dc4003cb70
There are various reasons that the HPA will decide not the change the current scale. Two important ones are when missing metrics might change the direction of scaling, and when the recommended scale is within tolerance of the current scale.
...
The way that ReplicaCalculator signals it's desire to not change the
current scale is by returning the current scale. However the current
scale is from scale.Status.Replicas and can be larger than
scale.Spec.Replicas (e.g. during Deployment rollout with configured
surge). This causes a positive feedback loop because
scale.Status.Replicas is written back into scale.Spec.Replicas,
further increasing the current scale.
This PR fixes the feedback loop by plumbing the replica count from
spec through horizontal.go and replica_calculator.go so the calculator
can punt with the right value.
2019-07-03 15:42:14 +02:00
602fc4771a
Merge pull request #78725 from waynepeking348/waynepeking348-master
...
fix bug of replica_calculator to initialize replicaCount with currentReplicas
2019-06-06 10:48:27 -07:00
b8b1720f12
Fix bug of ObjectPerPodMetricReplicas to initialize replicaCount with currentReplicas
2019-06-05 11:54:03 +00:00
88f8c785b4
Merge pull request #78711 from MrHohn/lb-finalizer-patch-err
...
Don't mask k8s error returned from patch
2019-06-04 22:31:52 -07:00
89a7a35607
Don't mask k8s error returned from patch
2019-06-04 16:40:51 -07:00
6a7e7fc156
Fix TestStressingCascadingDeletion flake
2019-06-04 17:26:36 -04:00
9f85c5c4b5
Merge pull request #78503 from gjtempleton/Multiple-Metrics-HPA
...
Multiple metrics hpa
2019-06-03 19:34:59 -07:00
ac458e9d89
Merge pull request #78581 from andrewsykim/route-rate-limit
...
route controller: respect rate limit for delete
2019-06-01 12:13:07 -07:00
bc32307314
Merge pull request #78262 from MrHohn/svc-finalizer-cleanup2
...
Add Service Load Balancer finalizer support
2019-05-31 22:02:59 -07:00
b3c1bbb471
Merge pull request #78061 from yuxiangqian/metrics
...
Enhance to record more accurate e2e provision/deletion latency metric
2019-05-31 20:45:13 -07:00
6feea43bcc
Merge pull request #77475 from liucimin/fix_cronjob_controller_pageerr
...
Fix cronjob controller page list err
2019-05-31 12:22:40 -07:00
380c6cbd0d
route controller: respect rate limit for delete
...
Signed-off-by: Andrew Sy Kim <kiman@vmware.com>
Co-authored-by: Khaled Henidak <khnidk@microsoft.com>
2019-05-31 08:26:36 -04:00
52f3380ef3
change preempting to PreemptionPolicy
2019-05-31 12:42:05 +08:00
5c9438c691
non-preempting-priorityclass
...
Co-authored-by: Vallery Lancey <vallery@zeitgeistlabs.io>
Co-authored-by: Tan shanshan <tan.shanshan@zte.com.cn>
2019-05-31 12:37:07 +08:00
64198a43e7
Auto-generated BUILD files
2019-05-30 21:09:04 -07:00
2ab3681648
Fix-up and add unit test for Load Balancer finalizer
2019-05-30 21:09:04 -07:00
aa3f81d657
Add Load Balancer finalizer support
...
- Always try to remove finalizer upon load balancer cleanup
- Add finalizer prior to load balancer creation (feature gated)
- Cache logic fix-ups
- Event type/message fix-ups
- Use runtime.HandleError() on eaten errors
Co-authored-by: Josh Horwitz <horwitzja@gmail.com>
2019-05-30 21:09:04 -07:00
38a884aa04
using storage class provisioner instead of annotation of a pv for external provisioner
...
minor updates
2019-05-30 10:19:36 -07:00
05df640f33
Merge pull request #77994 from gnufied/csi-resize-migration
...
Handle CSI volume resize migration.
2019-05-29 21:28:43 -07:00
1efbde2815
Handle invalid metrics when scaling on multiple metrics
...
Handle a case in the Horizontal Pod Autoscaler Controller when scaling
on multiple metrics and one or more is missing or invalid.
If all metrics are missing - return an error and leave the isScalingActive
condition as that for the last invalid metric.
If some metrics are missing/invalid and some are valid and found -
if a scale up would be triggered by the valid metrics ignore the missing
metrics and scale up, if a scale down would be triggered, return an error
and leave the isScalingActive condition as that for the last invalid metric.
2019-05-29 23:20:40 +01:00
ee4dbbcbff
Add tests for handling scaling on unavailable metrics
...
