Jesse Suen
16ff90c070
Defer deletion of app object until all resources have been deleted (issue #636 ) ( #807 )
...
Purge app cache after deletion is successful (issue #802 )
2018-11-19 12:25:45 -08:00
Jesse Suen
925f9486e3
Restructure application sources to separate types ( #799 )
2018-11-17 16:20:25 -08:00
Alexander Matyushentsev
275b9e194d
Issue #355 - Treat 'crd-install' hooks as normal k8s resource ( #792 )
2018-11-16 17:12:21 -08:00
Alexander Matyushentsev
44087fa1b5
Issue #621 - Remove resources state from application CRD ( #758 )
2018-11-16 17:10:04 -08:00
Alexander Matyushentsev
317d2a8aa8
Issue #536 - Declarative setup and configuration of ArgoCD ( #745 )
...
* Issue #536 - Declarative setup and configuration of ArgoCD
* Add missing rules to application-controller role
* Fix broken test; update install manifests
2018-11-09 09:58:07 -08:00
Taylor D. Edmiston
34bb60f064
Make Argo CD naming consistent ( #694 )
...
* Make Argo CD naming consistent
* Change ArgoCD to Argo CD on new lines
2018-11-05 11:29:01 -08:00
dthomson25
2c8c2fcd64
Support adding name prefix in helm and kustomize ( #735 )
2018-10-29 16:05:22 -07:00
Jesse Suen
2498f60c57
Update dependencies to k8s v1.12 and client-go v9.0 ( #729 )
...
Update dependencies to k8s v1.12 and client-go v9.0 (resolves #353 )
Fix issue where applications could not be deleted on k8s v1.12 (resolves #718 )
Refactor k8s dynamic resource libraries to promote code reuse
2018-10-28 22:46:13 -07:00
Alexander Matyushentsev
c76db90437
Issue #628 - Remove RollbackOperation in favor of Sync with ParameterOverrides ( #706 )
2018-10-18 20:23:22 -07:00
Alexander Matyushentsev
9c79af9340
Issue #682 - Operation stuck in 'in progress' state if application has no resources ( #684 )
2018-10-11 11:18:17 -04:00
Andrew Merenbach
1ba52c8880
Allow more fine-grained sync ( closes #508 ) ( #666 )
2018-10-10 10:12:20 -07:00
Alexander Matyushentsev
93a808e65a
Issue #627 - Cluster watch needs to be restarted when CRDs get created ( #678 )
2018-10-05 13:18:12 -04:00
Jesse Suen
e163177a12
Switch to go-git for all remote git interactions including auth (issue #651 )
2018-09-27 12:48:23 -07:00
dthomson25
1ba10a1a20
Remove default params from app history ( #649 )
2018-09-27 11:24:25 -07:00
Jesse Suen
0af77a0706
Add more event sources and provide better detail in event messages (issue #635 ) ( #637 )
...
* Expand SyncOperation to also store parameter overrides
Fix auto-sync when used with parameter overrides
* Add more event sources and provide better detail in event messages (issue #635 )
2018-09-24 08:52:43 -07:00
Alexander Matyushentsev
70b9db68b4
Issue #599 - Lazy enforcement of unknown cluster/namespace restricted resources ( #612 )
2018-09-20 09:48:54 -07:00
Jesse Suen
bfb558eb92
Fix issue where helm hooks were being deployed as part of sync (issue #605 )
2018-09-17 11:29:44 -07:00
Jesse Suen
505866a4c6
Support helm charts with dependencies and namespace sensitivity (issue #582 )
2018-09-17 11:29:44 -07:00
Jesse Suen
02116d4bfc
Fix comparison failure when app contains unregistered custom resource (issue #583 ) ( #596 )
2018-09-13 14:02:04 -07:00
Alexander Matyushentsev
aa2afcd47b
Issue #330 - Projects need controls on cluster-scoped resources ( #558 )
...
* Issue #330 - Projects need controls on cluster-scoped resources
* Issue #330 - Introduce namespace resources black-list
2018-09-11 15:10:47 -07:00
Jesse Suen
fd510e7933
Support an automated sync policy upon detection of OutOfSync status from git ( #571 )
2018-09-11 14:28:53 -07:00
Jesse Suen
bc98b65190
Fix controller hot loop when app source contains bad manifests (issue #568 ) ( #570 )
2018-09-10 10:58:13 -07:00
dthomson25
954706570c
Reorder K8s resources to correct creation order ( #551 )
2018-09-10 10:14:14 -07:00
Alexander Matyushentsev
e2faf6970f
Issue #527 - Support --in-cluster authentication without providing a kubeconfig ( #559 )
...
* Issue #527 - Support --in-cluster authentication without providing a kubeconfig
* Issue #527 - make sure resources are watched for 'local' cluster
2018-09-10 08:20:17 -07:00
Jesse Suen
cf630055b0
API discovery becomes best effort when partial resource list is returned ( resolves #524 ) ( #526 )
2018-08-21 13:53:42 -07:00
Jesse Suen
00299707e5
Expand RBAC role to be able to create application events. Fix username claims extraction. ( #479 )
2018-07-31 11:15:44 -07:00
Jesse Suen
a7c7523a8c
Support helm charts and yaml directories as an application source ( #432 )
...
