github/kubernetes
4
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
5,957 Commits 1 Branch 31 Tags 1,019 MiB
d7e89f9867
Commit Graph

569 Commits

Author SHA1 Message Date
Dawn Chen
c55921676c Merge pull request #4271 from ArtfulCoder/logging_etcd 
Updated etcd-pod-dir-fetch logging.
 2015-02-10 13:41:06 -08:00
Abhishek Shah
febf8c9b2c Fixed formattting issues in etcd.go 2015-02-10 12:03:40 -08:00
Mike Danese
c72c2a0d1e added tests 2015-02-10 08:30:13 -08:00
Mike Danese
3d0cd81feb refactor pkg/probe ProbeRunners to interfaces and move global probers into kubelet field for testability 2015-02-10 08:30:11 -08:00
Mike Danese
043794492e implement readiness. 2015-02-10 08:30:09 -08:00
Abhishek Shah
f136886dde Updated logging. If etcd-pod-dir-fetch fails because dir was absent, an info message is logged. 2) For all other reasons, the regular error message is logged. 2015-02-09 18:03:58 -08:00
Brendan Burns
cc7279f6a8 Actually use the safeFormatAndMounter... 2015-02-06 17:01:05 -08:00
Zach Loafman
b77abb5945 Merge pull request #4221 from brendandburns/pd_fix 
Make gce_safe_format_and_mount actually work correctly.
 2015-02-06 14:50:20 -08:00
Brendan Burns
3d6c9fc9e8 Fix gce_pd to only use safe_format_and_mount when mounting block devices. 2015-02-06 12:41:21 -08:00
Brendan Burns
35fa143b29 Revert "Revert "Start using GCE safe format and mount for mounting disks."" 
This reverts commit 858a85e8fe.
 2015-02-06 10:09:00 -08:00
Brendan Burns
94d852e8a6 Merge pull request #4214 from brendandburns/pd_fix 
Revert "Start using GCE safe format and mount for mounting disks."
 2015-02-06 09:58:51 -08:00
Brendan Burns
858a85e8fe Revert "Start using GCE safe format and mount for mounting disks." 
This reverts commit 3f109d6d46.
 2015-02-06 09:58:03 -08:00
Dawn Chen
05de54249f Merge pull request #4191 from simon3z/master 
api: add image's id to ContainerStatus
 2015-02-06 09:35:44 -08:00
Federico Simoncelli
e01df69565 api: add image's id to ContainerStatus 
Sometimes for external applications it is important to identify
exactly what images are running. Since tags can be moved to point
to newer builds this information can be used to identify old images
running.

Signed-off-by: Federico Simoncelli <fsimonce@redhat.com>
 2015-02-06 09:01:14 -05:00
Brian Grant
175e9caa55 Merge pull request #4194 from dchen1107/podstatus 
Kubelet decides podStatus
 2015-02-05 17:18:39 -08:00
Brendan Burns
3f109d6d46 Start using GCE safe format and mount for mounting disks. 
This removes the need for disks to be formatted.
 2015-02-05 17:01:37 -08:00
Mike Danese
14bfec92f2 rename probe.Status to probe.Result. 2015-02-05 15:49:58 -08:00
Dawn Chen
f762e062f2 Kubelet decides podStatus 2015-02-05 15:49:41 -08:00
Brendan Burns
8e6f5c7201 Merge pull request #3593 from commonlisp/master 
Add timeouts to HealthChecks and retry checks
 2015-02-05 15:40:22 -08:00
Brendan Burns
8b25b43039 Merge pull request #4117 from wojtek-t/validate_docker_version 
Check Docker version in Kubelet /healthz handler
 2015-02-05 10:06:33 -08:00
George Kuan
e8c33b7916 Add timeouts to HealthChecks and retry checks 
Fixes issue #3532. Added timeouts for HTTP and TCP checks
and enabled kubelet/probe to kubelet#maxRetries times
before declaring Failure.

Added Probe.TimeoutSecs to API

Probe variants now check container.LivenessProbe.TimeoutSeconds
Also added a test for timeouts in http_test.go.
 2015-02-05 06:04:45 -08:00
Wojciech Tyczynski
1a3419f970 Fix the min allowed Docker version. 2015-02-05 02:34:04 +01:00
Brendan Burns
cfe5b14114 Wait until containers actually finish running before trying to clean up 
volumes or pods.
 2015-02-04 17:28:41 -08:00
Wojciech Tyczynski
524cdba101 Fix hack/test-cmd.sh test. 2015-02-05 01:58:26 +01:00
Brendan Burns
6e897524be Merge pull request #4093 from nelcy/master 
Fix orphaned pod & volume cleanup order. Volumes need to be unmounted before pod cleanup.
 2015-02-04 14:38:20 -08:00
Wojciech Tyczynski
ead67108ce Use existing method to get Docker version. 2015-02-04 20:50:21 +01:00
Wojciech Tyczynski
78f66a6ce9 Check Docker version in Kubelet /healthz handler 2015-02-04 18:31:24 +01:00
chakri
9508b2bf12 Fix orphaned pod & volume cleanup order. Volumes need to be unmounted before pod cleanup. 2015-02-04 01:46:28 +00:00
Yu-Ju Hong
4a72addaeb Factor out API defaulting from validation logic 
Currently, the validation logic validates fields in an object and supply default
values wherever applies. This change factors out defaulting to a set of
defaulting callback functions for decoding (see #1502 for more discussion).

