* Improve dynamic flush scheduler for otel data
The changes introduce a more flexible and adaptive dynamic flush scheduler to address production issues where the system wasn't flushing data fast enough, causing memory growth and crashes. This issue arises from the existing scheduler handling only a single flush at a time, limiting concurrency and failing to cope with the influx of logs.
- Added configuration options for setting minimum and maximum concurrency levels, maximum batch size, and memory pressure threshold. These parameters ensure that flush operations adjust dynamically based on workload and pressure.
- Implemented `pLimit` to facilitate concurrent flush operations, with adjustments made according to batch queue length and memory pressure.
- Metrics reporting improvements were added to monitor the dynamic behavior of the flush scheduler, aiding in identifying performance issues and optimizing the operation accordingly.
* Implement load shedding for TaskEvent records
This change introduces load shedding mechanisms to manage TaskEvent
records, particularly those of kind LOG, when the system experiences
high volumes and is unable to flush to the database in a timely
manner. The addition aims to prevent overwhelming the system and
ensure critical tasks are prioritized.
- Added configuration options for `loadSheddingThreshold` and
`loadSheddingEnabled` in multiple modules to activate load shedding.
- Introduced `isDroppableEvent` function to allow specific events to
be dropped when load shedding is enabled.
- Ensured metrics are updated to reflect dropped events and load
shedding status, providing visibility into system performance
during high load conditions.
- Updated loggers to inform about load shedding state changes,
ensuring timely awareness of load management activities.
* Fix undefined 'queuePressure' variable in DynamicFlushScheduler
The 'queuePressure' variable was being used without being defined
in the DynamicFlushScheduler class, causing potential runtime
errors. This commit adds the missing definition and ensures that
the variable is correctly calculated based on the 'totalQueuedItems'
and 'memoryPressureThreshold'.
- Addressed code inconsistencies and improved formatting.
- Defined 'queuePressure' in the 'adjustConcurrency' method
to prevent potential undefined errors.
- Enhanced readability by maintaining consistent spacing and
format across the file, contributing to the stability and
maintainability of the code.
- Adjusted batch size logic based on the newly defined 'queuePressure'
variable.
* Refactor concurrency adjustment logic in scheduler
The concurrency adjustment logic in the dynamic flush scheduler has been refactored to improve clarity and maintainability. This change moves the calculation of pressure metrics outside of the conditional blocks to ensure they are always determined prior to decision-making.
- The queue pressure and time since last flush calculations were moved up in the code to be independent of the 'backOff' condition.
- This refactor sets up the groundwork for more reliable concurrency scaling and better performance monitoring capabilities. The overall logic of adjusting concurrency based on system pressure metrics remains unchanged.
This adjustment addresses ongoing issues with the scheduler that were not resolved by previous changes.
* Some tweaks
* Sentry WIP
* Configure sentry for uploading and releasing during the publish webapp step
* Delete source maps after uploading
* Forward logger.error calls to sentry through Logger.onError
* Couple tweaks to the dockerfile
* Premerge run batch before sending it to clickhouse
* Pre-order batch items before sending to clickhouse in favor of performance
* existing
* Emit event on batch flushes
* When merging batches, keep the last occurrence items with the same version
* Add a couple of tests
* Initial burst changes to queue page
* Added tooltip and changed wording around
* View runs from Queues page
* Fix for ugly Version filter "Current" badge
The normal database needs to be mocked because an import somewhere is
using it which causes issues with the test container dbs and hanging at
the end.
Strategy copied from the trigger test file.
