Commit Graph

338 Commits

Author SHA1 Message Date
Chao Xu
a309d3e652 Update pkg/api/testapi because namespace is always in the path since we remove v1beta1/2 2015-06-16 14:12:03 -07:00
Abhi Shah
2c93add90b Merge pull request #9613 from brendandburns/scheduler
Fix the scheduler to ignore terminated pods.
2015-06-12 10:05:09 -07:00
Brendan Burns
97634c7fbf Fix the scheduler to ignore terminated pods. 2015-06-10 14:38:08 -07:00
James DeFelice
7d66559725 added comments at the top of native k8s files which, if changed, might affect some of the k8sm code 2015-06-10 20:31:22 +00:00
Prashanth Balasubramanian
4bb3efaaad Scheduler doesn't send all fit predicate failures as events 2015-06-05 18:52:55 -07:00
Prashanth B
04209c4371 Merge pull request #9274 from lavalamp/logSpam
Demote log spam
2015-06-04 15:51:16 -07:00
Daniel Smith
32caedd882 Demote log spam
These messages are only useful if you want to debug a particular
scheduler assigment, and they are extremely verbose-- they each print
out a line per host per assignment. Let's try to keep our log messages
linear in the number of assigments.
2015-06-04 14:08:11 -07:00
derekwaynecarr
14d2128238 Fix namespace controller on delete to not care if item not found 2015-06-03 22:59:07 -04:00
feihujiang
ebc8eab590 get Events to be logged 2015-06-03 15:05:07 +08:00
Filip Grzadkowski
98115facfd Revert "Gracefully delete pods from the Kubelet" 2015-06-02 23:40:05 +02:00
Brian Grant
808932c079 Merge pull request #9056 from runningwild/master
Change isVolumeConflict so read-only gce persistent disks don't confl…
2015-06-02 13:14:13 -07:00
Clayton Coleman
72ee028cab Gracefully delete pods from the Kubelet
This commit wires together the graceful delete option for pods
on the Kubelet.  When a pod is deleted on the API server, a
grace period is calculated that is based on the
Pod.Spec.TerminationGracePeriodInSeconds, the user's provided grace
period, or a default.  The grace period can only shrink once set.
The value provided by the user (or the default) is set onto metadata
as DeletionGracePeriod.

When the Kubelet sees a pod with DeletionTimestamp set, it uses the
value of ObjectMeta.GracePeriodSeconds as the grace period
sent to Docker.  When updating status, if the pod has DeletionTimestamp
set and all containers are terminated, the Kubelet will update the
status one last time and then invoke Delete(pod, grace: 0) to
clean up the pod immediately.
2015-06-01 19:23:59 -04:00
Jonathan Wills
882462ad87 Change isVolumeConflict so read-only gce persistent disks don't conflict with each other 2015-06-01 08:13:48 -04:00
Rohit Jnagal
ef3d03e6f1 Merge pull request #8875 from derekwaynecarr/limit_range_not_use_requests
Do not set container requests in limit ranger for Kube 1.0
2015-05-29 13:26:00 -07:00
Rohit Jnagal
ff51f0b2e1 Merge pull request #8696 from derekwaynecarr/force_namespace_creation
Force explicit namespace provision, update e2e for failures
2015-05-29 09:28:47 -07:00
Tim Hockin
d0daabb34b Merge pull request #8471 from bprashanth/scheduler_burst
Limit the scheduler to a burst qps of 30
2015-05-28 16:55:25 -07:00
Chao Xu
9c1153322e update PodSpec.Host to PodSpec.NodeName in /pkg/api/types.go and /pkg/api/v1beta3/types.go 2015-05-28 14:10:06 -07:00
derekwaynecarr
3e8b1d5e01 Update all salt providers to force explicit namespace creation; update e2e 2015-05-28 13:45:49 -04:00
Prashanth Balasubramanian
4004a68ae9 Limit the scheduler to a burst qps of 30 and rate limit pod binding 2015-05-27 16:35:09 -07:00
derekwaynecarr
05f099211f Do not set container.requests for v1 2015-05-27 14:27:01 -04:00
deads2k
590bd048a5 add pull secrets to service accounts 2015-05-22 14:05:19 -04:00
Derek Carr
9b1fb6dca1 Merge pull request #8233 from csrwng/connect_adm_ctrl
Admission control to prevent exec on privileged pods
2015-05-21 14:35:28 -04:00
Cesar Wong
e95d9c416d Admission control to prevent exec on privileged pods 2015-05-21 13:56:51 -04:00
Cesar Wong
68ad63b5e2 Add operation checking to admission control handlers
Adds a new method to the handler interface that returns true only if the
admission control handler handles that operation.