* Support helm charts and yaml directories as an application source
* Run e2e test in parallel and increase timeout
2018-07-24 16:37:12 -07:00
Alexander Matyushentsev
b3af671803
Issue #340 - create application/project events for audit ( #440 )
...
* Issue #340 - create application/project events for audit
* Issue 340 - move username to message field
* Reviewer notes: fix possible panic
2018-07-24 18:48:13 +03:00
JazminGonzalez-Rivero
2343818ab5
Resolves 398 -> errors set status to known, status details set to error message ( #437 )
...
* if healthcheck fails, return unknown health status
* pass error through
* revert error pass
* fix error pass
* save first error
2018-07-23 11:32:35 -07:00
Jesse Suen
c0367ed595
Add support for hook deletion policies (OnSuccess, OnFailure) ( resolves #374 ) ( #412 )
2018-07-16 10:15:53 -07:00
Jesse Suen
39b9f4d31a
Add ability to terminate a running operation ( resolves #379 ) ( #409 )
2018-07-13 17:13:31 -07:00
Jesse Suen
078b2339bd
Apply logic was ignoring kubectl apply failures ( #395 )
2018-07-13 10:03:56 -07:00
Jesse Suen
f65859bc25
Label hooks so that the cluster resource watch will be notified about completions ( #387 )
2018-07-12 14:01:54 -07:00
Alexander Matyushentsev
543ae3cf13
Issue #364 - Assess health of StatefulSets, DaemonSets, ReplicaSets ( #391 )
2018-07-13 00:01:28 +03:00
Alexander Matyushentsev
610a4510cf
Issue #385 - Sync and health status should be unknown if controller unable to load target/live state ( #386 )
2018-07-12 22:40:21 +03:00
Jesse Suen
7b92977889
Improve logging. Prevent some unecessary patches to app ( #383 )
2018-07-12 12:39:46 -07:00
Jesse Suen
9d921f65f3
Various fixes to sync logic ( #370 )
...
* move hook resource state into SyncResult (from operation state)
* fix rollback to use apply based sync
* re-assess sync/health status between each sync phases
* PostSync hook should wait until application is Healthy (resolves #363 )
2018-07-11 19:12:30 -07:00
Andrew Merenbach
9cddd4c368
Force app refresh after sync ( #373 )
2018-07-11 13:45:41 -07:00
Alexander Matyushentsev
8db465c699
Issue #277 - Warning message if controller detect resources which belongs to multiple application ( #372 )
...
* Issue #277 - Warning message if controller detect resources which belongs to multiple application
* Reviewer notes: add missing nil check
2018-07-11 23:00:48 +03:00
Jesse Suen
eb1caf2231
Use hook strategy as the default when performing a sync ( #368 )
2018-07-10 16:02:55 -07:00
Alexander Matyushentsev
adb84f3d03
Issue #210 - Application controller should not erase previously collected info about resources if repo or cluster is not available ( #367 )
2018-07-11 00:45:18 +03:00
Alexander Matyushentsev
7d4dd0fdd5
Issue #304 - Add spec validation condition to application CRD ( #360 )
2018-07-09 20:45:03 +03:00
Jesse Suen
364415f83a
Move k8s health assessment into a stand-alone library. Use kubectl convert to statically assess health ( #356 )
2018-07-09 10:13:56 -07:00
Jesse Suen
d633f6b299
Support for PreSync, Sync, PostSync resource hooks ( #350 )
...
* Rewrite controller sync logic to support workflow-based sync
* Redesign hook implementation to support generic resources as hooks
2018-07-07 00:54:06 -07:00
Andrew Merenbach
bf2fe9d33f
Add missing health status ( #354 )
...
* Add missing health status
* Update swagger.json
* Default app and service health to Missing
* Fill in Missing status if status field is blank
* Simplify missing injection
2018-07-06 16:33:25 -07:00
Alexander Matyushentsev
16fa41d25b
Issue #275 - Application controller fails to get app state if app has resource without name ( #285 )
2018-06-14 09:08:22 -07:00
Andrew Merenbach
df0e2e4015
Fail app sync if prune flag is required ( #276 )
...
* Add status field to resource details
* Update generated code
* Set up const message responses
* Check number of resources requiring pruning
* Fix imports
* Use string, thanks @alexmt
* Update generated code
2018-06-12 10:54:11 -07:00
Alexander Matyushentsev
aa4291183b
Take into account number of unavailable replicas to decided if deployment is healthy or not ( #270 )
...
* Take into account number of unavailable replicas to decided if deployment is healthy or not
* Run one controller for all e2e tests to reduce tests duration
* Apply reviewer notes: use logic from kubectl/rollout_status.go to check deployment health
2018-06-07 11:05:46 -07:00
Alexander Matyushentsev
f0be1bd251
Fix bug secret controller which is causing update loop in secret controller ( #251 )
2018-05-31 16:06:41 -07:00
Alexander Matyushentsev
e785abeb8f
Issue #244 - Cluster/Repository connection status ( #248 )
2018-05-31 13:44:19 -07:00
