jprdonnelly/kestra
1
0
Fork 0
mirror of https://github.com/kestra-io/kestra.git synced 2025-12-19 18:05:41 -05:00
Code Issues Packages Projects Releases Wiki Activity
8,707 Commits 164 Branches 427 Tags
plugin/template_task_example
Commit Graph

563 Commits

Author SHA1 Message Date
Nicolas K.
7bca8b4924 fix(repositories): unwanted integer to string conversion (#13586) 
* fix(repositories): unwanted integer to string conversion

* fix(repositories): clean code

---------

Co-authored-by: nKwiatkowski <nkwiatkowski@kestra.io>
 2025-12-10 18:59:52 +01:00
Loïc Mathieu
56febfb415 fix(executions): don't remove worker task result for killed execution 
As killing an executuion is asynchronous, it is inherently racy.
So when we kill an execution it move to the rerminal state which then will remove any worker task result to purge eagerly the queue table.

But if such worker task result arrives late and was not already processed by the executor, it will be purged before ever been able to be processed so the task would nevert be updated as KILLED.

Note: this may un-flaky some falky unit tests

Fixes https://github.com/kestra-io/kestra-ee/issues/6040
 2025-12-10 17:27:11 +01:00
YannC
018c22918f Revert "feat(jdbc): method without auditlog registered for setting repository (#13543)" (#13581) 
This reverts commit 3e9c8cf7da.
 2025-12-10 16:00:55 +01:00
YannC
3e9c8cf7da feat(jdbc): method without auditlog registered for setting repository (#13543) 2025-12-10 15:52:42 +01:00
Nicolas K.
008404e442 feat(filters): add new operation to filters (#13424) 
* feat(filters): add new operation to filters

* feat(filters): add flowId to flow search

* fix(test): fix unit test

---------

Co-authored-by: nKwiatkowski <nkwiatkowski@kestra.io>
 2025-12-10 15:17:32 +01:00
char zheng
bea3d63d89 fix(executions): concurrency limit exceeded for KILLED execution 
Fixes #13211
 2025-12-09 09:57:49 +01:00
mustafatarek
354873e220 chore(core): remove unnecessary attempt list copying 2025-12-08 09:41:10 +01:00
brian-mulier-p
a8635108b7 fix(core): safeguard for null flow when trying to reset trigger in JdbcExecutor (#13381) 2025-12-04 12:46:53 +01:00
Loïc Mathieu
56fb304ff6 fix(execution): NORMAL kind should also be retrieved 
Fixes #13262
 2025-12-03 11:43:50 +01:00
brian-mulier-p
d8e3a9bd44 fix(executions): avoid infinite loop in some cases of execution failure (#13293) 2025-12-02 13:09:34 +01:00
Loïc Mathieu
ea44128d2b chore(system): refactor RunnerUtils to be a static utils 
Part-of: https://github.com/kestra-io/kestra-ee/issues/4228
 2025-12-02 10:37:46 +01:00
brian.mulier
682d258e7b feat(ns-files): add a metadata layer on top for better performance & versioned ns files 
part of https://github.com/kestra-io/kestra/issues/5617
 2025-12-01 18:13:49 +01:00
brian.mulier
d20f7039c7 fix(tests): avoid flakiness with random worker group in JdbcServiceLivenessCoordinatorTest 2025-12-01 16:56:01 +01:00
brian.mulier
2191331750 fix(tests): JdbcRunnerRetryTest.retryFailedFlowDuration flaky 2025-12-01 16:56:01 +01:00
Roman Acevedo
40880bf7d8 feat(system): force nullsFirst for ASC and opposite globally in JDBC 2025-12-01 16:06:35 +01:00
Roman Acevedo
7aca309be5 feat(dashboard): calculate running execution duration on the fly 2025-12-01 16:06:35 +01:00
brian-mulier-p
f06b1c5347 fix(core): concurrency limit on JDBC was decrementing when using FAIL or CANCEL behavior (#13220) 
closes https://github.com/kestra-io/kestra/issues/13141
 2025-12-01 12:44:30 +01:00
Florian Hussonnois
5a8552ad36 refactor(core): remove LogService 
Move static methods to Logs utility class
Move purge method to existing ExecutionLogService
 2025-11-28 17:48:23 +01:00
brian.mulier
b68dcb7bf5 fix(tests): larger timeout for JdbcServiceLivenessCoordinatorTest.shouldReEmitTasksToTheSameWorkerGroup 2025-11-27 19:23:17 +01:00
brian.mulier
d6933b8e49 fix(executions): avoid JdbcExecutor from being stuck due to missing flow 
In tests it can occur for example
 2025-11-27 19:23:17 +01:00
brian.mulier
15da58dbf4 fix(tests): use another db name on webserver to avoid colliding with repositories 2025-11-27 19:23:17 +01:00
brian.mulier
c3d94dc8ff fix(tests): remove JdbcTestUtils.drop usages as it defeats concurrent test runs 2025-11-27 19:23:17 +01:00
YannC
546039e30a fix(webserver): export test with correct path + DateField for execution export (#13197) 2025-11-27 15:43:03 +01:00
Loïc Mathieu
8f4bafc666 feat(execution): add an attemps on skipped tasks 2025-11-26 18:11:07 +01:00
YannC
7fd16b24e0 feat: allows to export flows/triggers/executions as a streamed CSV files (#13137) 2025-11-26 14:11:14 +01:00
Loïc Mathieu
9d6694f807 fix(system): WorkerTask should not FAILED when interrupting so they would be resubmitted 
When a Worker is stopping, it will first wait for all running tasks to stop, then kill them. For those that didn't implement kill their thread would be interrupted.