* Set the default replication concurrency to 2 for self-hosters
100 was a bit crazy
* Made the run repository an interface, deferring just to CH for now
* Added run repository feature flag, allowing passing a default when getting a flag
* Switch run repository using a feature flag
* Added spans
* Pass the default repository in, so we can try Postgres in the tests
* Fallback to Postgres if ClickHouse errors
* Update feature flags API endpoint
* Queue in run table and filtering
* Debounce the filter changes
* Remove console log
* Added machine filtering
* Added version filtering
* Filter by version in the db
* Removed duplicate classes
* Version filtering hasFilters consistency
* Added queues and machines to the bulk action summary
* runs.list filtering for queue and machine
* Fix for machine errors
* Input field now has accessory instead of shortcut
* First experiments with the UI
* Got the fake filtering working
* AI filtering is working pretty well ✨
* Started working on tool calling
* Tool calling is working
* Styling progress
* Working on the error
* Errors work, improved the styling
* Nice glow effect
* Tweak the darkness of the text field
* Re-ordered the UI, set AI settings to use system prompt and telemetry
* Refactored to make it testable
* Added basic evals
* Better time inputs and evals
* Removed some code comments
* Remove unused useSearchParam change
* Tidy imports
* If no OpenAI API key send json back
* Fix for merge conflict with duplicate query filters
* Another conflict resolved
* Another merge conflict resolved
* Pass the model in, allow changing it
* Queue in run table and filtering
* Debounce the filter changes
* Remove console log
* Added machine filtering
* Added version filtering
* Filter by version in the db
* Removed duplicate classes
* Version filtering hasFilters consistency
* Added queues and machines to the bulk action summary
* runs.list filtering for queue and machine
* Fix for machine errors
* Added experimental_devProcessCwdInBuildDir config option
Added experimental_devProcessCwdInBuildDir config option to opt-in to new process.cwd behavior when executing tasks in the dev CLI. Currently process.cwd maps to the "root" of your trigger.dev project (the directory that contains your trigger.config.ts file). Setting experimental_devProcessCwdInBuildDir to true changes process.cwd to instead be the temporary build directory inside of the .trigger directory.
This makes it so the files added via the additionalFiles extension can be read using the same path in dev and deployed tasks.
* Remove claude code reference because it breaks windows tests 😡
* WIP
* Make release concurrency system extremely simple, everything just releases all the time
* update the deadlock detection to use the new lockedQueueReleaseConcurrencyOnWaitpoint column
* WIP new release concurrency system
* Remove releaseConcurrency and releaseConcurrencyOnWaitpoint
Also removed deadlock detection, and added environment burst concurrency
* Added new DEQUEUED status
Cleaned up the API run statuses, including now detecting new clients and not breaking older clients by adding an API version header to all requests
* Introduce the new "current dequeued concurrency set"
* Remove QUEUED_EXECUTING because we no longer "eagerly" release before checkpointing
* Remove waitpoint test for QUEUED_EXECUTING
* Add isWaiting
* Add changeset
* Use createdAt for ordering realtime runs instead of number
* Clarify the envCurrentDequeuedKey usage
* mock the db.server file to fix the tests
* Updated changset "EXECUTED" -> "EXECUTING"
---------
Co-authored-by: Matt Aitken <matt@mattaitken.com>
* Access machinePreset from the run list presenter
* New icons for machine presets
* New icon + name combo label for the machine preset
* Adds new “Machine” column to the runs list
* Make a separate component for the machine tooltip info
* add machinePreset to the span presenter
* Show the Machine in the Details tab in the Run inspector
* Show an admin only separator
* Fix docs icon in the button
* Small padding tweak
* Move the Machine nearer the costs
* Don't cast the machine preset
* Remove typecast, better to have a bad label if we add a new machine and don't update thos
---------
Co-authored-by: Matt Aitken <matt@mattaitken.com>
* Use sexy scrollbars for templates and recent runs popover content
@Nick this is for you!!1
* Fix TabButton component disabled state
* Disable payload tab when replaying runs with large payloads
* Fix for filtering by run id