2015-05-21 13:51:43 -04:00
Dawn Chen
40480d8504 Merge pull request #8532 from derekwaynecarr/improve_event_readability
Improve event readability for generic scheduler
2015-05-21 09:25:32 -07:00
Wojciech Tyczynski
5fbe58b2c8 Revert "Make scheduler optimistic about its bindings" 2015-05-19 14:27:28 -07:00
derekwaynecarr
8ee06a9983 Improve event readability 2015-05-19 17:26:24 -04:00
David Oppenheimer
44b1e1ca0c Fix stylistic isues with #5547. Closes #4910. 2015-05-18 23:32:06 -07:00
Victor Marmol
45874d5f76 Merge pull request #5547 from gmarek/client3
Add a resource specifying number of Pods that are allowed to run on Kubelet.
2015-05-18 11:11:43 -07:00
Vish Kannan
969023f3ec Merge pull request #8389 from smarterclayton/chatty_endpoints_controller
Endpoints controller is logging too much
2015-05-18 09:40:57 -07:00
Derek Carr
eb12565960 Merge pull request #8203 from pweil-/expose-context-to-admission
Expose user info to admission controllers
2015-05-18 08:25:10 -04:00
Clayton Coleman
d00f705652 Tone down scheduler logging
Having no nodes in the cluster is unusual and is likely a test
environment, and when a pod is deleted there is no need to log
information about our inability to schedule it.
2015-05-16 19:46:50 -04:00
Prashanth B
abd0d7b2cb Merge pull request #8261 from lavalamp/scheduler
Make scheduler optimistic about its bindings
2015-05-15 17:06:42 -07:00
gmarek
27d660d0ac Add a resource specifying number of Pods that are allowed to run on Kubelet. 2015-05-15 10:57:46 +02:00
Daniel Smith
3a7086fd8a Make scheduler optimistic about its bindings 2015-05-14 10:55:49 -07:00
Paul Weil
aaeb1dad93 expose user info to admission controllers 2015-05-13 21:31:51 -04:00
hurf
6203ce9f85 Move pkg/scheduler to plugin/pkg/scheduler
As the TODO in plugin/pkg/scheduler/scheduler.go described:

move everything from pkg/scheduler into this package. Remove
references from registry.
2015-05-13 14:03:25 +08:00
Jordan Liggitt
7e14a80f63 ServiceAccount admission plugin 2015-05-11 17:18:06 -04:00
Prashanth Balasubramanian
4b0607cf0b Scheduler ignored nodes with unknown condition status 2015-05-06 18:32:52 -07:00
Zach Loafman
875e83a741 Revert "Revert "Security context - types, kubelet, admission"" 2015-05-05 16:02:13 -07:00
Zach Loafman
f48904fd5e Revert "Security context - types, kubelet, admission" 2015-05-05 15:20:39 -07:00
Paul Weil
982bf19c20 security context initial implementation - squash 2015-05-05 13:46:13 -04:00
Daniel Smith
4a6ab7fd36 Merge pull request #7597 from eparis/copyright
Use generic copyright holder name in license boilerplate
2015-05-01 17:01:39 -07:00
Eric Paris
6b3a6e6b98 Make copyright ownership statement generic
Instead of saying "Google Inc." (which is not always correct) say "The
Kubernetes Authors", which is generic.