But if the task is properly killed, or support interrupts (like the Sleep task), it would ends in FAILED then a WorkerTaskWould be send that would fail the flow preventing the WorkerTask to be resubmitted.

We nows check if the worker is terminating and should resubmit, in this case we didn't emit any WorkerTaskResult

Fixes #13108
Part-of: https://github.com/kestra-io/kestra-ee/issues/5556
 2025-11-21 16:58:36 +01:00
Loïc Mathieu
735697ac71 chore(system): share JDBC repository code in an abstract CRUD repository 2025-11-18 11:13:16 +01:00
Loïc Mathieu
a4ca3498f3 fix(execution): use jdbcRepository.findOne to be tolerant of multiple results 
It uses findAny() under the cover which does not throw if more than one result is returned.

Fixes #12943
 2025-11-18 10:21:51 +01:00
Loïc Mathieu
41f83949f0 chore(system): replace Flow by FlowInterface where possible 
Part-of: https://github.com/kestra-io/kestra/issues/8274
 2025-11-18 09:50:49 +01:00
Florian Hussonnois
0f340a9a29 fix(scheduler): mysql convert 'now' to UTC to avoid any offset error on next_execution_date 
Fixed a previous commit to only apply the change for MySQL

Related-to: kestra-io/kestra-ee#5611
 2025-11-17 21:52:58 +01:00
Florian Hussonnois
ae204a03b0 fix(scheduler): convert 'now' to UTC to avoid any offset error on next_execution_date 
Related-to: kestra-io/kestra-ee#5611
 2025-11-17 13:12:31 +01:00
Loïc Mathieu
9957b1659e fix(flow): flow trigger with both conditions and preconditions 
When a flow have both a condition and a precondition, the condition was evaluated twice which lead to double execution triggered.

Fixes
 2025-11-14 16:35:15 +01:00
Loïc Mathieu
03b1b1be8c fix(system): consume the trigger queue so it is properly cleaned 
Fixes https://github.com/kestra-io/kestra/issues/11671
 2025-11-13 11:10:27 +01:00
YannC
7d1f064fe9 fix: when removing a queued execution, directly delete instead of fetching then delete to reduce deadlock (#12789) 2025-11-10 10:31:41 +01:00
shashank.agrawal
33299dc3ec Issue - 10984 | Full dependency graphs for parent child namespaces. 2025-11-06 17:03:50 +01:00
Loïc Mathieu
9d3d40ade8 fix(system): trigger an execution once per condition on flow triggers 
Fixes #12560
 2025-11-05 15:30:26 +01:00
Roman Acevedo
d620dd7dec test: set retryWithFlowableErrors as FlakyTest 2025-11-04 13:46:14 +01:00
Loïc Mathieu
f8b0d4217f fix(executions): Flow triggered twice when there are two multiple conditions 
Fixes #12560
 2025-10-31 16:22:13 +01:00
Roman Acevedo
71e49f9eb5 feat(executions): add IN, NOT_IN, CONTAINS LABELS #11916 
- advance on https://github.com/kestra-io/kestra/issues/11587
- companion PR: https://github.com/kestra-io/kestra-ee/pull/5617
 2025-10-31 10:20:05 +01:00
Loïc Mathieu
9920d190c8 feat(system): add an administration page for concurrency limit 
Closes  #11250
 2025-10-30 17:49:33 +01:00
brian.mulier
07e90de835 fix(core): CrudEvent should not be done on the repository side for KV 2025-10-30 16:57:57 +01:00
Loïc Mathieu
c06ffb3063 feat(system): set taskrun attempt to resubmitted when a taskrun is resubmitted to a worker 
Closes https://github.com/kestra-io/kestra/issues/12481
 2025-10-30 15:46:05 +01:00
brian.mulier
d9144c8c4f feat(core): introduce KV Metadata in-repository storing (#12342) 
part of https://github.com/kestra-io/kestra/issues/12341
 2025-10-29 17:18:43 +01:00
brian.mulier
4ed44754ab fix(core): use index by adding deleted everytime in query 2025-10-29 17:18:43 +01:00
Loïc Mathieu
4502c52d2b fix(executions): remove errors and finally tasks when restarting 
Otherwize we would detect that an error or a finally branch is processing and the flowable state would not be correctly taken.

Moreover, it prevent this branch to be taken again after a restart.

Fixes #11731
 2025-10-28 14:29:27 +01:00
mustafatarek
85dc3ec788 fix(core): synchronize creation/running flowable attempt states in Jdbc executor 2025-10-22 14:26:13 +02:00
mustafatarek
c6e7ff9436 fix(core): transfer flowable attempt creation from saveFlowableOutput() to TaskRun to fix tests 2025-10-22 14:26:13 +02:00
mustafatarek
6e7d6de2e2 test: try to fix 2025-10-22 14:26:13 +02:00
mustafatarek
1771955717 feat(core): handle flowable task attempts state transitions 2025-10-22 14:26:13 +02:00
杨利伟
7e68274cf4 fix(jdbc): add comment 2025-10-22 10:36:15 +02:00