Zexi Li
833dfcc6e2
fix(scheduler): avoid_same_host 策略分数过高导致调度分配不均 ( #21415 )
2024-10-16 10:23:22 +08:00
wanyaoqi
2c51ca5e1e
fix(region,scheduler): allow req cpu count zero on allocate cpunumapin ( #21345 )
2024-10-03 20:03:23 +08:00
wanyaoqi
6f3f9cb058
fix(region,scheduler): sub pending usage on select cpu numa pin ( #20948 )
...
- reset guest cpu numa pin before start guest
- sub pending usage on select cpu numa pin
- hypervisor pod record pending usage
- add core:threads pair allocate, use same core threads first
Signed-off-by: wanyaoqi <d3lx.yq@gmail.com >
2024-08-02 10:22:53 +08:00
wanyaoqi
4690c2307d
feat(region,host,scheduler): scheduler support allocate cpu numa pin ( #20372 )
2024-07-30 19:10:18 +08:00
wanyaoqi
f4c6eadae8
fix(scheduler): optimize get guests by hosts ids ( #20700 )
2024-07-02 00:16:51 +08:00
Qiu Jian
8159a6d86c
fix: host nic index should be unique
2024-04-10 01:33:02 +08:00
屈轩
7156c24d68
fix(scheduler): skip network exit check ( #19665 )
2024-03-06 19:10:43 +08:00
Zexi Li
7cbff93e0d
fix(scheduler): unsued key
2024-02-28 10:26:50 +08:00
Jian Qiu
eef9e48e02
feature: allow manager to preapre query context ( #19455 )
...
Co-authored-by: Qiu Jian <qiujian@yunionyun.com >
2024-02-06 00:01:26 +08:00
Zexi Li
c8aed17eb2
optimized(scheduler): remove fillResidentTenants function loading data from db
2024-01-25 17:07:05 +08:00
Zexi Li
fff91b8eca
fix(scheduler): reload cloudprovider's hosts when syncing finished
2023-12-15 20:19:32 +08:00
guoguangwu
65d66edbc7
chore: unnecessary use of fmt.Sprintf d
...
Signed-off-by: guoguangwu <guoguangwu@magic-shield.com >
2023-11-13 20:42:42 +08:00
Qiu Jian
335ff385c5
fix: scheduler ping/pong response mime
2023-10-30 13:17:09 +08:00
Zexi Li
bdaf659a28
fix(scheduler): fetch host storages directly
2023-10-24 14:26:49 +08:00
Zexi Li
e97408dca8
fix(scheduler): improve scheduler prefer and avoid tags strategy ( #18373 )
2023-10-20 20:46:14 +08:00
Jian Qiu
b6e1e9d226
fix: allow vmware network associate with multiple wires ( #18341 )
...
Co-authored-by: Qiu Jian <qiujian@yunionyun.com >
2023-10-19 09:15:40 +08:00
屈轩
d5a3913a1f
fix: docgengo ( #18351 )
2023-10-18 06:19:10 -05:00
wanyaoqi
0fa98f7b82
fix(scheduler): isolated device predicate miss sriov nic
2023-10-16 16:22:19 +08:00
Zexi Li
7735720a5e
fix(scheduler): some joint resources are not synced
2023-09-27 16:47:36 +08:00
Jian Qiu
c5dd7939fc
fix: vmware host wire recode ( #17818 )
...
* fix: vmware host wire recode
* update vendor
* update vendor
---------
Co-authored-by: Qiu Jian <qiujian@yunionyun.com >
2023-08-24 16:36:55 +08:00
Jian Qiu
b7b2a630f5
fix: remove hostwire to cleanup host-wire codes ( #17670 )
...
Co-authored-by: Qiu Jian <qiujian@yunionyun.com >
2023-08-04 23:40:54 +08:00
Zexi Li
57751a20bf
fix(scheduler): auto allocation network selection is not working
2023-07-27 12:20:23 +08:00
Zexi Li
22c764246a
Merge pull request #17509 from wanyaoqi/feat/ovs-offload-vpc-support
...
fix(region,scheduler): ovs offload support vpc network
2023-07-13 21:45:07 +08:00
wanyaoqi
9e6b8ed538
fix(region,scheduler): ovs offload support vpc network
2023-07-13 14:37:59 +08:00
吴典秋
6e06d6eea2
fix fillStorage get storage cause scheduler panic
2023-07-13 08:34:24 +08:00
Zexi Li
e0c47dd5df
feat(apimap): fetch vpcagent net topo info from DB ( #17038 )
2023-05-15 14:48:24 +08:00
Zexi Li
55fa9ef1d0
Merge pull request #16845 from wanyaoqi/feat/lvm-disk
...