Simplified to have a single filter for runId which accepts an array. This unifies the filtering by multiple runs (used by the waitpoint page) and the individual run filtering. It fixes a bug with bulk actions
* Truncate long bulk action titles
* Fix regression that broke superjson behavior for the replay flow
* Ignore undefined type metadata when overriding superjson payloads
Addresses https://github.com/triggerdotdev/trigger.dev/issues/1968
* Add a few unit tests for replaceSuperJsonPayload
* Add ID of the replayedFrom run to the TaskRun schema
* Propagate the replayedFrom run ID in the replay flow
* Link the replayed run in the run details pane
* useSearchParams has
* useSearchParams has
* useSearchParams has
* Consistent way to get the run filters
* Consistent way to get the run filters
* Consistent way to get the run filters
* Initial work on the new bulk actions
* Initial work on the new bulk actions
* Initial work on the new bulk actions
* WIP actions and filtering
* WIP actions and filtering
* WIP actions and filtering
* Empty filter arrays are set to undefined
* Empty filter arrays are set to undefined
* Empty filter arrays are set to undefined
* WIP prisma schema
Removed extra runtimeEnvironmentId
* WIP prisma schema
Removed extra runtimeEnvironmentId
* WIP prisma schema
Removed extra runtimeEnvironmentId
* Migrations
* Migrations
* Migrations
* BulkActionGroup changed some columns around
* BulkActionGroup changed some columns around
* BulkActionGroup changed some columns around
* New badge variant, removed unused ones
* New badge variant, removed unused ones
* New badge variant, removed unused ones
* Bulk action button
* Bulk action button
* Bulk action button
* Make the next runs page the default now
* Make the next runs page the default now
* Make the next runs page the default now
* Improved the RadioButton style
* Improved the RadioButton style
* Improved the RadioButton style
* Remove the old bulk action bar
* Remove the old bulk action bar
* Remove the old bulk action bar
* More UI progress
* More UI progress
* More UI progress
* Lots of UI changes to the Runs page
* Lots of UI changes to the Runs page
* Lots of UI changes to the Runs page
* Fixed period filter resetting everything
* Fixed period filter resetting everything
* Fixed period filter resetting everything
* Improved the Switch secondary style
* Improved the Switch secondary style
* Improved the Switch secondary style
* Buggy filter fixes
* Buggy filter fixes
* Buggy filter fixes
* Improved the filter display and fixed a bug with search param from object
* Improved the filter display and fixed a bug with search param from object
* Improved the filter display and fixed a bug with search param from object
* Clear button is minimal
* Clear button is minimal
* Clear button is minimal
* Using a presenter now
* Using a presenter now
* Using a presenter now
* Bulk actions are created, but not actually processed (yet)
* Bulk actions are created, but not actually processed (yet)
* Bulk actions are created, but not actually processed (yet)
* Bulk replay/cancel is working
* Bulk replay/cancel is working
* Bulk replay/cancel is working
* Multiple fixes, added bulk column to PG
* Multiple fixes, added bulk column to PG
* Multiple fixes, added bulk column to PG
* Bulk action run filtering working using CH
* Bulk action run filtering working using CH
* Bulk action run filtering working using CH
* Replay setting the bulk id on the runs
* Replay setting the bulk id on the runs
* Replay setting the bulk id on the runs
* Properly cap the time when doing a bulk action
* Properly cap the time when doing a bulk action
* Properly cap the time when doing a bulk action
* If the bulk action isn't recent, add it to the dropdown anyway
* If the bulk action isn't recent, add it to the dropdown anyway
* If the bulk action isn't recent, add it to the dropdown anyway
* Blank version of the bulk actions page
* Blank version of the bulk actions page
* Blank version of the bulk actions page
* Individually selected runs working
* Individually selected runs working
* Individually selected runs working
* Use selected mode if runs are checked
* Use selected mode if runs are checked
* Use selected mode if runs are checked
* Added the modal
* Added the modal
* Added the modal
* Marked the old bulk actions stuff as deprecated
* Marked the old bulk actions stuff as deprecated
* Marked the old bulk actions stuff as deprecated
* Renamed bulk action file
* Renamed bulk action file
* Renamed bulk action file
