
|
|
||
| When running in development mode, the orchestrator exposes a gRPC endpoint on port 9000. | ||
| gRPC reflection is enabled, and protobuf definitions are in [lib/assets](../lib/assets) directory. | ||
|
|
||
| **distributed mode**: requests **MUST** have the following 3 headers set: | ||
| Requests **MUST** have the following header set: |
There was a problem hiding this comment.
From what I understand from the documentation, this whole paragaph about headers can be dropped I think?
docs/architecture.md
Outdated
| The same level of consistency is provided in standalone mode by leveraging [postgresql's isolation level](https://www.postgresql.org/docs/current/transaction-iso.html#XACT-SERIALIZABLE). | ||
| By enforcing serializable transaction, we make sure we don't rely on inconsistent data. |
There was a problem hiding this comment.
I think we want to keep these bullet points from the concurrent request processing.
There was a problem hiding this comment.
Not sure why this file is added here? 🤔
Was is not generated from a previous PR?
There was a problem hiding this comment.
Ah yes I didn't see it. Yes it surely because of that !
| @@ -8,34 +8,30 @@ Unless specified, all settings are mandatory. | |||
|
|
|||
| | Env Var | mode | type | usage | | |||
There was a problem hiding this comment.
shouldn't we just drop the mode column in this table ? 🤔
| @@ -9,6 +9,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0 | |||
|
|
|||
| ### Removed | |||
|
|
|||
| - BREAKING: remove all code related to the `distributed` mode, and mentions in schemas and documentation ([#341](https://github.com/Substra/orchestrator/pull/341)) | |||
There was a problem hiding this comment.
Is it breaking if we already removed the functionality? 🤔
There was a problem hiding this comment.
Did we ? I was thinkin that as the code was still there, the feature was too (even if we already cut some access to it). But if we consider the feature was already deactivated, then I fully agree with you :)
Signed-off-by: ThibaultFy <thibault.fouqueray@gmail.com>
Signed-off-by: ThibaultFy <thibault.fouqueray@gmail.com>
ThibaultFy
force-pushed
the
feat/decreciate-distributed-mode
branch
from
9546535
to
c6cf08f
Compare
Signed-off-by: ThibaultFy <thibault.fouqueray@gmail.com>
|
/e2e --tests sdk,mnist,camelyon,substrafl |
Description
closes FL-1329
How has this been tested?
Checklist