feat(region,host): storage backend lvm and nvme
2023-04-27 21:07:32 +08:00
Zexi Li
b667a4ccc3
fix(scheduler): skip load esxi network free-port cache
2023-04-27 09:27:26 +08:00
wanyaoqi
d394d9bb55
feat(region,host): nvme device passthrough
2023-04-24 17:55:39 +08:00
Zexi Li
3d778311aa
fix(scheduler): storage failed reason isn't showed in forecast result
2023-03-08 13:03:36 +08:00
Zexi Li
407c90ccfb
optimized(scheduler): reduce sql query of mysql ( #15878 )
2023-01-31 19:30:58 +08:00
Qiu Jian
1d5a43d66d
fix: init shceduler schedinfo project and domain
2023-01-05 15:26:33 +08:00
Zexi Li
076b6e5387
fix(scheduler): do migrate checking when hypervisor is not esxi
2023-01-03 18:14:51 +08:00
Zexi Li
b4c08e23f7
fix(scheduler): reduce calculation of storage's free_capacity
2022-12-30 12:40:54 +08:00
Jian Qiu
21716cefb5
fix: remove mutual dependency of cloudmux on onecloud ( #15621 )
...
Co-authored-by: Qiu Jian <qiujian@yunionyun.com >
2022-12-27 01:21:26 +08:00
Zexi Li
2643c8916e
optimize(scheduler): reduce calculation of network free address ( #15595 )
...
- only calculate non-emulated host's network on building
- calculate emulated host's network on scheduling
2022-12-20 20:54:16 +08:00
Zexi Li
15912a9882
fix(scheduler): consider storage backend and mediumType size ( #15469 )
2022-12-01 01:22:36 +08:00
wanyaoqi
ad3f592747
fix(scheduler): ignore fetch disk image failed ( #15361 )
2022-11-15 00:49:46 +08:00
wanyaoqi
70755a9a1b
feat(region,host): sriov nic support ( #15342 )
...
Support sriov nic passthrough to kvm guest.
Signed-off-by: wanyaoqi <d3lx.yq@gmail.com >
Signed-off-by: wanyaoqi <d3lx.yq@gmail.com >
2022-11-13 18:17:30 +08:00
Zexi Li
beb2951523
Merge pull request #15348 from swordqiu/hotfix/qj-scheduler-ignore-storage-capacity-constraint-if-migrating
...
fix: scheduler ignore storage capacity constraint if migrating
2022-11-12 18:40:52 +08:00
Qiu Jian
4fe9dcec16
fix: scheduler ignore storage capacity constraint if migrating
2022-11-11 19:49:57 +08:00
Zexi Li
af0dcee9e3
fix(scheduler): check capacity of disk_schedtag_predicate ( #15336 )
...
* only check free_capacity when not migrating or non schared storage
* only check actual_free_capacity when storage is local
2022-11-10 11:42:38 +08:00
Zexi Li
0e2c520ecd
feat(misc): use cloudmux package ( #15254 )
2022-10-28 19:53:03 +08:00
Zexi Li
8d28143a80
fix(scheduler): do free capacity check when storage filter enabled
2022-10-20 14:57:01 +08:00
Zexi Li
053944c265
fix(scheduler): check free capacity of each storage
2022-10-14 12:42:16 +08:00
Zexi Li
9952c9ad8d
Merge pull request #15168 from wanyaoqi/fix/hugepages
...
feat(region,host): hugepage optimize
2022-10-13 09:47:18 +08:00
wanyaoqi
3d0c09195e
feat(region,host): hugepage optimize
...
- host agent will not allocate hugepage on init
- except hugepage, the remaining memory update to reserved memory
- default hugepage size 1024M
- host model add page size field
Signed-off-by: wanyaoqi <d3lx.yq@gmail.com >
2022-10-12 14:57:13 +08:00
Zexi Li
48bd8d72d3
Automated cherry pick of #15120 : optimize(scheduler): global cache for schedtag ( #15121 )
...
* optimize(scheduler): global cache for schedtag
* feat(scheduler): informer of schedtag
2022-10-08 20:49:38 +08:00
Zexi Li
8f73e2c19d
fix(scheduler): testing failed ( #14943 )
2022-09-07 10:47:54 +08:00
Zexi Li
33e70266b0
fix(scheduler): fill storage and network results
2022-09-06 12:04:23 +08:00
