8fd45d8300
clean up func podTimestamp
2019-03-27 11:05:19 +08:00
49346c1e04
shorten scheduler package alias for better readability
...
- schedulerinternalcache -> internalcache
2019-03-21 18:14:19 -07:00
2b554b4286
scheduler: cleanup unused parameters in unschedulableQ
2019-03-13 11:54:16 -07:00
7bcdbc5a51
Merge pull request #74446 from yanghaichao12/comm0223
...
Fix typos
2019-03-05 23:24:00 -08:00
b475cdf7fc
Fix typos
...
change SupportsPreemption
change comment again
2019-03-05 02:12:55 -05:00
f6d05d6315
Merge pull request #74611 from denkensk/fix-unable-find-backoff-value
...
fix the flake in scheduling_queue_test
2019-02-28 16:06:57 -08:00
471679f8d2
fix the flake in scheduling_queue_test
2019-02-28 07:23:27 +08:00
df4d65d2e1
Ensure that pods obey backoff timers.
...
The function AddUnschedulableIfNotPresent is responsible for
initializing or updating backoff timers for pods that could not be
scheduled. The helper function backoffPod does that work, but was not
being called in all cases.
This moves that call to be (mostly) unconditional, while cleaning up
comments and error handling.
2019-02-26 10:21:18 -08:00
bf3b5e5563
Merge pull request #73700 from denkensk/no-updae-timestamp-each-scheduling-attempt
...
Don't update the Pod object after each scheduling attempt
2019-02-22 01:47:48 -08:00
ea9e1a4118
not updae timestamp for each scheduling attempt
2019-02-22 09:46:18 +08:00
337cb7036c
Add tests for the new cache snapshotting mechanism.
2019-02-20 13:38:37 -08:00
e0ad271891
Optimize scheduler cache snapshotting using an MRU structure.
2019-02-20 13:38:37 -08:00
ec0739bd03
Fix races in scheduling queue tests
2019-02-11 16:38:26 -08:00
b975573385
move pkg/kubelet/apis/well_known_labels.go to staging/src/k8s.io/api/core/v1/
...
Co-Authored-By: Weibin Lin <linweibin1@huawei.com>
Change-Id: I163b2f2833e6b8767f72e2c815dcacd0f4e504ea
2019-02-05 13:39:07 -05:00
f63f772867
Presize map with proper capacity in cache
2019-02-04 17:00:05 -08:00
ba47beffd2
Fix weakness of current receivedMoveRequest
...
- add incremental scheduling cycle
- instead of set a flag on move reqeust, we cache current scheduling
cycle in moveRequestCycle
- when unschedulable pods are added back, compare its cycle with
moveRequestCycle to decide whether it should be added into active queue
or not
2019-01-30 10:13:49 +08:00
02cdc24cd3
enforce unscheduable pod has proper chance to be retried
2019-01-22 09:23:51 +08:00
28e6bbc175
acquire lock before operating unschedulablepodsmap
2019-01-18 04:24:18 +08:00
d857790d36
Merge pull request #72558 from denkensk/add-goroutine-move-unschedulablepods-to-activeq
...
Move unschedulable pods to the active queue if they are not retried for more than 1 minute
2019-01-16 17:15:16 -08:00
de8cfdcd79
add goroutine to move unschedulablepods to activeq regularly
2019-01-16 12:08:19 +08:00
db60e4f20c
Merge pull request #72632 from danielqsj/sch_pod
...
Move getNextPod to closure
2019-01-08 19:22:19 -08:00
5a708017e9
Merge pull request #72619 from everpeace/fix-starvation-by-sorting-queue-with-last-probe-time
...
change sort function of scheduling queue to avoid starvation when a lot of unscheduleable pods are in the queue
2019-01-07 20:24:24 -08:00
8dc1e83946
Add func comment for MakeNextPodFunc
2019-01-08 11:07:47 +08:00
554ad383e3
Move getNextPod to closure
2019-01-07 17:54:49 +08:00
22079a79d4
change sort function of scheduling queue to avoid starvation when unschedulable pods are in the queue
...
When starvation heppens:
- a lot of unschedulable pods exists in the head of queue
- because condition.LastTransitionTime is updated only when condition.Status changed
- (this means that once a pod is marked unschedulable, the field never updated until the pod successfuly scheduled.)
What was changed:
- condition.LastProbeTime is updated everytime when pod is determined
unschedulable.
- changed sort function so to use LastProbeTime to avoid starvation
described above
Consideration:
- This changes increases k8s API server load because it updates Pod.status whenever scheduler decides it as
unschedulable.
