This fixes a bug in the pruning (soft-delete). The pruning was a bit too
aggressive, as it also pruned executions that weren't in an end state
yet. This only becomes an issue if there are long-running executions
(e.g. workflow with Wait node) or the prune parameters are set to keep
only a tiny number of executions.
`tsBuildInfoFile` is supposed to be relative to `tsconfig` like `outDir`
is.
Because of this, we are currently saving the TS incremental build cache
for all packages in the same file. This is likely causing issues where
the built backend code sometimes does not accurately map to the current
source code.
This PR changes the incremental build setup to keep the cache in
individual `dist` folders, like it used to be up until a 2 months ago,
before https://github.com/n8n-io/n8n/pull/6816.
This PR adds a message for queue mode which triggers an external secrets
provider reload inside the workers if the configuration has changed on
the main instance.
It also refactors some of the message handler code to remove cyclic
dependencies, as well as remove unnecessary duplicate redis clients
inside services (thanks to no more cyclic deps)
Depends on https://github.com/n8n-io/n8n/pull/7220 | Story:
[PAY-840](https://linear.app/n8n/issue/PAY-840/introduce-object-store-service-and-manager-for-binary-data)
This PR introduces an object store service for Enterprise edition. Note
that the service is tested but currently unused - it will be integrated
soon as a binary data manager, and later for execution data.
`amazonaws.com` in the host is temporarily hardcoded until we integrate
the service and test against AWS, Cloudflare and Backblaze, in the next
PR.
This is ready for review - the PR it depends on is approved and waiting
for CI.
---------
Co-authored-by: कारतोफ्फेलस्क्रिप्ट™ <aditya@netroy.in>
all commands sent between main instance and workers need to contain a
server id to prevent senders from reacting to their own messages,
causing loops
this PR makes sure all sent messages contain a sender id by default as
part of constructing a sending redis client.
---------
Co-authored-by: कारतोफ्फेलस्क्रिप्ट™ <aditya@netroy.in>
Depends on: https://github.com/n8n-io/n8n/pull/7195 | Story:
[PAY-837](https://linear.app/n8n/issue/PAY-837/implement-object-store-manager-for-binary-data)
This PR includes `workflowId` in binary data writes so that the S3
manager can support this filepath structure
`/workflows/{workflowId}/executions/{executionId}/binaryData/{binaryFilename}`
to easily delete binary data for workflows. Also all binary data service
and manager methods that take `workflowId` and `executionId` are made
consistent in arg order.
Note: `workflowId` is included in filesystem mode for compatibility with
the common interface, but `workflowId` will remain unused by filesystem
mode until we decide to restructure how this mode stores data.
---------
Co-authored-by: कारतोफ्फेलस्क्रिप्ट™ <aditya@netroy.in>
Story: [PAY-846](https://linear.app/n8n/issue/PAY-846) | Related:
https://github.com/n8n-io/n8n/pull/7225
For the S3 backend for external storage of binary data and execution
data, the `getAsStream` method in the binary data manager interface used
by FS and S3 will need to become async. This is a breaking change for
nodes-base.
Story: https://linear.app/n8n/issue/PAY-839
This is a longstanding bug, fixed now so that the S3 backend for binary
data can use execution IDs as part of the filename.
To reproduce:
1. Set up a workflow with a POST Webhook node that accepts binary data.
2. Activate the workflow and call it sending a binary file, e.g. `curl
-X POST -F "file=@/path/to/binary/file/test.jpg"
http://localhost:5678/webhook/uuid`
3. Check `~/.n8n/binaryData`. The binary data and metadata files will be
missing the execution ID, e.g. `11869055-83c4-4493-876a-9092c4708b9b`
instead of `39011869055-83c4-4493-876a-9092c4708b9b`.
Depends on: #7092 | Story:
[PAY-768](https://linear.app/n8n/issue/PAY-768)
This PR:
- Generalizes the `IBinaryDataManager` interface.
- Adjusts `Filesystem.ts` to satisfy the interface.
- Sets up an S3 client stub to be filled in in the next PR.
- Turns `BinaryDataManager` into an injectable service.
- Adjusts the config schema and adds new validators.
Note that the PR looks large but all the main changes are in
`packages/core/src/binaryData`.
Out of scope:
- `BinaryDataManager` (now `BinaryDataService`) and `Filesystem.ts` (now
`fs.client.ts`) were slightly refactored for maintainability, but fully
overhauling them is **not** the focus of this PR, which is meant to
clear the way for the S3 implementation. Future improvements for these
two should include setting up a backwards-compatible dir structure that
makes it easier to locate binary data files to delete, removing
duplication, simplifying cloning methods, using integers for binary data
size instead of `prettyBytes()`, writing tests for existing binary data
logic, etc.