2015-05-01 17:49:56 -04:00
derekwaynecarr
6d08cc30ed Retry incrementing quota if there is a conflict 2015-05-01 15:23:21 -04:00
Robert Bailey
6d85dcb4a0 Add support for HTTP basic auth to the kube-apiserver. 2015-04-28 10:33:51 -07:00
Tim Hockin
a3d45fada8 Change flags to use dashes in help 2015-04-27 15:11:03 -07:00
nikhiljindal
dcc368c781 Removing more references to v1beta1 from pkg/ 2015-04-24 00:45:17 -07:00
derekwaynecarr
ac97e9fb67 normalize to lower resource names 2015-04-23 17:39:45 -04:00
Vish Kannan
ee06097c07 Merge pull request #7003 from derekwaynecarr/enforce_unbounded
Reject unbounded cpu and memory pods if quota is restricting it
2015-04-22 11:19:52 -07:00
Brian Grant
740d2a5f8a Merge pull request #6967 from HaiyangDING/ListAlgos
Listing existing algorithm providers in scheduler
2015-04-22 07:04:05 -07:00
dingh
71b98840c8 Listing available algorithm providers in scheduler
List the available algorithm providers with 'kube-scheduler --help' under field `algorithm_provider`
2015-04-22 06:54:24 +08:00
Abhishek Gupta
c203117e42 Including balanced resource allocation priority in the default set 2015-04-21 14:25:42 -07:00
derekwaynecarr
06eb45fb75 ResourceQuota add object count support for secret and volume claims 2015-04-21 09:49:40 -04:00
Jeff Lowdermilk
aba5d9aca5 Merge pull request #6969 from erictune/use_kubeconfig
Use kubeconfig in several components
2015-04-17 15:07:04 -07:00
derekwaynecarr
75482cabb4 Reject unbounded cpu and memory pods if quota is restricting it 2015-04-17 17:02:47 -04:00
Kris Rousey
81497f3ed2 Changing the scheduler package to use *api.Pod instead of api.Pod to
avoid unnecessary shallow copies. The change rippled through a lot of
code.
2015-04-17 13:34:31 -07:00
Eric Tune
6081fa52a7 controller, scheduler, proxy use kubeconfig
--master option still supported.
--kubeconfig option added to kube-proxy,
kube-scheduler, and kube-controller-manager
binaries.

Kube-proxy now always makes some kind of API
source, since that is its only kind of config.
Warn if it is using a default client, which probably won't work.

Uses the clientcmd builder.
2015-04-17 12:44:17 -07:00
derekwaynecarr
81dcd8c836 Improve error message when name is omitted but generateName is available 2015-04-16 13:06:19 -04:00
Daniel Smith
835a87aed2 Merge pull request #6857 from wojtek-t/fix_scheduler_nodes
Fix listing nodes in scheduler
2015-04-15 17:05:04 -07:00
Wojciech Tyczynski
20f39cd788 Merge pull request #6836 from lavalamp/fix4
add missing syncronization
2015-04-15 15:25:43 +02:00
Wojciech Tyczynski
afb94c43b1 Fix listing nodes in scheduler 2015-04-15 14:26:25 +02:00
Daniel Smith
fb0df935f6 add missing syncronization 2015-04-14 15:59:50 -07:00
Dawn Chen
9b1b5c5447 Merge pull request #6824 from derekwaynecarr/fix_quota_bug
Fix quota bug in LimitRanger
2015-04-14 14:56:55 -07:00
derekwaynecarr
a50e53257d Fix nil pointer that can happen if no container resources are supplied 2015-04-14 16:12:24 -04:00
Daniel Smith
dd26cdf0f7 Fix comments; switch to watching nodes 2015-04-10 17:49:51 -07:00
Daniel Smith
6835318d1e switch to require 'go controller.Run()' 2015-04-10 14:05:08 -07:00
Daniel Smith
5f7715f0e9 Make scheduler not miss deletion events even in the case of a resync. 2015-04-10 14:05:08 -07:00
Daniel Smith
e2093798cb Merge pull request #6675 from wojtek-t/nodes_support_field_selector
FieldSelector for node List()
2015-04-10 10:30:55 -07:00
Timothy St. Clair
2b60111fca Performance change to option enable client.QPS, client.Burst
and change default on max_requests_inflight.
2015-04-10 07:53:54 -05:00
Wojciech Tyczynski
88eb0b0295 FieldSelector for node List() 2015-04-10 12:53:16 +02:00
Dawn Chen
fd4fbef519 Install /healthz to kube-scheduler 2015-04-09 15:52:27 -07:00
Dawn Chen
d3d2948976 Revert "Revert "Fix scheduler crashes when profiling is enabled""
This reverts commit a945068eb4.
