Eru是一款简单，无状态，灵活的企业级容器协调器
Eru是一款简单，无状态，灵活的企业级容器协调器，可轻松集成到现有工作流程中。 可以在很长或很短的时间内运行任何集装箱化的东西。
Changelog
1ee5e6f add ERU_CPU env to containers for their perception
2825832 add test cases, resolve dead lock
21bc787 bulky operate remove and diss by node
6d2d3ef cluster layer calls remap on resource changes
6825f39 fix after_create hook
f1a5a3c fix embeded etcd issue for tests
cd04cf4 fix: missing wal while make
5944f69 force logging with color
8754425 implement RemapResource for docker engine
113786d intro goroutine pool to restrain max concurrency
f67f8af keep error log stack concise
da7a7e3 pass unittests
2ec7b86 stipulate engine interface: ResourceRemap
Assets
5
Changelog
da065cf fix nil error, NodeStatus won't have an error when loading from etcd
Assets
5
Changelog
76081c8 accelerate workload force stop (#350)
f44a0e0 add get node status method (#351)
a3bafe3 feat: WAL for creating workload.
02fd27e fix build image from exist tests
e67535c fix submodule auth issue
3a53018 fix: no such container means WAL has been handled (#348)
cbb0e3c recify logging system (#349)
Assets
5
Changelog
86cafe8 feat: WAL for creating labmda (#342)
c6bd2aa fix issue #338, not expose restart policy to upper layer later
b84b401 ignore core.wal
7a3f6f5 intro goroutine pool to speed up engine API call (#343)
cd5681d redefine restart and keep backward compatible (#345)
cb6eaa9 redefine restart policy (#344)
0d7342e use grpc code indicate errors (#340)
Assets
5
Changelog
c927f5d Build Debug Version of Binary in Github Action (#335)
d192194 bugfix: set image tags for debug version image (#337)
8afc19a demultiplex exec output into stdout / stderr (#327)
13064e4 refactor: wraps grant and keepalive.
14b8e7c refines DAL from DAO (#330)
0a6e110 remove useless copy in strategy
923b6ee round cpu when reallocing
566ff66 set cpu request=limit when cpubind=true (#334)
Assets
5
Changelog
6aab298 add SetNodeStatus and NodeStatusStream API (#323)
fa0fab2 bugfix: make correction of total available calculation (#322)
Assets
5
Changelog
6aab298 add SetNodeStatus and NodeStatusStream API (#323)
fa0fab2 bugfix: make correction of total available calculation (#322)
58385f1 improve scheduler getFullResult (#324)
06419bc remove fill global, fill now can calcuate with capacity
Assets
5
Changelog
9afaeb5 bugfix: node limit effect in deploy strategy (#252)
600dc8b minor revise strategy rpc define
cbde8d3 refactor service discovery (#251)
30eb518 split strategy and scheduler (#249)
7dbe69e support multiple nodenames (#255)
Assets
5
Changelog
dc13f79 decrease service put frequency
16bf528 support get all nodes by ListPodNodes API
35bd9fc support store status forever
Assets
5
Changelog
9afaeb5 bugfix: node limit effect in deploy strategy (#252)
600dc8b minor revise strategy rpc define
cbde8d3 refactor service discovery (#251)
30eb518 split strategy and scheduler (#249)
7dbe69e support multiple nodenames (#255)
Assets
5
Changelog
1631542 Abstract Resources (#256)
9afaeb5 bugfix: node limit effect in deploy strategy (#252)
f2c5f6a minor revise strategy rpc define
cbde8d3 refactor service discovery (#251)
30f925f refactor: revise a lot (#259)
51a0fa5 resource request & limit (#258)
30eb518 split strategy and scheduler (#249)
7dbe69e support multiple nodenames (#255)
Assets
5
Changelog
9afaeb5 bugfix: node limit effect in deploy strategy (#252)
600dc8b minor revise strategy rpc define
cbde8d3 refactor service discovery (#251)
30eb518 split strategy and scheduler (#249)
7dbe69e support multiple nodenames (#255)
Assets
5
Assets
5
Assets
5
Changelog
117b5f7 add RPC WatchServiceStatus into protobuf
ce496b6 check cpu suffience when realloc (#236)
8a1956c client: discovery watches service status change
333f861 client: fast resolve address (#244)
6b3e03a client: resolver updates svc addresses periodically
abfaba5 client: retry watching stream
e6b5d62 client: service discovery can watch core with auth
c1c82a6 client: set keepalive and balancer options
da4e6ef cluster: WatchServiceStatus, RegisterService
81e2e4e cluster: register service can fast fail (1s)
ec90ff2 get nodes to speed up bind methods (#237)
b9b1ab9 minor refactor service discovery
6d7e5a3 move grpc related configures
fd80ffd refactor network interface, will response subnets (#240)
50cedea regenerate go.mod
6eb0177 store, cluster: add unittests for service
4868f94 store: RegisterService, UnregisterService
cd44595 support network control (#239)
9cf7783 virt attach can specify command (#235)
Assets
5
Changelog
22e0c91 apply lint suggestions, add rpc Info API
50b80e2 async lambda outputs messages by line (#224)
4af90c7 bugfix: fix repo username when image building (#229)
f52a97d due to docker issue, realloc unbind cpu will bind all cpu to targets (fixed #228)
a9d896f pod resource return volumes usage (#232)
fdfc485 support fix resource when get node resource (#231)
f5fa132 support override memory limit options when reallocing (#226)
83ddc24 switch the git lib from libgit2 to go-git (#225)
92447db use tri options to revise cpu bind and memory limit (#227)
d9015db use triOpt in node set (#230)
Assets
5
Changelog
59d9065 introduce ReaderManager to address race condition in smaller critical (#220)
ec48556 labels pass-through
f6c8321 replace etcd deps with v3 version(github.com/coreos/etcd => go.etcd.io/etcd/v3) (#221)
94b8f94 resolve race condition during creation (#219)
Assets
4
| Watchers：11 |
| Star：148 |
| Fork：26 |
| 创建时间： 2017-09-07 15:10:04 |
| 最后Commits： 前天 |
| 许可协议：MIT |
7624030Compare
Verified
Changelog
549c0f7 Revert "workload id will be returned as the first message (#364)"
6e5e014 add err handling, avoid panic (#374)
84f4705 add error for node set down (#359)
9759655 add trylock for mutex
8902e3b bugfix: doRemoveWorkload isn't for rollback of deploy (#368)
ab9cb9f bugfix: fix nil store in tests
5914b8f enable redis as store
6d5814b fix: make golint works (#373)
7624030 fix: wal deadlock (#375)
feaf46d make mock engine rework again
a2ba499 redis as broker (#369)
4bb9861 remove external redis dependency for testing (#371)
41c527b support excluding nodes when creating workloads (#365)
c90580c support ulimit for docker engine #362
789ca9d to remove ctx which unused (#372)
5399d07 wider coverage of error stack (#358)
f87a4b0 workload id will be returned as the first message (#364)