* Bulk run filter with the name and a default
* Bulk run filter with the name and a default
* Bulk run filter with the name and a default
* WIP on bulk actions page
* WIP on bulk actions page
* WIP on bulk actions page
* Updated panel, added new truncated id component
* Updated panel, added new truncated id component
* Updated panel, added new truncated id component
* Style improvements to the radio buttons
* Style improvements to the radio buttons
* Style improvements to the radio buttons
* Added an option action completion email
* Added an option action completion email
* Added an option action completion email
* Adds a blank state for the bulk actions page
* Adds a blank state for the bulk actions page
* Adds a blank state for the bulk actions page
* Nicer completed email
* Nicer completed email
* Nicer completed email
* Don't open the bulk action panel if there are no runs
* Don't open the bulk action panel if there are no runs
* Don't open the bulk action panel if there are no runs
* Runs blank state and bulk action accordion
* Runs blank state and bulk action accordion
* Runs blank state and bulk action accordion
* Updates secondary/small switch style
* Updates secondary/small switch style
* Updates secondary/small switch style
* Pagination buttons no longer split in twain (WIP)
* Pagination buttons no longer split in twain (WIP)
* Pagination buttons no longer split in twain (WIP)
* Aborting working
* Aborting working
* Aborting working
* Bulk action live reloading
* Bulk action live reloading
* Bulk action live reloading
* ListPagination works correctly in all states
* ListPagination works correctly in all states
* ListPagination works correctly in all states
* Run page, show friendlyId instead of number
* Run page, show friendlyId instead of number
* Run page, show friendlyId instead of number
* Bulk action help open by default if you have none
* Bulk action help open by default if you have none
* Bulk action help open by default if you have none
* Extra status filtering step because of replication delay
* Extra status filtering step because of replication delay
* Extra status filtering step because of replication delay
* Wider bulk action onboarding
* Wider bulk action onboarding
* Wider bulk action onboarding
* More sensible widths on the bulk action side panel
* More sensible widths on the bulk action side panel
* More sensible widths on the bulk action side panel
* Border color tweak to the RadioButton
* Border color tweak to the RadioButton
* Border color tweak to the RadioButton
* Improved the accordion component hover states
* Improved the accordion component hover states
* Improved the accordion component hover states
* Updates the bulk action blank state images to the latest UI
* Updates the bulk action blank state images to the latest UI
* Updates the bulk action blank state images to the latest UI
* Added R and C shortcuts back in
* Added R and C shortcuts back in
* Added R and C shortcuts back in
* Fix for selecting a single run
* Fix for selecting a single run
* Fix for selecting a single run
* Improved exit icon, added shortcut to modal
* Improved exit icon, added shortcut to modal
* Improved exit icon, added shortcut to modal
* Tidy imports
* Tidy imports
* Tidy imports
* Tidy imports
* Tidy imports
* Tidy imports
* Tidy imports
* Tidy imports
* Tidy imports
* Tidy imports
* Fix for grid layout when 1 page of bulk actions visible
* Fix for grid layout when 1 page of bulk actions visible
* Fix for grid layout when 1 page of bulk actions visible
* Removed the ... on the abort button
* Removed the ... on the abort button
* Removed the ... on the abort button
* Removed the ... on the abort button
* Animate the progress bar
* Set TZ="UTC" in the env example
* Filter summary in the bulk inspector
* Improves the pagination styling
* Improves the pagination styling
* Delete old bulk action routes
* Removed old Postgres RunListPresenter
* Retry any replication error where the message contains "timeout"
* Increase wait to make test less flaky
* The test was using run id instead of friendly id
* Safer array access
* Remove error log if there's a bad status
* Nicer frontend type safety with the bulk action and mode
* Switched a log to a debug log
* Retry replication unless the error is a known non-retry error
Flip the strategy to retry by default
* Make ClickHouse required
* Backfill run replication admin API endpoint
* Set a CLICKHOUSE_URL for unit tests
---------
Co-authored-by: James Ritchie <james@trigger.dev>
nicktrn