 * This change is based on pull request 2587.

 * Most defaulting has been migrated to defaults.go where the defaulting
   functions are added.

 * validation_test.go and converter_test.go have been adapted to not testing the
   default values.

 * Fixed all tests with that create invalid objects with the absence of
   defaulting logic.
 2015-02-03 00:55:42 -08:00
Brian Grant
2903fce2b8 Merge pull request #4031 from jbeda/hyperkube-kubelet 
Convert kubelet and proxy to hyperkube
 2015-02-02 18:08:59 -08:00
Tim Hockin
1ee1702be0 Merge pull request #3808 from brendandburns/master 
Add a mounter that uses google's safe_format_and_mount.
 2015-02-02 16:29:10 -08:00
Brendan Burns
ac21ac24fa Add a mounter that uses google's safe_format_and_mount. 2015-02-02 16:16:20 -08:00
Tim Hockin
ac399fb354 Merge pull request #3967 from brendandburns/fixer 
Be a little more careful when removing mount points.
 2015-02-02 15:46:18 -08:00
Joe Beda
76df5471b5 Covert the kubelet binary to hyperkube. 
This leaves `pkg/kubelet/server/server.go` looking a little ugly as there is an extra layer of "config" structs that isn't needed.  This is left as a TODO for now.
 2015-02-02 14:03:13 -08:00
Joe Beda
d96afdd645 Move 'pkg/standalone' package to 'pkg/kubelet/server'. 
This paves the way to hyperkube for the kubelet.
 2015-02-02 14:03:13 -08:00
Brendan Burns
7f06fbb081 Be a little more careful when removing mount points. 2015-02-02 12:35:24 -08:00
Vishnu Kannan
5e36f63f8b Adding ResourceRequirementSpec to v1beta1, v1beta2, and v1beta3 APIs. The old resource 
quantities 'CPU' and 'Memory' will be preserved until support for v1beta1 and v1beta2 APIs are
dropped.
Improved resource validation in the process.
 2015-02-01 02:19:55 +00:00
Clayton Coleman
556e59b512 Merge pull request #3810 from ironcladlou/cache-namespacing 
Support namespacing in cache.Store
 2015-01-30 13:19:41 -05:00
Brian Grant
207d0f5d40 Merge pull request #3878 from thockin/infra_name 
Use a constant for the pod infra container name
 2015-01-30 09:18:00 -08:00
Ye Yin
248624fc1f Add docker's add/drop capabilites 2015-01-30 15:09:31 +08:00
Brendan Burns
a0b8429d2e Reverse the order of unmounting and removing the pd disk. 
This should mean that we retry detaching if the original detach fails for some reason.
 2015-01-29 16:13:12 -08:00
Dan Mace
5ee943d683 Support namespacing in cache.Store implementations 
Support namespacing in cache.Store by framing the interface functions
around interface{} and providing a key function to each Store implementation.

Implementation of a fix for #2294.
 2015-01-29 17:39:49 -05:00
Mrunal Patel
19e0605f5b Replace "net" by "pod infra" in docs and format strings. 2015-01-28 15:03:06 -08:00
Tim Hockin
804643a04b Merge pull request #3818 from mikedanese/probe-refactor 
break api.Probe out of api.LivenessProbe
 2015-01-28 12:41:35 -08:00
Dawn Chen
b0bcee7b40 Increase timeout for all testcases in file_test.go 2015-01-28 11:27:15 -08:00
Mike Danese
78f33e950a rename api.LivenessProbe to api.Probe and break out Actions 2015-01-28 11:20:44 -08:00
Tim Hockin
c8f61885df Merge pull request #3695 from mikedanese/ready 
refactor pkg/health into more reusable pkg/probe
 2015-01-28 11:00:32 -08:00
Tim Hockin
70ab406751 Use a constant for the pod infra container name 
This is an implementation detail that has leaked out of kubelet.  Use a
constant to prevent divergence, for now.  Plan to get rid of it.

@dchen1107
 2015-01-28 10:13:02 -08:00
Paul Morie
962bdda0dc Fix typo in kubelet 2015-01-28 12:00:53 -05:00
Tim Hockin
371637a5fc Merge pull request #3845 from brendandburns/disk 
Detach the PD disk from the VM in some failure cases.
 2015-01-27 15:46:36 -08:00