Commit Graph

1449 Commits

Author SHA1 Message Date
Tim Hockin
bca90f4866 Reduce and comment exports 2014-08-11 00:11:59 -07:00
Tim Hockin
7beac7a9af Split health_check.go into smaller parts.
Distinct files for core, http, and tcp.
2014-08-11 00:01:17 -07:00
Tim Hockin
24c516ec1d Move HTTP code out of health.go
Put it with the related HTTP code.
2014-08-10 23:50:06 -07:00
Tim Hockin
c67c1edfb4 Use IntOrString for TCP health check ports
Clean up code to be more testable.  Add test cases for named and numeric
ports in TCP health checks.  Improve tests.
2014-08-10 23:44:42 -07:00
Tim Hockin
7201227cb1 Use IntOrString for HTTP health check ports
Clean up code to be more testable.  Add test cases for named and numeric
ports in HTTP health checks.  Improve tests.
2014-08-10 23:26:42 -07:00
Daniel Smith
7d605467dc New scheduler API
This commit adds a Binding object. The idea is that schedulers can write
these to cause pods to be asssigned to hosts. I'll provide an implementation
along with a rudimentary scheduler plugin.

This continues k8s' tradition of phrasing all APIs as RESTful handlers.
2014-08-10 19:05:03 -07:00
Daniel Smith
5cdce0e35a Prepare for external scheduler
1. Change names of Pod statuses (Waiting, Running, Terminated).
2. Store assigned host in etcd.
3. Change pod key to /registry/pods/<podid>. Container location remains
   the same (/registry/hosts/<machine>/kublet).
2014-08-10 15:05:36 -07:00
Daniel Smith
e35dfedd79 Merge pull request #840 from smarterclayton/missing_return_in_watch
Watch handler not returning after 404
2014-08-09 16:05:03 -07:00
Clayton Coleman
d419eedb06 Watch handler not returning after 404 2014-08-09 10:46:56 -04:00
Yuki Yugui Sonoda
d359f3c396 Make storage.Delete return *api.Status instead of api.Status
as apiserver.APIServer.finishReq expects.