Add three tests for handling invalid metrics when scaling on
multiple metrics - one for scaling up successfully (new behaviour)
and two for ensuring we don't scale down (existing behaviour).
2019-05-29 23:11:32 +01:00
9688511595
record provision and deletion latency metric
...
instead of using provisioner from storage class directly, uses plugin name firstly
2019-05-29 14:15:30 -07:00
7563b4d01b
Address comments about retry in online resizing block
2019-05-29 11:01:45 -04:00
d8fd232ea1
Merge pull request #78327 from caesarxuchao/pointer-remainingItemCount
...
Make RemainingItemCount a pointer
2019-05-29 07:59:29 -07:00
306b6f330d
Merge pull request #78248 from cwdsuzhou/optimize_err_log
...
Add error info for plugin do not support attachment
2019-05-29 00:53:17 -07:00
501092c7d0
Merge pull request #78232 from SataQiu/fix-golint-20190523
...
Fix golint failures of pkg/controller/namespace/deletion pkg/controller/serviceaccount
2019-05-29 00:53:02 -07:00
d119899e22
handle review comments
2019-05-28 15:22:41 -04:00
369314959c
generated
2019-05-24 18:14:53 -07:00
f1883c9e8c
Support scale subresource for PDBs ( #76294 )
...
* Support scale subresource for PDBs
* Check group in finder functions
* Small fixes and more tests
2019-05-23 22:24:17 -07:00
cdff17a96b
Merge pull request #78166 from qingsenLi/k8s-190522
...
duplicated klog.V(5) when had a if klog.V(5)
2019-05-23 20:58:43 -07:00
5c23e393cf
duplicated klog.V(5) when had a if klog.V(5)
2019-05-24 00:49:38 +08:00
e39ec09975
Add error info for plugin do not support attachment
2019-05-23 18:32:49 +08:00
b87d006a50
fix golint failures of pkg/controller/namespace/deletion pkg/controller/serviceaccount
2019-05-23 12:09:01 +08:00
0cea0675ec
Merge pull request #78159 from byteor/master
...
[pkg/controller/serviceaccount]: refactor to more idiomatic go
2019-05-22 15:14:37 -07:00
4f98c6fc7f
Merge pull request #78104 from Pingan2017/update-translate-time
...
update LastTransitionTime for pod ready condition when node notready
2019-05-22 08:34:14 -07:00
2fd4f883d6
[pkg/controller/serviceaccount]: refactor to more idiomatic go
...
Signed-off-by: Zhou Peng <p@ctriple.cn>
2019-05-21 17:55:30 +08:00
0a8d7aadc0
update LastTransitionTime for pod ready condition when node become notready
2019-05-21 17:20:08 +08:00
8211cabfb2
Merge pull request #77647 from fisherxu/patch-1
...
Add pending status for pastBackoffLimitOnFailure
2019-05-20 21:56:39 -07:00
9feb0df370
Add pending status for pastBackoffLimitOnFailure
2019-05-21 09:45:29 +08:00
33b95ebddb
add some tests for expand controller
...
Update bazel file
2019-05-20 10:55:23 -04:00
6ba13bf151
Merge pull request #77773 from DaiHao/daemon
...
fix daemon set rolling update hang
2019-05-20 03:11:12 -07:00
05ebe91277
Switched to use dynamic shared informer for Garbage Collector.
2019-05-17 15:41:36 -07:00
9cbbf59265
Merge pull request #77984 from MrHohn/svc-finalizer-alpha
...
`patch` service status instead of `update` in service controller
2019-05-17 10:12:06 -07:00
405d33eae4
Add code to handle in-tree to CSI migration for resizing
2019-05-17 10:57:44 -04:00
27410955e2
Merge pull request #77952 from liggitt/delete-on-update
...
Handle updates removing remaining finalizers on deleted objects
2019-05-17 02:04:32 -07:00
da4d6f5829
Merge pull request #77883 from gnufied/mat-expand-workqueue
...
Use workqueues for volume expansion
2019-05-17 02:04:18 -07:00
4886e8782c
name the queue properly and use lower case for error message
2019-05-16 21:34:57 -04:00
cda73ebbb0
Fixes service controller unit test for patching
...
Co-authored-by: Josh Horwitz <horwitzja@gmail.com>
2019-05-16 17:30:43 -07:00
a1f07049e5
Patch service instead of update in service controller
...
Co-authored-by: Josh Horwitz <horwitzja@gmail.com>
2019-05-16 17:30:43 -07:00
673773cbde
check for pvc objects
2019-05-16 12:23:52 -04:00
Joseph Burnett