Signed-off-by: Shingo Omura <everpeace@gmail.com>
2019-01-07 11:05:45 +09:00
95254d5457
Move CacheDebugger signal handling into the package.
...
This moves the signal handling for CacheDebugger from the factory
package into the CacheDebugger's package. That makes it easier to reuse
from packages other than factory.
2019-01-03 18:15:51 -08:00
222d7837a8
Merge pull request #72303 from cofyc/fix72013-cleanup-scheduler
...
Update comments to reflect current logic
2019-01-03 15:51:34 -08:00
fa88bcc154
Update comments for PriorityQueue.
...
Log condition reason which is useful for debugging.
2019-01-03 09:25:05 +08:00
7044145920
Fix race in setting nominated node
2018-12-29 22:54:15 -08:00
aee1ab34ab
Merge pull request #72260 from bsalamat/fix_waiting_pods
...
Add pods in the backoff queue to the list of pending pods
2018-12-27 00:03:18 -08:00
000aa06912
Make scheduleinternalcache.PodFilter as public
2018-12-21 23:09:56 +08:00
48b6f75829
Add pods in the backoff queue to the list of pending pods
2018-12-20 17:28:23 -08:00
d27d28a44e
Flatten nominated pod logic in PriorityQueue.
...
This replaces deeply nested ifs & fors with early returns & continues.
2018-12-14 18:11:12 -08:00
fae4f69d36
Fix return value of PriorityQueue.Add.
...
This function was returning a non-nil error for the common, non-failure
case. The fix is to properly scope local error values and add early
returns.
2018-12-14 16:44:21 -08:00
05183bffe5
Merge pull request #71872 from yuexiao-wang/scheduler-nodeinfo
...
[scheduler cleanup phase 2]: Rename `pkg/scheduler/cache` to `pkg/scheduler/nodeinfo`
2018-12-12 08:08:33 -08:00
7b6f60f085
modify BUILD
...
Signed-off-by: yuexiao-wang <wang.yuexiao@zte.com.cn>
2018-12-11 11:22:06 +08:00
f3353c358d
[scheduler cleanup phase 2]: Rename to
...
Signed-off-by: yuexiao-wang <wang.yuexiao@zte.com.cn>
2018-12-11 11:21:12 +08:00
2fe9b1438c
activate unschedulable pods only if the node became more schedulable
2018-12-10 09:48:23 +08:00
5e4ccede4c
Reschedule with backoff
...
With the alpha scheduling queue we move pods from unschedulable to
active on certain events without a backoff. As a result we can cause
starvation issues if high priority pods are in the unschedulable queue.
Implement a backoff mechanism for pods being moved to active.
Closes #56721
2018-12-06 16:38:34 +00:00
c821f2ed2f
Move scheduling Heap in to scheduler.core.utils
...
The Heap data structure is useful for our backoff system in addition to
scheduling queue. Move it to somewhere it can be consumed by both
systems and properly export needed names. Also adding unit tests
from client-go/tools/cache/heap.go.
2018-12-04 18:01:42 +00:00
82abbdc11a
Merge pull request #71488 from bsalamat/queue-sort
...
Change sort function of the scheduling queue to avoid starvation
2018-12-01 03:59:51 -08:00
79e5cb2cb7
Merge pull request #71302 from liggitt/verify-unit-test-feature-gates
...
Split mutable and read-only access to feature gates, limit tests to readonly access
2018-11-29 21:45:12 -08:00
527d1c34cc
Merge pull request #70947 from Adirio/nodetree-thread-safety
...
Scheduler internal NodeTree thread-safe NumNodes
2018-11-29 07:36:48 -08:00
36f8859fa0
autogenerated files
2018-11-27 17:08:58 -08:00
554acf2b38
Change sort function of the scheduling queue to avoid starvation
2018-11-27 17:08:40 -08:00
2498ca7606
drop VerifyFeatureGatesUnchanged
2018-11-21 11:51:33 -05:00
1f3057b7fb
Merge pull request #70898 from Huang-Wei/preemption-issue
...
ensure scheduler preemptor behaves in an efficient/correct path
2018-11-16 20:27:35 -08:00
b4fd11512a
ensure scheduler preemptor behaves in an efficient/correct path
...
- don't update nominatedMap cache when Pop() an element from activeQ
- instead, delete the nominated info from cache when it's "assumed"
- unit test behavior adjusted
- expose SchedulingQueue in factory.Config
2018-11-16 14:22:15 -08:00
733dd9dfd7
Add tests to ensure feature gate changes don't escape kubelet/scheduler packages
2018-11-16 10:52:53 -05:00
de8bf9b63d
fix scheduler and kubelet unit tests leaking feature flag changes
2018-11-16 10:52:53 -05:00
wangqingcan