This solves the warning in finishReq:
"programmer error: use *api.Status as a result, not api.Status."
2014-08-09 22:18:24 +09:00
Brad Wells
0ef6b9eb19 Add explicit instruction to vagrant example docs
Building kubecfg is required for kubecfg.sh. kubecfg.sh will warn
of this if it hasn't been built, but it's a better user experience
to make people explicitly aware of all required steps.
2014-08-08 23:09:58 -06:00
Brad Wells
e3fd1672ac (Re)generate the build version when preparing a release.
If hack/build-go.sh has never been run before release/build-release.sh
the subsequent build will fail. Also, when preparing builds the
version should be forced to be up to date with the current checkout.
2014-08-08 22:43:31 -06:00
brendandburns
c71866164f Merge pull request #837 from lavalamp/gofuzz
Gofuzz
2014-08-08 16:47:54 -07:00
Clayton Coleman
0083fae453 Provide helpers and tests for common error types
Unify error handling in apiserver into a single path - RESTStorage
objects must provide appropriate errors individually.  Ensure ALL
errors which can be traced to logical faults with RESTStorage are
returned as api.Status objects.
2014-08-08 19:44:47 -04:00
Daniel Smith
079c9043bd Switch to new external fuzz package 2014-08-08 15:54:09 -07:00
Daniel Smith
a0e9cf575f bump(github.com/google/gofuzz): 6897f01a51d9afe3f54e4911f7e1ce6dd399202a 2014-08-08 15:54:02 -07:00
Daniel Smith
723df867c4 Insert gofuzz as dep 2014-08-08 15:31:58 -07:00
Clayton Coleman
71c6e082d4 Merge pull request #805 from lavalamp/serverWatch
Improve watch
2014-08-08 17:30:28 -04:00
brendandburns
4c00acdde0 Merge pull request #827 from brendandburns/kubelet
Add container hashing to the container name, and restart containers on changes.
2014-08-08 14:17:11 -07:00
Daniel Smith
5dd130a350 Prevent accidental setting of sync or timeout 2014-08-08 14:09:14 -07:00
Daniel Smith
49cded3800 Simplify ResourceWatcher interface to one function. 2014-08-08 14:09:13 -07:00
Daniel Smith
71709ae09e Make replication controller use client 2014-08-08 14:09:13 -07:00
Daniel Smith
51caf759c3 Add WatchReplicationControllers to kubecfg's fake 2014-08-08 14:09:13 -07:00
Daniel Smith
85ff1d3e7f Add fake client to make testing easier. 2014-08-08 14:09:13 -07:00
Daniel Smith
097147545c Allow more general parameters to be made by client. Also fix style and comments. 2014-08-08 14:09:13 -07:00
Daniel Smith
283fdba6ab Add more parameters to Watch
* Add labels selector (same as List)
* Add fields selector
 * Plan to let you select pods by Host and/or Status
* Add resourceVersion to let you resume a watch where you left off.
2014-08-08 14:09:13 -07:00
Daniel Smith
d52492111f Merge pull request #835 from smarterclayton/apiserver_cleanup
Centralize path magic in apiserver into the New method
2014-08-08 13:53:15 -07:00
Clayton Coleman
a74fac7d70 Centralize path magic in apiserver into the New method
Make OperationHandler and WatchHandler properly encapsulate their
concerns and make them not depend on the global path
2014-08-08 16:33:44 -04:00
Brendan Burns
2986db9885 Add container hashing to the container name, and restart containers on changes. 2014-08-08 13:32:07 -07:00
Daniel Smith
2297bf8cea Merge pull request #830 from brendandburns/net
Delete containers for a pod if we have to create the network container.
2014-08-08 13:12:42 -07:00
brendandburns
a7a571e1ce Update README.md 2014-08-08 12:49:06 -07:00
brendandburns
959b40bd54 Point the docs at a specific API html revision. 2014-08-08 12:48:41 -07:00
brendandburns
d0e6d00c30 Add a link to the HTML-ified API Documentation. 2014-08-08 12:46:36 -07:00
Brendan Burns
38900a9c58 Delete containers for a pod if we have to create the network container. 2014-08-08 12:34:59 -07:00
Daniel Smith
ac6d6ec974 Merge pull request #807 from smarterclayton/abstract_apiserver_encoding
Decouple apiserver from codec implementation
2014-08-08 11:57:04 -07:00
Tim Hockin
bff752bb24 Merge pull request #832 from VojtechVitek/api_typo
Fix typo in API
2014-08-08 08:30:10 -07:00
Vojtech Vitek (V-Teq)
ce4fcc4ad8 Fix typo in API
Signed-off-by: Vojtech Vitek (V-Teq) <vvitek@redhat.com>
2014-08-08 13:16:10 +02:00
Victor Marmol
adc57da3f3 Merge pull request #825 from brendandburns/runin
Add support for "run in"
2014-08-07 23:47:19 -07:00
Brendan Burns
d31d0781b2 Add support for "run in" 2014-08-07 20:27:58 -07:00
Tim Hockin
b7a8fabb32 Merge pull request #824 from derekwaynecarr/fix_kubelet_in_vagrant_env
Enforce -address flag is set on all Linux flavors
2014-08-07 17:39:02 -07:00
derekwaynecarr
c8c5435656 Enforce --address flag is set on all Linux flavors 2014-08-07 17:54:49 -05:00
brendandburns
76489b7fa0 Merge pull request #813 from thockin/cleanups
Remove duplicate -config param
2014-08-07 15:06:06 -07:00
Tim Hockin
7e9c142392 Remove duplicate -config param
This is duplicated in cluster/saltbase/salt/kubelet/default
cluster/saltbase/salt/kubelet/initd
2014-08-07 10:53:08 -07:00
Clayton Coleman
c9fc0bcf3d Decouple apiserver from codec implementation
The apiserver on initialization must be provided with a codec
for encoding and decoding all handled objects including api.Status
and api.ServerOp.  In addition, the RESTStorage Extract() method
has been changed to New(), which returns a pointer object that the
codec must decode into (the internal object).  Switched registry
methods to use pointers for Create/Update instead of values.
2014-08-07 13:35:35 -04:00
Daniel Smith
14c379d27b Merge pull request #812 from smarterclayton/sync_loop_max_wait
Kubelet should have a max think time before auto resync
2014-08-07 10:08:40 -07:00
Daniel Smith
9ede47299a Merge pull request #821 from csrwng/fix_local_up_cluster_controller
Remove etcd servers flag from controller manager in local-up-cluster.sh
2014-08-07 10:05:59 -07:00
Victor Marmol
c501e09d14 Merge pull request #820 from brendandburns/nsenter
Install nsinit that is vendored into docker.
2014-08-07 08:36:49 -07:00
Clayton Coleman
4123a44653 Merge pull request #819 from thockin/restarts
Disable docker restarts, for now
2014-08-07 10:41:56 -04:00
Clayton Coleman
d7f46718a8 Kubelet should have a max think time before auto resync
The sync frequency should be part of the syncLoop and resync no
less often than every X seconds.  The current implementation runs
even if a config update was delivered less than X seconds ago.
2014-08-07 10:40:03 -04:00
csrwng
c65470583e Remove etcd servers flag from controller manager in local-up-cluster 2014-08-07 10:02:15 -04:00