---------
Co-authored-by: कारतोफ्फेलस्क्रिप्ट™ <aditya@netroy.in>
Github issue / Community forum post (link here to close automatically):
---------
Co-authored-by: Omar Ajoue <krynble@gmail.com>
Co-authored-by: कारतोफ्फेलस्क्रिप्ट™ <aditya@netroy.in>
Based on #7065 | Story: https://linear.app/n8n/issue/PAY-771
n8n on filesystem mode marks binary data to delete on manual execution
deletion, on unsaved execution completion, and on every execution
pruning cycle. We later prune binary data in a separate cycle via these
marker files, based on the configured TTL. In the context of introducing
an S3 client to manage binary data, the filesystem mode's mark-and-prune
setup is too tightly coupled to the general binary data management
client interface.
This PR...
- Ensures the deletion of an execution causes the deletion of any binary
data associated to it. This does away with the need for binary data TTL
and simplifies the filesystem mode's mark-and-prune setup.
- Refactors all execution deletions (including pruning) to cause soft
deletions, hard-deletes soft-deleted executions based on the existing
pruning config, and adjusts execution endpoints to filter out
soft-deleted executions. This reduces DB load, and keeps binary data
around long enough for users to access it when building workflows with
unsaved executions.
- Moves all execution pruning work from an execution lifecycle hook to
`execution.repository.ts`. This keeps related logic in a single place.
- Removes all marking logic from the binary data manager. This
simplifies the interface that the S3 client will meet.
- Adds basic sanity-check tests to pruning logic and execution deletion.
Out of scope:
- Improving existing pruning logic.
- Improving existing execution repository logic.
- Adjusting dir structure for filesystem mode.
---------
Co-authored-by: कारतोफ्फेलस्क्रिप्ट™ <aditya@netroy.in>
This PR implements the updated license SDK so that worker and webhook
instances do not auto-renew licenses any more.
Instead, they receive a `reloadLicense` command via the Redis client
that will fetch the updated license after it was saved on the main
instance
This also contains some refactoring with moving redis sub and pub
clients into the event bus directly, to prevent cyclic dependency
issues.
PR adds a new field to the SourceControlPreferences as well as to the
POST parameters for the `source-control/preferences` and
`source-control/generate-key-pair` endpoints. Both now accept an
optional string parameter `keyGeneratorType` of `'ed25519' | 'rsa'`
Calling the `source-control/generate-key-pair` endpoint with the
parameter set, it will also update the stored preferences accordingly
(so that in the future new keys will use the same method)
By default ed25519 is being used. The default may be changed using a new
environment parameter:
`N8N_SOURCECONTROL_DEFAULT_SSH_KEY_TYPE` which can be `rsa` or `ed25519`
RSA keys are generated with a length of 4096 bytes.
# Motivation
In Queue mode, finished executions would cause the main instance to
always pull all execution data from the database, unflatten it and then
use it to send out event log events and telemetry events, as well as
required returns to Respond to Webhook nodes etc.
This could cause OOM errors when the data was large, since it had to be
fully unpacked and transformed on the main instance’s side, using up a
lot of memory (and time).
This PR attempts to limit this behaviour to only happen in those
required cases where the data has to be forwarded to some waiting
webhook, for example.
# Changes
Execution data is only required in cases, where the active execution has
a `postExecutePromise` attached to it. These usually forward the data to
some other endpoint (e.g. a listening webhook connection).
By adding a helper `getPostExecutePromiseCount()`, we can decide that in
cases where there is nothing listening at all, there is no reason to
pull the data on the main instance.
Previously, there would always be postExecutePromises because the
telemetry events were called. Now, these have been moved into the
workers, which have been given the various InternalHooks calls to their
hook function arrays, so they themselves issue these telemetry and event
calls.
This results in all event log messages to now be logged on the worker’s
event log, as well as the worker’s eventbus being the one to send out
the events to destinations. The main event log does…pretty much nothing.
We are not logging executions on the main event log any more, because
this would require all events to be replicated 1:1 from the workers to
the main instance(s) (this IS possible and implemented, see the worker’s
`replicateToRedisEventLogFunction` - but it is not enabled to reduce the
amount of traffic over redis).
Partial events in the main log could confuse the recovery process and
would result in, ironically, the recovery corrupting the execution data
by considering them crashed.
# Refactor
I have also used the opportunity to reduce duplicate code and move some
of the hook functionality into
`packages/cli/src/executionLifecycleHooks/shared/sharedHookFunctions.ts`
in preparation for a future full refactor of the hooks
This PR adds new endpoints to the REST API:
`/orchestration/worker/status` and `/orchestration/worker/id`
Currently these just trigger the return of status / ids from the workers
via the redis back channel, this still needs to be handled and passed
through to the frontend.
It also adds the eventbus to each worker, and triggers a reload of those
eventbus instances when the configuration changes on the main instances.