2015-04-09 15:37:59 -07:00
Fabio Yeon
a945068eb4 Revert "Fix scheduler crashes when profiling is enabled" 2015-04-09 15:06:32 -07:00
Wojciech Tyczynski
51a2d2db45 Fix scheduler profiling 2015-04-09 18:49:40 +02:00
Prashanth Balasubramanian
a7864aa230 Scheduler uses TTLStore for assumed pods 2015-04-08 15:01:04 -07:00
Masahiro Sano
d04cc5ced4 List in NodesInterface takes label selector 2015-04-08 23:41:27 +09:00
Wojciech Tyczynski
8a5c76356c Revert "Stop the bleeding for #6059"
This reverts commit 6be09c5e97.
2015-04-08 13:29:29 +02:00
Daniel Smith
c1390a0836 version field selector field names in the client 2015-04-07 16:59:57 -07:00
Clayton Coleman
51db3bd654 Create a new testclient package that can be backed by disk files
Standardize how our fakes are used so that a test case can use a
simpler mechanism for providing large, complex data sets, as well
as represent queries over time.
2015-04-07 14:56:15 -04:00
gmarek
ef56dca6b6 Remove ConditionSchedulable 2015-04-07 17:01:07 +02:00
Prashanth B
620af16892 Merge pull request #6097 from derekwaynecarr/limit_range_default_value
Limit range supporting a default resource limits value
2015-04-02 16:04:09 -07:00
Daniel Smith
6be09c5e97 Stop the bleeding for #6059 2015-04-02 13:54:10 -07:00
Daniel Smith
4a3b1a2b9d Merge pull request #6375 from abhgupta/abhgupta-dev
Validating against negative weights for priority functions
2015-04-02 13:22:55 -07:00
Brian Grant
414be101b6 Merge pull request #6352 from piosz/api
Removed PodStatus.Host
2015-04-02 13:16:24 -07:00
Abhishek Gupta
8deedf8e1f Validating against negative weights for priority functions 2015-04-02 12:30:22 -07:00
Filip Grzadkowski
25f95b0672 Add basic latency metrics to scheduler. 2015-04-02 19:24:55 +02:00
Piotr Szczesniak
5c6439d449 Removed PodStatus.Host
Fixes #6165
2015-04-02 16:38:45 +02:00
derekwaynecarr
c2b670513c Introduce concept of a default limit 2015-04-01 16:31:25 -04:00
Deyuan Deng
6bb9701489 Remove Reachable Condition 2015-04-01 13:56:43 -04:00
Abhishek Gupta
8a01aed83f Adding validations for scheduler Policy 2015-03-31 10:54:45 -07:00
derekwaynecarr
9bd7f05ffc Remove ResourceDefaults plugin 2015-03-31 10:05:28 -04:00
saadali
e0f71cb21f Make each new instance of kubelet generate a new event channel (instead of reusing existing). 2015-03-30 14:22:16 -07:00
Daniel Smith
10b4fe6f30 Merge pull request #6188 from abhgupta/abhgupta-dev
Minor fixes for scheduler priority configuration
2015-03-30 14:05:28 -07:00
Victor Marmol
05452f12ef Merge pull request #6110 from lavalamp/fix2
Remove pods from the assumed pod list when they are deleted
2015-03-30 13:49:52 -07:00
Abhishek Gupta
490dd1bb67 Fixing accidental hardcoding of priority function weight 2015-03-30 13:47:27 -07:00
Abhishek Gupta
08a0563933 Removing EqualPriority from the list of default priorities 2015-03-30 13:47:27 -07:00
Daniel Smith
fd952862c3 Remove pods from the assumed pod list when they are deleted 2015-03-30 11:16:55 -07:00
Abhishek Gupta
142fd773b0 Updating priority function weight based on specified configuration 2015-03-27 17:56:40 -07:00
Piotr Szczesniak
8d94c43e70 Merge pull request #5725 from rrati/schedule-to-available-nodes-5545
Only schedule to pods that are available.  This turns on the node
2015-03-27 10:20:35 +01:00