5ff9bd43df
Various fixes for the operator
...
1. Fixed SparkApplication controller to ignore reconcilation caused by SparkApplication not found errors.
2. Fixed the PySpark example.
3. Fixed code for updating the status of a SparkApplication in updateAndExportMetrics.
4. Logging refactoring.
2018-12-05 14:50:26 -08:00
78fdb0f315
fixed error propagation
2018-11-30 14:42:38 +01:00
b7a219587a
Refactoring around logging code
2018-11-19 16:44:40 -08:00
8c7fdbb306
Operator State Management + Ingress Creation ( #291 )
...
* SparkOperator: Prometheus Metrics Integration
* Prometheus Metric Update
* Spark Operator:Prometheus Metric Integration
* PositiveGauge rework
* remove unwanted dependencies
* Propogating ScheduledSpark App Labels
* Doc update
* Metric Description update
* fix app wait
* SparkOperator: Prometheus Metrics Integration
* Spark Operator metrics:PR Comments
* SparkOperator: Set completion time for Failed App
* Operator Metrics: PR comments
* Spark Operator: PR Comments
* Controller Update
* PR Comments
* Docs Update
* Driver State Transition Check Update
* Operator State Management
* Clean-up
* Exposing Spark Application Id in Operator
* SparkAppId updates
* Add Lyft as a user and contributor to operator
* Spark Operator Rework
* Reworking Restart-Policy
* Documentation update
* PR comments
* PR comments
* Ingress impl
* Ingress Tests + Updates
* go fmt
* PR Comments
* missing files
* AppId removal: Doc Update
* Doc update
* Delete UI/Ingress + Other minor changes
* Add PENDING_RETRY State
* PR comments
* PR comments
* Clean-up
* Update controller.go
* Add Terminal State
* Terminal State
* Spark improvements
* event type
* Events update
* Update controller.go
* Update controller.go
* PR Comments
* PR comments
* Support Best-effort Spec updates
* New State
* PR comments
* PR comments
* go fmt
* Docs update
* PR feedback
* PR Feedback
* PR comments
2018-11-19 15:25:55 -08:00
90e246faab
Fixed go import path
2018-09-13 12:32:58 -07:00
5bb1ca8b2e
Add support for automatic configuration for exposing metrics to Prometheus
2018-08-20 12:55:21 -07:00
046efef400
Removed the superficious error log
2018-08-09 14:20:17 -07:00
55b7b95b7a
Extend sparkctl to stream SparkApplication events ( #243 )
...
* Trigger events on driver/executor pending, running and unknown
* Extend sparkctl to stream SparkApplication events
2018-08-09 10:09:49 -07:00
d44d6c1e9a
Some minor refactoring
2018-08-03 15:22:31 -07:00
ceb230481e
SparkOperator: Prometheus Metrics Integration ( #227 )
...
* SparkOperator: Prometheus Metrics Integration
* Spark Operator metrics:PR Comments
* SparkOperator: Set completion time for Failed App
* Operator Metrics: PR comments
* Spark Operator: PR Comments
* Controller Update
* PR Comments
* Docs Update
* Driver State Transition Check Update
2018-08-03 13:56:58 -07:00
00c6a7fd58
Do not check restart eligibility if the application state has not changed
2018-07-10 23:30:03 -07:00
fca2c5552f
Fixed RBAC for the spark operator
2018-06-02 16:06:34 -07:00
b7351eac9b
make apiv1.NamespaceAll the default namespace
2018-06-01 07:23:09 -07:00
a89239e94f
add -namespace flag to manage only CRDs in that namespace
2018-06-01 07:23:09 -07:00
ca786de4b5
Explicitly delete the driver pod and UI service when a SparkApp gets deleted
2018-05-24 14:18:15 -07:00
44219cf645
Skip restart of the pod or UI service of the old run was not found
2018-05-21 12:57:30 -07:00
35b51dfc87
Fixed a bug that may lead to duplicated restarts when RestartPolicy=Always
2018-05-19 15:28:35 -07:00
96ecf5e12d
Fixed typos in string format
2018-04-23 12:09:23 -07:00
eff4c59cb2
Add cron support
2018-04-19 12:01:37 -07:00
Yinan Li
Grzegorz Lyczba
akhurana001
Piotr Mrówczyński
Scott Reisor